Partial save of the draft agent config (instructions, model fields, agentTools, tool_search, browser, guardrails, workflows, memory, scorers, subagents, etc). Only the fields you send are written. Subagents are [{ agentId }]. The model fields are model_slug, policy_slug, model_options and model_settings; the older requesty_model_slug, requesty_policy_slug and requesty_options spellings are still accepted and mapped onto them (the canonical name wins if you send both). prompt, model_name, model_provider and model_openrouter_slug no longer exist and are ignored — the system prompt is instructions. name and description are the agent’s own, not the draft’s: they are saved here for convenience but apply immediately to every version, published ones included, and need no publish. The response carries the updated agent alongside the draft.
Organization API key, generated in studio.sidenet.ai. Backend only — never in a browser.
Path Parameters
id
string
required
Example:
"6f1c0f4e-2b7a-4a51-9a2f-0c9d1b3e5a10"
Body
application/json
instructions
object[]
The agent's system prompt, as an ordered list of blocks. Each entry is one of: { type: "text", content } — inline text; { type: "prompt_block", content, rules? } — inline text with an optional display condition; { type: "prompt_block_ref", id } — a reference to a /v1/prompt-blocks block in the same organization. Content may contain {{variable}}, {{a.b}} and {{variable || 'default'}} placeholders, resolved per request from the chat body's variables plus the now.* builtins; an unsupplied placeholder is left visible. rules is a { operator: "AND"|"OR", conditions: [...] } group (max 3 levels) — the block is included only when it evaluates true. On the draft a ref floats to the block's active version; publishing PINS it, so a published agent version keeps its exact prompt until you republish (see /v1/prompt-blocks/:id/rollout). Pins are server-owned — do not send versionId.
The concrete model to run, e.g. anthropic/claude-sonnet-4-5. Ignored when policy_slug is set.
Example:
"anthropic/claude-sonnet-4-5"
policy_slug
string
A routing policy to run instead of a fixed model, e.g. policy/fastest-sonnet. Overrides model_slug when set.
Example:
"policy/fastest-sonnet"
model_options
object
Model-level options passed through to the router — reasoning effort, native provider options, and the like.
Example:
{ "reasoningEffort": "medium" }
model_settings
object
Generation settings: temperature, maxTokens, and similar.
Example:
{ "temperature": 0.2 }
max_steps
integer
Max LLM round-trips per run for the agentic tool loop. The final step is forced tool-free so runs always end with a text answer. null = platform default (20).
Required range: 1 <= x <= 100
Example:
8
browser
object | null
The built-in browser, off by default. { "enabled": true } gives the agent a web browser that runs in the EU: it can open pages, read them and act on them in a browser session of the conversation, which the person can watch, take over and hand back (/v1/browser-sessions). viewport ({ width, height } in pixels, default 1280 × 800) is the size of its window. With it on, the agent's instructions gain a short block of rules: use the browser only when no tool does the job, treat what a page says as data, never fill a payment form, and send anything on a website only through browser_submit, which asks the organization's submit check first (or, without one, the person in the conversation). Unknown keys are rejected. null or enabled: false turns it off.
Tool search for agents with a large roster (a big API import, a wide MCP server, several connected-app toolkits). ON BY DEFAULT: with more than threshold tools attached to the agent (default 40, connected-app tools included; platform tools such as the workflow builder are not counted), only an always-loaded set stays in the prompt and the rest is reached through two meta-tools: search_tools (keywords → matching tool names and descriptions) and load_tool (exact names → callable for the rest of the conversation). One search surface per agent: API, MCP and connected-app tools share it. providers scopes it to a list of provider ids (API / MCP tools) and connected-app toolkit slugs — only their tools are searchable, everything else stays in the prompt; omit it to search the whole roster. The always-loaded set is the rendering components, the large-response retrieval tools, the workflow builder, every tool with require_approval, the connection tools, discovery: false entries, and whatever always_loaded names (tool names or ids). top_k is the number of search results (1–20, default 8). show_catalog: true also lists every searchable tool's name and a short description in the prompt, trading tokens for one fewer round-trip — worth it for a few dozen tools, not for hundreds. Under the threshold, or with enabled: false, your own API and MCP tools all go into the prompt and connected-app toolkits keep their automatic pooling above ~15 direct tools (see agentTools). Send null to go back to the platform default.
Guardrails: safety checks that run on every turn of this agent, one entry per kind (see the guardrails guide). kind: pii (personal data in messages: emails, phone numbers, cards, ids…), moderation (hate, harassment, violence, sexual content, self-harm), prompt_injection (jailbreaks and attempts to override the instructions), system_prompt_scrub (the agent quoting its own instructions back), language (messages outside the target languages) and unicode (control characters and odd whitespace in the user message). mode says what happens when a guardrail fires: warn (default) records the violation and lets the turn continue; block ends the turn with an error the chat shows. All kinds but unicode ask a small judge model; that spend bills the organization's "LLM Judges" group, never a customer group, and a judge-backed guardrail is skipped (with a warning in the rebuild report) when the organization has no judge key. In block mode a judge that fails also stops the turn (fail-closed); in warn mode it lets the message through. enabled: false keeps an entry but turns it off. options tunes the kind — every key optional, unknown keys are rejected: threshold (0–1 confidence to fire; the defaults are 0.6 for pii, 0.5 for moderation, 0.7 for prompt_injection and language), model (a judge model slug, default a small fast one), instructions (custom judge instructions, up to 4000 characters), strategy (an action other than warn/block when the guardrail fires: redact or filter for pii and system_prompt_scrub, rewrite or filter for prompt_injection, filter for moderation, translate or detect for language — mode then only decides what a judge failure does), stage (input, output or both; pii and moderation only, default input), detection_types (pii, prompt_injection), redaction_method (pii: mask, hash, remove, placeholder; system_prompt_scrub: mask, placeholder, remove), categories (moderation), target_languages and min_text_length (language), custom_patterns and placeholder_text (system_prompt_scrub), strip_control_chars, collapse_whitespace, preserve_emojis and trim (unicode, all default true). Send null to remove every guardrail.
Workflows the agent can run, as { workflow_id, config? } entries. workflow_id is the /v1/workflows id and must belong to the same organization. The workflow's ACTIVE version is always what runs, so activating a new version reaches the agent on its next turn with no republish. The agent-facing tool name and its description come from the workflow itself — the model calls it as workflow-<workflow name>. The whole workflow runs inside one tool call, and its step progress is streamed to the client as data-tool-workflow / data-tool-workflow-step parts by /v1/chat.
Skills the agent can load, as { skill_id } entries. skill_id is the /v1/skills id and must belong to the same organization. A skill is a named block of instructions the agent loads on demand — it calls its skill tool with the name when the skill's description says to — instead of carrying the text in every prompt. The agent always gets the skill's ACTIVE version, so publishing a new one reaches the agent on its next turn with no republish; a skill that is not published yet is skipped and reported in the rebuild summary. Send null to remove every skill.
Tools assigned to the agent. Each entry is { tool_id, provider_id?, config?, composio_connected_account_id?, discovery?, loop_responses?, require_approval? }. The tool_name / provider_name a read returns are ignored here (resolved again on the way out), so an entry can be sent back as it was read. discovery (integration tools only) controls where the tool schema lives at runtime: true puts it behind the agent's search_tools/load_tool discovery tools, false pins it into the prompt, and unset lets the platform decide — small toolkits stay in the prompt, large ones (beyond ~15 total tools) become discoverable so the prompt is not flooded with schemas. require_approval: true pauses every call to that tool for human approval: the chat stream emits a data-tool-call-approval chunk instead of executing (carrying a model-written reason sentence for the approval card), and the client continues the run by POSTing /v1/chat with the approval field. Set loop_responses: true on a tool to let one call cover a whole result set: it gains _fields (project only the fields needed), _all_pages (fetch and merge every page) and _for_each (call once per value of a parameter and merge the rows), and the agent is given the large-response retrieval tools automatically. Because those re-issue the request, enable it only on tools whose endpoint is safe to call repeatedly.
The agent's name. Agent-level, not versioned — it takes effect immediately, with no publish, and is what other agents see when this one is a subagent.
Example:
"Support agent"
description
string
The agent's description. Agent-level, not versioned — same immediate effect as name. Send null to clear it.
Example:
"Answers billing and account questions"
scorers
object[]
Scorers for this version. Each entry: name (unique key), scorerType (catalogue id), model (judge slug, LLM scorers only), sampling_rate (0-1, share of eligible runs scored, default 1), options, and an optional filter — a predicate over the run, evaluated before sampling, over requestContext.* (groupId, environment, copilotId, source, userId…), entity.*, threadId, resourceId. Operators: eq/ne/lt/lte/gt/gte (left/right as { path } or { literal }), in/notIn (value, set), exists/notExists (path), truthy/falsy (value), and/or (args), not (arg). An invalid filter answers 400.
Memory settings of this version, as { options } — see the memory guide. options: lastMessages (recent messages read back, default 10), generateTitle (first title from the first message; true, false or { model, instructions }), semanticRecall (true, false or { topK, messageRange, scope }), workingMemory (true, false or { enabled, scope, template | schema, agentManaged } — scope: "resource", the default, is one record per user shared by every conversation and automation run), observationalMemory (true, false or { model, activateAfterIdle, retrieval, observation: { messageTokens, bufferTokens, bufferActivation, blockAfter, bufferOnIdle, manageWorkingMemory, threadTitle, instruction, activateAfterIdle }, reflection: { observationTokens, bufferActivation, blockAfter, activateAfterIdle, instruction } }; its scope is thread only). Unknown keys and out-of-range values answer 400 naming the path. Null = platform defaults. Model slugs bill the organization's key.
The saved draft, and the agent record it belongs to
agent
object
The agent after the save — name/description reflect this call, and the version summary says whether the draft now differs from what is published.
Show child attributes
draft
object
The whole draft after the merge, not just the fields you sent. rev has been bumped by the DB.
Show child attributes
{"docsConfig":{"theme":"mint","$schema":"https://mintlify.com/docs.json","name":"Sidenet API","colors":{"primary":"#4F46E5","light":"#818CF8","dark":"#4338CA"},"logo":{"light":"https://mintcdn.com/sidenetai/IsfHUeiSCKNFr_6U/images/logo-light.svg?fit=max\u0026auto=format\u0026n=IsfHUeiSCKNFr_6U\u0026q=85\u0026s=206c6acd674e00cc951ff99d54600daa","dark":"https://mintcdn.com/sidenetai/IsfHUeiSCKNFr_6U/images/logo-dark.svg?fit=max\u0026auto=format\u0026n=IsfHUeiSCKNFr_6U\u0026q=85\u0026s=db3396c278572b71489a4c1be475acd4","href":"https://www.sidenet.ai"},"favicon":{"light":"https://mintcdn.com/sidenetai/bYiM267pd74qNSPc/images/3bc48416-a174-468b-9d1a-56046f8bb9b4.png?fit=max\u0026auto=format\u0026n=bYiM267pd74qNSPc\u0026q=85\u0026s=b651ac95041904f0eaf2678e8d2704ab","dark":"https://mintcdn.com/sidenetai/bYiM267pd74qNSPc/images/3bc48416-a174-468b-9d1a-56046f8bb9b4.png?fit=max\u0026auto=format\u0026n=bYiM267pd74qNSPc\u0026q=85\u0026s=b651ac95041904f0eaf2678e8d2704ab"},"api":{"examples":{"prefill":true}},"navbar":{"links":[{"href":"https://www.sidenet.ai","label":"Back to website"},{"href":"https://www.sidenet.ai/book-demo/","label":"Book a demo"}],"primary":{"type":"button","label":"Sidenet Studio","href":"https://studio.sidenet.ai"}},"footer":{"socials":{"website":"https://www.sidenet.ai"}},"seo":{"metatags":{"canonical":"https://www.sidenet.ai"},"indexing":"all"},"styling":{"latex":false},"navigation":{"pages":[]}},"deployment":{"deploymentMetadataValue":{"subdomain":"sidenetai","hasAudioNarration":false,"actualSubdomain":"sidenetai","customDomains":["sidenet.ai"],"gitSource":{"type":"github","owner":"alexdanan","repo":"sidenet-api","deployBranch":"master","contentDirectory":"docs","isPrivate":true,"mountPath":"","label":""},"entitlements":{"AI_CHAT":{"status":"DISABLED"}},"buildId":"6af9f11af1de041cd01487d6","clientVersion":"0.0.4143","searchFilterCounts":[{"language":"default","tag":"API Reference","count":163},{"language":"default","tag":"Guides","count":23},{"language":"default","tag":"Browser","count":17},{"language":"default","tag":"Chat SDK","count":13},{"language":"default","tag":"Chat \u0026 Messaging","count":12},{"language":"default","tag":"Workflow Schedules","count":12},{"language":"default","tag":"Dashboards","count":11},{"language":"default","tag":"Networks","count":11},{"language":"default","tag":"Prompt Blocks","count":11},{"language":"default","tag":"Workflows","count":10}]},"assistantConfig":{"enableAskAiButton":false,"enableStarterQuestions":false,"enableContextualStarterQuestions":false,"enableAskAiSearchBar":true},"basePath":"/docs","hasChatPermissions":false,"hasWebMcpPermissions":false,"toggles":[{"name":"kill-isr","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"custom-rbac-roles","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"agent-mintlify-slack","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"agent-integrations-settings","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"rbac-expansion","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"automations-pr-grouping","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"custom-prefix","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"analytics-v7","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"user-flows-analytics","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"no-publish-flow","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-onboarding","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"snapshot-serving-live","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"stripTemperatureMiddleware","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"support-platform-pylon-zendesk","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"snapshot-serving","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"instant-rollback","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"granular-notifications-server","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"granular-notifications","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"automations-fable","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"fixed-ai-action-pricing","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"automation-resolve-merge-conflict","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"scoped-publishing","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"mintql","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"page-level-analytics","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"editor-version-history-v2","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"billing-page-v2","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"full-page-search","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workspace-projects","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"assistant-v5","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"fork-pr-previews","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"revalidation-jwt","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-sso-scim","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"hobby-abuse-scan","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"dashboard-inbox","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-signup","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"turbopuffer-dual-write","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"workos-enabled","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"approve-review-flow","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"editor-metadata","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"sso-setup-deadline-banner","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"turbopuffer-serve","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"live-agents","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false},{"name":"llms-txt-v2","enabled":true,"variant":{"name":"disabled","enabled":false,"feature_enabled":true},"impressionData":false}],"googleFontsInlined":false},"customCss":[],"page":{"pagePath":"api-reference/agents/save-agent-draft","slug":"/api-reference/agents/save-agent-draft","pageMetadata":{"title":"Save agent draft","description":"Partial save of the draft agent config (instructions, model fields, agentTools, tool_search, browser, guardrails, workflows, memory, scorers, subagents, etc). Only the fields you send are written. Subagents are `[{ agentId }]`. The model fields are `model_slug`, `policy_slug`, `model_options` and `model_settings`; the older `requesty_model_slug`, `requesty_policy_slug` and `requesty_options` spellings are still accepted and mapped onto them (the canonical name wins if you send both). `prompt`, `model_name`, `model_provider` and `model_openrouter_slug` no longer exist and are ignored — the system prompt is `instructions`. `name` and `description` are the agent's own, not the draft's: they are saved here for convenience but apply immediately to every version, published ones included, and need no publish. The response carries the updated `agent` alongside the `draft`.","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/agents/{id}/draft","href":"/api-reference/agents/save-agent-draft","autogeneratedByOpenApi":true},"description":{"compiledSource":"\"use strict\";\nconst {jsx: _jsx, jsxs: _jsxs} = arguments[0];\nconst {useMDXComponents: _provideComponents} = arguments[0];\nfunction _createMdxContent(props) {\n const _components = {\n code: \"code\",\n p: \"p\",\n ..._provideComponents(),\n ...props.components\n };\n return _jsxs(_components.p, {\n children: [\"Partial save of the draft agent config (instructions, model fields, agentTools, tool_search, browser, guardrails, workflows, memory, scorers, subagents, etc). Only the fields you send are written. Subagents are \", _jsx(_components.code, {\n children: \"[{ agentId }]\"\n }), \". The model fields are \", _jsx(_components.code, {\n children: \"model_slug\"\n }), \", \", _jsx(_components.code, {\n children: \"policy_slug\"\n }), \", \", _jsx(_components.code, {\n children: \"model_options\"\n }), \" and \", _jsx(_components.code, {\n children: \"model_settings\"\n }), \"; the older \", _jsx(_components.code, {\n children: \"requesty_model_slug\"\n }), \", \", _jsx(_components.code, {\n children: \"requesty_policy_slug\"\n }), \" and \", _jsx(_components.code, {\n children: \"requesty_options\"\n }), \" spellings are still accepted and mapped onto them (the canonical name wins if you send both). \", _jsx(_components.code, {\n children: \"prompt\"\n }), \", \", _jsx(_components.code, {\n children: \"model_name\"\n }), \", \", _jsx(_components.code, {\n children: \"model_provider\"\n }), \" and \", _jsx(_components.code, {\n children: \"model_openrouter_slug\"\n }), \" no longer exist and are ignored — the system prompt is \", _jsx(_components.code, {\n children: \"instructions\"\n }), \". \", _jsx(_components.code, {\n children: \"name\"\n }), \" and \", _jsx(_components.code, {\n children: \"description\"\n }), \" are the agent’s own, not the draft’s: they are saved here for convenience but apply immediately to every version, published ones included, and need no publish. The response carries the updated \", _jsx(_components.code, {\n children: \"agent\"\n }), \" alongside the \", _jsx(_components.code, {\n children: \"draft\"\n }), \".\"]\n });\n}\nfunction MDXContent(props = {}) {\n const {wrapper: MDXLayout} = {\n ..._provideComponents(),\n ...props.components\n };\n return MDXLayout ? _jsx(MDXLayout, {\n ...props,\n children: _jsx(_createMdxContent, {\n ...props\n })\n }) : _createMdxContent(props);\n}\nreturn {\n default: MDXContent\n};\n","frontmatter":{},"scope":{}},"pageType":"default","mdx":{"mdxSource":{"compiledSource":"\"use strict\";const{Fragment:_Fragment,jsx:_jsx}=arguments[0];const{useMDXComponents:_provideComponents}=arguments[0];function _createMdxContent(props){return _jsx(_Fragment,{})}function MDXContent(props={}){const{wrapper:MDXLayout}={..._provideComponents(),...props.components};return MDXLayout?_jsx(MDXLayout,{...props,children:_jsx(_createMdxContent,{...props})}):_createMdxContent(props)}return{default:MDXContent}","frontmatter":{},"scope":{"config":{},"pageMetadata":{"title":"Save agent draft","description":"Partial save of the draft agent config (instructions, model fields, agentTools, tool_search, browser, guardrails, workflows, memory, scorers, subagents, etc). Only the fields you send are written. Subagents are `[{ agentId }]`. The model fields are `model_slug`, `policy_slug`, `model_options` and `model_settings`; the older `requesty_model_slug`, `requesty_policy_slug` and `requesty_options` spellings are still accepted and mapped onto them (the canonical name wins if you send both). `prompt`, `model_name`, `model_provider` and `model_openrouter_slug` no longer exist and are ignored — the system prompt is `instructions`. `name` and `description` are the agent's own, not the draft's: they are saved here for convenience but apply immediately to every version, published ones included, and need no publish. The response carries the updated `agent` alongside the `draft`.","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/agents/{id}/draft","href":"/api-reference/agents/save-agent-draft","autogeneratedByOpenApi":true}}},"mdxSourceWithNoJs":null},"mdxExtracts":{"tableOfContents":[],"codeExamples":{}},"jsLinks":["/docs/_mintlify/custom/sidenetai/8e3ad2809aba5c1a/analytics.js"],"openApiReferenceData":{"operation":{"summary":"Save agent draft","description":"Partial save of the draft agent config (instructions, model fields, agentTools, tool_search, browser, guardrails, workflows, memory, scorers, subagents, etc). Only the fields you send are written. Subagents are `[{ agentId }]`. The model fields are `model_slug`, `policy_slug`, `model_options` and `model_settings`; the older `requesty_model_slug`, `requesty_policy_slug` and `requesty_options` spellings are still accepted and mapped onto them (the canonical name wins if you send both). `prompt`, `model_name`, `model_provider` and `model_openrouter_slug` no longer exist and are ignored — the system prompt is `instructions`. `name` and `description` are the agent's own, not the draft's: they are saved here for convenience but apply immediately to every version, published ones included, and need no publish. The response carries the updated `agent` alongside the `draft`.","tags":["Agents"],"parameters":["876efdfd-1ead-4bdd-9a56-81f39b39571e"],"requestBody":"ad3fcf31-96c5-4603-93c8-c2ae39efe88c","responses":{"200":"0df57298-0ace-408a-8f6f-aa15ea4f85f0","400":"913bfefa-1d42-4af0-908b-b78127fe9800","404":"54dea7b5-ddc3-4d3f-8fd4-9f2ed66c5dbe"},"method":"patch","type":"path","servers":["5a9ef6cd-d896-47ca-98f0-fbb8d6f1aa96"],"security":["b98d762b-7a65-4629-83ec-12fce5b08580"],"path":"/v1/agents/{id}/draft","baseUrlOptions":["https://api.sidenet.ai"],"title":"Save agent draft","codeSamples":[],"requiredOnlyExamples":false,"requestExampleType":"generated","responseExampleType":"generated","prefillPlaygroundWithExample":true,"prefillPlaygroundExamples":[{"body.instructions.items.0":{"type":"prompt_block_ref","id":"d41a8f6b-9c27-4b13-a5e8-2f60c1d7b394"},"body.instructions.items.1":{"type":"text","content":"Answer in {{locale || 'English'}}."},"body.model_slug":"anthropic/claude-sonnet-4-5","body.policy_slug":"policy/fastest-sonnet","body.model_options":{"reasoningEffort":"medium"},"body.model_settings":{"temperature":0.2},"body.max_steps":8,"body.browser.oneOf.0":{"enabled":true,"viewport":{"width":1280,"height":800}},"body.tool_search.oneOf.0.threshold":40,"body.tool_search.oneOf.0.providers.items.0":"8d3b1a75-6c02-4e59-b84f-27a9d5e10c63","body.tool_search.oneOf.0.always_loaded.items.0":"BackyouApi_request_hotel_classics","body.guardrails.oneOf.0.items.0.kind":"prompt_injection","body.guardrails.oneOf.0.items.0.mode":"block","body.guardrails.oneOf.0.items.1.kind":"pii","body.guardrails.oneOf.0.items.1.mode":"warn","body.guardrails.oneOf.0.items.1.options":{"strategy":"redact","detection_types":["email","phone","credit-card"]},"body.guardrails.oneOf.0.items.2.kind":"moderation","body.guardrails.oneOf.0.items.2.mode":"block","body.guardrails.oneOf.0.items.2.options":{"categories":["hate","harassment","violence"],"threshold":0.7},"body.workflows.items.0.workflow_id":"4c2f9e18-7a63-4d05-b1e8-93a7c0f2d641","body.skills.oneOf.0.items.0.skill_id":"9a4e2c71-5b08-4f3d-a6c2-1e7d8b0f4c23","body.agentTools.items.0":{"tool_id":"c07f2b95-8d41-4e63-a029-5f7b1c8d3e40","provider_id":"8d3b1a75-6c02-4e59-b84f-27a9d5e10c63"},"body.name":"Support agent","body.description":"Answers billing and account questions","body.scorers.items.0":{"name":"faithfulness","scorerType":"faithfulness","model":"openai/gpt-4o-mini","sampling_rate":0.2,"filter":{"op":"and","args":[{"op":"eq","left":{"path":"requestContext.environment"},"right":{"literal":"production"}},{"op":"in","value":{"path":"requestContext.groupId"},"set":["grp_vip","grp_enterprise"]}]}},"body.agents.items.0.agentId":"6f1c0f4e-2b7a-4a51-9a2f-0c9d1b3e5a10","body.memory.oneOf.0.options.lastMessages.oneOf.1":20,"body.memory.oneOf.0.options.workingMemory.oneOf.1.enabled":true,"body.memory.oneOf.0.options.workingMemory.oneOf.1.scope":"resource","body.memory.oneOf.0.options.observationalMemory.oneOf.1.enabled":true,"body.memory.oneOf.0.options.observationalMemory.oneOf.1.observation.bufferOnIdle":true,"body.memory.oneOf.0.options.observationalMemory.oneOf.1.observation.manageWorkingMemory":true,"body.memory.oneOf.0.options.observationalMemory.oneOf.1.observation.threadTitle":true}]},"dependencies":{"parameters":{"876efdfd-1ead-4bdd-9a56-81f39b39571e":{"name":"id","in":"path","required":true,"schema":{"type":"string","example":"6f1c0f4e-2b7a-4a51-9a2f-0c9d1b3e5a10","isRequired":true,"uniqueKey":"","typeLabel":"string"}}},"requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"instructions":{"type":"array","description":"The agent's system prompt, as an ordered list of blocks. Each entry is one of: `{ type: \"text\", content }` — inline text; `{ type: \"prompt_block\", content, rules? }` — inline text with an optional display condition; `{ type: \"prompt_block_ref\", id }` — a reference to a /v1/prompt-blocks block in the same organization. Content may contain `{{variable}}`, `{{a.b}}` and `{{variable || 'default'}}` placeholders, resolved per request from the chat body's `variables` plus the `now.*` builtins; an unsupplied placeholder is left visible. `rules` is a `{ operator: \"AND\"|\"OR\", conditions: [...] }` group (max 3 levels) — the block is included only when it evaluates true. On the draft a ref floats to the block's active version; publishing PINS it, so a published agent version keeps its exact prompt until you republish (see /v1/prompt-blocks/:id/rollout). Pins are server-owned — do not send `versionId`.","items":{"type":"object","uniqueKey":"instructions.items.[INDEX]","typeLabel":"object"},"example":[{"type":"prompt_block_ref","id":"d41a8f6b-9c27-4b13-a5e8-2f60c1d7b394"},{"type":"text","content":"Answer in {{locale || 'English'}}."}],"uniqueKey":"instructions","typeLabel":"object[]"},"model_slug":{"type":"string","description":"The concrete model to run, e.g. `anthropic/claude-sonnet-4-5`. Ignored when `policy_slug` is set.","example":"anthropic/claude-sonnet-4-5","uniqueKey":"model_slug","typeLabel":"string"},"policy_slug":{"type":"string","description":"A routing policy to run instead of a fixed model, e.g. `policy/fastest-sonnet`. Overrides `model_slug` when set.","example":"policy/fastest-sonnet","uniqueKey":"policy_slug","typeLabel":"string"},"model_options":{"type":"object","description":"Model-level options passed through to the router — reasoning effort, native provider options, and the like.","example":{"reasoningEffort":"medium"},"uniqueKey":"model_options","typeLabel":"object"},"model_settings":{"type":"object","description":"Generation settings: temperature, maxTokens, and similar.","example":{"temperature":0.2},"uniqueKey":"model_settings","typeLabel":"object"},"max_steps":{"type":"integer","minimum":1,"maximum":100,"description":"Max LLM round-trips per run for the agentic tool loop. The final step is forced tool-free so runs always end with a text answer. null = platform default (20).","example":8,"uniqueKey":"max_steps","typeLabel":"integer"},"browser":{"oneOf":[{"type":"object","description":"The built-in browser, off by default. `{ \"enabled\": true }` gives the agent a web browser that runs in the EU: it can open pages, read them and act on them in a browser session of the conversation, which the person can watch, take over and hand back (`/v1/browser-sessions`). `viewport` (`{ width, height }` in pixels, default 1280 × 800) is the size of its window. With it on, the agent's instructions gain a short block of rules: use the browser only when no tool does the job, treat what a page says as data, never fill a payment form, and send anything on a website only through `browser_submit`, which asks the organization's submit check first (or, without one, the person in the conversation). Unknown keys are rejected. `null` or `enabled: false` turns it off.","example":{"enabled":true,"viewport":{"width":1280,"height":800}},"uniqueKey":"browser.oneOf.0","typeLabel":"object"},{"type":"null","description":"The built-in browser, off by default. `{ \"enabled\": true }` gives the agent a web browser that runs in the EU: it can open pages, read them and act on them in a browser session of the conversation, which the person can watch, take over and hand back (`/v1/browser-sessions`). `viewport` (`{ width, height }` in pixels, default 1280 × 800) is the size of its window. With it on, the agent's instructions gain a short block of rules: use the browser only when no tool does the job, treat what a page says as data, never fill a payment form, and send anything on a website only through `browser_submit`, which asks the organization's submit check first (or, without one, the person in the conversation). Unknown keys are rejected. `null` or `enabled: false` turns it off.","example":{"enabled":true,"viewport":{"width":1280,"height":800}},"uniqueKey":"browser.oneOf.1","typeLabel":"null"}],"uniqueKey":"browser","typeLabel":"object | null"},"tool_search":{"oneOf":[{"type":"object","description":"Tool search for agents with a large roster (a big API import, a wide MCP server, several connected-app toolkits). ON BY DEFAULT: with more than `threshold` tools attached to the agent (default 40, connected-app tools included; platform tools such as the workflow builder are not counted), only an always-loaded set stays in the prompt and the rest is reached through two meta-tools: `search_tools` (keywords → matching tool names and descriptions) and `load_tool` (exact names → callable for the rest of the conversation). One search surface per agent: API, MCP and connected-app tools share it. `providers` scopes it to a list of provider ids (API / MCP tools) and connected-app toolkit slugs — only their tools are searchable, everything else stays in the prompt; omit it to search the whole roster. The always-loaded set is the rendering components, the large-response retrieval tools, the workflow builder, every tool with `require_approval`, the connection tools, `discovery: false` entries, and whatever `always_loaded` names (tool names or ids). `top_k` is the number of search results (1–20, default 8). `show_catalog: true` also lists every searchable tool's name and a short description in the prompt, trading tokens for one fewer round-trip — worth it for a few dozen tools, not for hundreds. Under the threshold, or with `enabled: false`, your own API and MCP tools all go into the prompt and connected-app toolkits keep their automatic pooling above ~15 direct tools (see `agentTools`). Send null to go back to the platform default.","properties":{"enabled":{"type":"boolean","default":true,"uniqueKey":"tool_search.oneOf.0.enabled","typeLabel":"boolean"},"threshold":{"type":"integer","minimum":0,"maximum":1000,"default":40,"uniqueKey":"tool_search.oneOf.0.threshold","typeLabel":"integer"},"providers":{"type":"array","items":{"type":"string","uniqueKey":"tool_search.oneOf.0.providers.items.[INDEX]","typeLabel":"string"},"uniqueKey":"tool_search.oneOf.0.providers","typeLabel":"string[]"},"always_loaded":{"type":"array","items":{"type":"string","uniqueKey":"tool_search.oneOf.0.always_loaded.items.[INDEX]","typeLabel":"string"},"uniqueKey":"tool_search.oneOf.0.always_loaded","typeLabel":"string[]"},"top_k":{"type":"integer","minimum":1,"maximum":20,"default":8,"uniqueKey":"tool_search.oneOf.0.top_k","typeLabel":"integer"},"show_catalog":{"type":"boolean","default":false,"uniqueKey":"tool_search.oneOf.0.show_catalog","typeLabel":"boolean"}},"example":{"threshold":40,"providers":["8d3b1a75-6c02-4e59-b84f-27a9d5e10c63"],"always_loaded":["BackyouApi_request_hotel_classics"]},"uniqueKey":"tool_search.oneOf.0","typeLabel":"object"},{"type":"null","description":"Tool search for agents with a large roster (a big API import, a wide MCP server, several connected-app toolkits). ON BY DEFAULT: with more than `threshold` tools attached to the agent (default 40, connected-app tools included; platform tools such as the workflow builder are not counted), only an always-loaded set stays in the prompt and the rest is reached through two meta-tools: `search_tools` (keywords → matching tool names and descriptions) and `load_tool` (exact names → callable for the rest of the conversation). One search surface per agent: API, MCP and connected-app tools share it. `providers` scopes it to a list of provider ids (API / MCP tools) and connected-app toolkit slugs — only their tools are searchable, everything else stays in the prompt; omit it to search the whole roster. The always-loaded set is the rendering components, the large-response retrieval tools, the workflow builder, every tool with `require_approval`, the connection tools, `discovery: false` entries, and whatever `always_loaded` names (tool names or ids). `top_k` is the number of search results (1–20, default 8). `show_catalog: true` also lists every searchable tool's name and a short description in the prompt, trading tokens for one fewer round-trip — worth it for a few dozen tools, not for hundreds. Under the threshold, or with `enabled: false`, your own API and MCP tools all go into the prompt and connected-app toolkits keep their automatic pooling above ~15 direct tools (see `agentTools`). Send null to go back to the platform default.","properties":{"enabled":{"type":"boolean","default":true},"threshold":{"type":"integer","minimum":0,"maximum":1000,"default":40},"providers":{"type":"array","items":{"type":"string"}},"always_loaded":{"type":"array","items":{"type":"string"}},"top_k":{"type":"integer","minimum":1,"maximum":20,"default":8},"show_catalog":{"type":"boolean","default":false}},"example":{"threshold":40,"providers":["8d3b1a75-6c02-4e59-b84f-27a9d5e10c63"],"always_loaded":["BackyouApi_request_hotel_classics"]},"uniqueKey":"tool_search.oneOf.1","typeLabel":"null"}],"uniqueKey":"tool_search","typeLabel":"object | null"},"guardrails":{"oneOf":[{"type":"array","maxItems":12,"description":"Guardrails: safety checks that run on every turn of this agent, one entry per kind (see the guardrails guide). `kind`: `pii` (personal data in messages: emails, phone numbers, cards, ids…), `moderation` (hate, harassment, violence, sexual content, self-harm), `prompt_injection` (jailbreaks and attempts to override the instructions), `system_prompt_scrub` (the agent quoting its own instructions back), `language` (messages outside the target languages) and `unicode` (control characters and odd whitespace in the user message). `mode` says what happens when a guardrail fires: `warn` (default) records the violation and lets the turn continue; `block` ends the turn with an error the chat shows. All kinds but `unicode` ask a small judge model; that spend bills the organization's \"LLM Judges\" group, never a customer group, and a judge-backed guardrail is skipped (with a warning in the rebuild report) when the organization has no judge key. In `block` mode a judge that fails also stops the turn (fail-closed); in `warn` mode it lets the message through. `enabled: false` keeps an entry but turns it off. `options` tunes the kind — every key optional, unknown keys are rejected: `threshold` (0–1 confidence to fire; the defaults are 0.6 for pii, 0.5 for moderation, 0.7 for prompt_injection and language), `model` (a judge model slug, default a small fast one), `instructions` (custom judge instructions, up to 4000 characters), `strategy` (an action other than warn/block when the guardrail fires: `redact` or `filter` for pii and system_prompt_scrub, `rewrite` or `filter` for prompt_injection, `filter` for moderation, `translate` or `detect` for language — `mode` then only decides what a judge failure does), `stage` (`input`, `output` or `both`; pii and moderation only, default `input`), `detection_types` (pii, prompt_injection), `redaction_method` (pii: `mask`, `hash`, `remove`, `placeholder`; system_prompt_scrub: `mask`, `placeholder`, `remove`), `categories` (moderation), `target_languages` and `min_text_length` (language), `custom_patterns` and `placeholder_text` (system_prompt_scrub), `strip_control_chars`, `collapse_whitespace`, `preserve_emojis` and `trim` (unicode, all default true). Send null to remove every guardrail.","items":{"type":"object","required":["kind"],"properties":{"kind":{"type":"string","enum":["pii","moderation","prompt_injection","unicode","language","system_prompt_scrub"],"isRequired":true,"uniqueKey":"guardrails.oneOf.0.items.[INDEX].kind","typeLabel":"enum\u003cstring>"},"mode":{"type":"string","enum":["warn","block"],"default":"warn","uniqueKey":"guardrails.oneOf.0.items.[INDEX].mode","typeLabel":"enum\u003cstring>"},"enabled":{"type":"boolean","default":true,"uniqueKey":"guardrails.oneOf.0.items.[INDEX].enabled","typeLabel":"boolean"},"options":{"type":"object","uniqueKey":"guardrails.oneOf.0.items.[INDEX].options","typeLabel":"object"}},"uniqueKey":"guardrails.oneOf.0.items.[INDEX]","typeLabel":"object"},"example":[{"kind":"prompt_injection","mode":"block"},{"kind":"pii","mode":"warn","options":{"strategy":"redact","detection_types":["email","phone","credit-card"]}},{"kind":"moderation","mode":"block","options":{"categories":["hate","harassment","violence"],"threshold":0.7}}],"uniqueKey":"guardrails.oneOf.0","typeLabel":"object[]"},{"type":"null","maxItems":12,"description":"Guardrails: safety checks that run on every turn of this agent, one entry per kind (see the guardrails guide). `kind`: `pii` (personal data in messages: emails, phone numbers, cards, ids…), `moderation` (hate, harassment, violence, sexual content, self-harm), `prompt_injection` (jailbreaks and attempts to override the instructions), `system_prompt_scrub` (the agent quoting its own instructions back), `language` (messages outside the target languages) and `unicode` (control characters and odd whitespace in the user message). `mode` says what happens when a guardrail fires: `warn` (default) records the violation and lets the turn continue; `block` ends the turn with an error the chat shows. All kinds but `unicode` ask a small judge model; that spend bills the organization's \"LLM Judges\" group, never a customer group, and a judge-backed guardrail is skipped (with a warning in the rebuild report) when the organization has no judge key. In `block` mode a judge that fails also stops the turn (fail-closed); in `warn` mode it lets the message through. `enabled: false` keeps an entry but turns it off. `options` tunes the kind — every key optional, unknown keys are rejected: `threshold` (0–1 confidence to fire; the defaults are 0.6 for pii, 0.5 for moderation, 0.7 for prompt_injection and language), `model` (a judge model slug, default a small fast one), `instructions` (custom judge instructions, up to 4000 characters), `strategy` (an action other than warn/block when the guardrail fires: `redact` or `filter` for pii and system_prompt_scrub, `rewrite` or `filter` for prompt_injection, `filter` for moderation, `translate` or `detect` for language — `mode` then only decides what a judge failure does), `stage` (`input`, `output` or `both`; pii and moderation only, default `input`), `detection_types` (pii, prompt_injection), `redaction_method` (pii: `mask`, `hash`, `remove`, `placeholder`; system_prompt_scrub: `mask`, `placeholder`, `remove`), `categories` (moderation), `target_languages` and `min_text_length` (language), `custom_patterns` and `placeholder_text` (system_prompt_scrub), `strip_control_chars`, `collapse_whitespace`, `preserve_emojis` and `trim` (unicode, all default true). Send null to remove every guardrail.","items":{"type":"object","required":["kind"],"properties":{"kind":{"type":"string","enum":["pii","moderation","prompt_injection","unicode","language","system_prompt_scrub"]},"mode":{"type":"string","enum":["warn","block"],"default":"warn"},"enabled":{"type":"boolean","default":true},"options":{"type":"object"}}},"example":[{"kind":"prompt_injection","mode":"block"},{"kind":"pii","mode":"warn","options":{"strategy":"redact","detection_types":["email","phone","credit-card"]}},{"kind":"moderation","mode":"block","options":{"categories":["hate","harassment","violence"],"threshold":0.7}}],"uniqueKey":"guardrails.oneOf.1","typeLabel":"null"}],"uniqueKey":"guardrails","typeLabel":"object[] | null"},"workflows":{"type":"array","description":"Workflows the agent can run, as `{ workflow_id, config? }` entries. `workflow_id` is the `/v1/workflows` id and must belong to the same organization. The workflow's ACTIVE version is always what runs, so activating a new version reaches the agent on its next turn with no republish. The agent-facing tool name and its description come from the workflow itself — the model calls it as `workflow-\u003cworkflow name>`. The whole workflow runs inside one tool call, and its step progress is streamed to the client as `data-tool-workflow` / `data-tool-workflow-step` parts by /v1/chat.","items":{"type":"object","required":["workflow_id"],"properties":{"workflow_id":{"type":"string","isRequired":true,"uniqueKey":"workflows.items.[INDEX].workflow_id","typeLabel":"string"},"config":{"type":"object","uniqueKey":"workflows.items.[INDEX].config","typeLabel":"object"}},"uniqueKey":"workflows.items.[INDEX]","typeLabel":"object"},"example":[{"workflow_id":"4c2f9e18-7a63-4d05-b1e8-93a7c0f2d641"}],"uniqueKey":"workflows","typeLabel":"object[]"},"skills":{"oneOf":[{"type":"array","description":"Skills the agent can load, as `{ skill_id }` entries. `skill_id` is the `/v1/skills` id and must belong to the same organization. A skill is a named block of instructions the agent loads on demand — it calls its `skill` tool with the name when the skill's description says to — instead of carrying the text in every prompt. The agent always gets the skill's ACTIVE version, so publishing a new one reaches the agent on its next turn with no republish; a skill that is not published yet is skipped and reported in the rebuild summary. Send null to remove every skill.","items":{"type":"object","required":["skill_id"],"properties":{"skill_id":{"type":"string","isRequired":true,"uniqueKey":"skills.oneOf.0.items.[INDEX].skill_id","typeLabel":"string"}},"uniqueKey":"skills.oneOf.0.items.[INDEX]","typeLabel":"object"},"example":[{"skill_id":"9a4e2c71-5b08-4f3d-a6c2-1e7d8b0f4c23"}],"uniqueKey":"skills.oneOf.0","typeLabel":"object[]"},{"type":"null","description":"Skills the agent can load, as `{ skill_id }` entries. `skill_id` is the `/v1/skills` id and must belong to the same organization. A skill is a named block of instructions the agent loads on demand — it calls its `skill` tool with the name when the skill's description says to — instead of carrying the text in every prompt. The agent always gets the skill's ACTIVE version, so publishing a new one reaches the agent on its next turn with no republish; a skill that is not published yet is skipped and reported in the rebuild summary. Send null to remove every skill.","items":{"type":"object","required":["skill_id"],"properties":{"skill_id":{"type":"string"}}},"example":[{"skill_id":"9a4e2c71-5b08-4f3d-a6c2-1e7d8b0f4c23"}],"uniqueKey":"skills.oneOf.1","typeLabel":"null"}],"uniqueKey":"skills","typeLabel":"object[] | null"},"agentTools":{"type":"array","description":"Tools assigned to the agent. Each entry is `{ tool_id, provider_id?, config?, composio_connected_account_id?, discovery?, loop_responses?, require_approval? }`. The `tool_name` / `provider_name` a read returns are ignored here (resolved again on the way out), so an entry can be sent back as it was read. `discovery` (integration tools only) controls where the tool schema lives at runtime: `true` puts it behind the agent's search_tools/load_tool discovery tools, `false` pins it into the prompt, and unset lets the platform decide — small toolkits stay in the prompt, large ones (beyond ~15 total tools) become discoverable so the prompt is not flooded with schemas. `require_approval: true` pauses every call to that tool for human approval: the chat stream emits a `data-tool-call-approval` chunk instead of executing (carrying a model-written `reason` sentence for the approval card), and the client continues the run by POSTing /v1/chat with the `approval` field. Set `loop_responses: true` on a tool to let one call cover a whole result set: it gains `_fields` (project only the fields needed), `_all_pages` (fetch and merge every page) and `_for_each` (call once per value of a parameter and merge the rows), and the agent is given the large-response retrieval tools automatically. Because those re-issue the request, enable it only on tools whose endpoint is safe to call repeatedly.","items":{"type":"object","uniqueKey":"agentTools.items.[INDEX]","typeLabel":"object"},"example":[{"tool_id":"c07f2b95-8d41-4e63-a029-5f7b1c8d3e40","provider_id":"8d3b1a75-6c02-4e59-b84f-27a9d5e10c63"}],"uniqueKey":"agentTools","typeLabel":"object[]"},"name":{"type":"string","description":"The agent's name. Agent-level, not versioned — it takes effect immediately, with no publish, and is what other agents see when this one is a subagent.","example":"Support agent","uniqueKey":"name","typeLabel":"string"},"description":{"type":"string","description":"The agent's description. Agent-level, not versioned — same immediate effect as `name`. Send null to clear it.","example":"Answers billing and account questions","uniqueKey":"description","typeLabel":"string"},"scorers":{"type":"array","description":"Scorers for this version. Each entry: `name` (unique key), `scorerType` (catalogue id), `model` (judge slug, LLM scorers only), `sampling_rate` (0-1, share of eligible runs scored, default 1), `options`, and an optional `filter` — a predicate over the run, evaluated before sampling, over `requestContext.*` (groupId, environment, copilotId, source, userId…), `entity.*`, `threadId`, `resourceId`. Operators: eq/ne/lt/lte/gt/gte (`left`/`right` as `{ path }` or `{ literal }`), in/notIn (`value`, `set`), exists/notExists (`path`), truthy/falsy (`value`), and/or (`args`), not (`arg`). An invalid filter answers 400.","items":{"type":"object","uniqueKey":"scorers.items.[INDEX]","typeLabel":"object"},"example":[{"name":"faithfulness","scorerType":"faithfulness","model":"openai/gpt-4o-mini","sampling_rate":0.2,"filter":{"op":"and","args":[{"op":"eq","left":{"path":"requestContext.environment"},"right":{"literal":"production"}},{"op":"in","value":{"path":"requestContext.groupId"},"set":["grp_vip","grp_enterprise"]}]}}],"uniqueKey":"scorers","typeLabel":"object[]"},"agents":{"type":"array","description":"Subagents this one may delegate to.","items":{"type":"object","properties":{"agentId":{"type":"string","uniqueKey":"agents.items.[INDEX].agentId","typeLabel":"string"}},"uniqueKey":"agents.items.[INDEX]","typeLabel":"object"},"example":[{"agentId":"6f1c0f4e-2b7a-4a51-9a2f-0c9d1b3e5a10"}],"uniqueKey":"agents","typeLabel":"object[]"},"memory":{"oneOf":[{"type":"object","properties":{"options":{"type":"object","properties":{"lastMessages":{"oneOf":[{"type":"boolean","description":"Recent messages read back on every turn. Default 10; `false` or 0 reads none.","enum":[false],"uniqueKey":"memory.oneOf.0.options.lastMessages.oneOf.0","typeLabel":"enum\u003cboolean>"},{"type":"integer","description":"Recent messages read back on every turn. Default 10; `false` or 0 reads none.","maximum":200,"minimum":0,"uniqueKey":"memory.oneOf.0.options.lastMessages.oneOf.1","typeLabel":"integer"}],"uniqueKey":"memory.oneOf.0.options.lastMessages","typeLabel":"enum\u003cboolean> | integer"},"readOnly":{"type":"boolean","description":"Read memory but save nothing. Default false.","uniqueKey":"memory.oneOf.0.options.readOnly","typeLabel":"boolean"},"generateTitle":{"oneOf":[{"type":"boolean","description":"Name the thread from its first message. Default on, with the platform title model.","uniqueKey":"memory.oneOf.0.options.generateTitle.oneOf.0","typeLabel":"boolean"},{"type":"object","description":"Name the thread from its first message. Default on, with the platform title model.","properties":{"model":{"type":"string","minLength":1,"maxLength":200,"description":"Model slug for the title. Default: the platform title model.","uniqueKey":"memory.oneOf.0.options.generateTitle.oneOf.1.model","typeLabel":"string"},"instructions":{"type":"string","minLength":1,"maxLength":2000,"description":"How to write the title.","uniqueKey":"memory.oneOf.0.options.generateTitle.oneOf.1.instructions","typeLabel":"string"}},"additionalProperties":false,"uniqueKey":"memory.oneOf.0.options.generateTitle.oneOf.1","typeLabel":"object"}],"uniqueKey":"memory.oneOf.0.options.generateTitle","typeLabel":"boolean | object"},"semanticRecall":{"oneOf":[{"type":"boolean","description":"Recall similar past messages by meaning. Default off.","uniqueKey":"memory.oneOf.0.options.semanticRecall.oneOf.0","typeLabel":"boolean"},{"type":"object","description":"Recall similar past messages by meaning. Default off.","properties":{"topK":{"type":"integer","minimum":1,"maximum":50,"description":"How many similar past messages to recall. Default 5.","uniqueKey":"memory.oneOf.0.options.semanticRecall.oneOf.1.topK","typeLabel":"integer"},"messageRange":{"oneOf":[{"type":"integer","description":"Context messages around each match, one number or `{ before, after }`. Default 2.","maximum":20,"minimum":0,"uniqueKey":"memory.oneOf.0.options.semanticRecall.oneOf.1.messageRange.oneOf.0","typeLabel":"integer"},{"type":"object","description":"Context messages around each match, one number or `{ before, after }`. Default 2.","required":["before","after"],"properties":{"before":{"type":"integer","minimum":0,"maximum":20,"isRequired":true,"uniqueKey":"memory.oneOf.0.options.semanticRecall.oneOf.1.messageRange.oneOf.1.before","typeLabel":"integer"},"after":{"type":"integer","minimum":0,"maximum":20,"isRequired":true,"uniqueKey":"memory.oneOf.0.options.semanticRecall.oneOf.1.messageRange.oneOf.1.after","typeLabel":"integer"}},"additionalProperties":false,"uniqueKey":"memory.oneOf.0.options.semanticRecall.oneOf.1.messageRange.oneOf.1","typeLabel":"object"}],"uniqueKey":"memory.oneOf.0.options.semanticRecall.oneOf.1.messageRange","typeLabel":"integer | object"},"scope":{"type":"string","enum":["thread","resource"],"description":"`resource` (default): across all of the user's conversations; `thread`: this conversation only.","uniqueKey":"memory.oneOf.0.options.semanticRecall.oneOf.1.scope","typeLabel":"enum\u003cstring>"}},"additionalProperties":false,"uniqueKey":"memory.oneOf.0.options.semanticRecall.oneOf.1","typeLabel":"object"}],"uniqueKey":"memory.oneOf.0.options.semanticRecall","typeLabel":"boolean | object"},"workingMemory":{"oneOf":[{"type":"boolean","description":"The agent's standing notes about the user. Default off.","uniqueKey":"memory.oneOf.0.options.workingMemory.oneOf.0","typeLabel":"boolean"},{"type":"object","description":"The agent's standing notes about the user. Default off.","properties":{"enabled":{"type":"boolean","description":"Default true when this object is sent.","uniqueKey":"memory.oneOf.0.options.workingMemory.oneOf.1.enabled","typeLabel":"boolean"},"scope":{"type":"string","enum":["thread","resource"],"description":"`resource` (default): one record per user, shared by every conversation and automation run; `thread`: one per conversation.","uniqueKey":"memory.oneOf.0.options.workingMemory.oneOf.1.scope","typeLabel":"enum\u003cstring>"},"template":{"type":"string","minLength":1,"maxLength":20000,"description":"Markdown template of the notes. Exclusive with `schema`.","uniqueKey":"memory.oneOf.0.options.workingMemory.oneOf.1.template","typeLabel":"string"},"schema":{"type":"object","additionalProperties":{"oneOf":[{}],"uniqueKey":"memory.oneOf.0.options.workingMemory.oneOf.1.schema.additionalProperties","typeLabel":"any"},"description":"JSON Schema of the notes, kept as a JSON object. Exclusive with `template`.","uniqueKey":"memory.oneOf.0.options.workingMemory.oneOf.1.schema","typeLabel":"object"},"agentManaged":{"type":"boolean","description":"Whether the agent itself gets the tool and instructions to update its notes. Default true; default false when the observer manages them (`observationalMemory.observation.manageWorkingMemory`).","uniqueKey":"memory.oneOf.0.options.workingMemory.oneOf.1.agentManaged","typeLabel":"boolean"}},"additionalProperties":false,"uniqueKey":"memory.oneOf.0.options.workingMemory.oneOf.1","typeLabel":"object"}],"uniqueKey":"memory.oneOf.0.options.workingMemory","typeLabel":"boolean | object"},"observationalMemory":{"oneOf":[{"type":"boolean","description":"The background observer. Default off.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.0","typeLabel":"boolean"},{"type":"object","description":"The background observer. Default off.","properties":{"enabled":{"type":"boolean","description":"Default true when this object is sent.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.enabled","typeLabel":"boolean"},"model":{"type":"string","minLength":1,"maxLength":200,"description":"Observer and reflector model. Default: the platform observer model. Exclusive with `observation.model` / `reflection.model`.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.model","typeLabel":"string"},"scope":{"type":"string","enum":["thread","resource"],"description":"`thread` only.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.scope","typeLabel":"enum\u003cstring>"},"activateAfterIdle":{"oneOf":[{"type":"string","description":"Activate buffered observations once the conversation has been idle this long, before the threshold. Milliseconds, `\"5m\"`, `\"auto\"` or `false`.","enum":["auto"],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.activateAfterIdle.oneOf.0","typeLabel":"enum\u003cstring>"},{"type":"boolean","description":"Activate buffered observations once the conversation has been idle this long, before the threshold. Milliseconds, `\"5m\"`, `\"auto\"` or `false`.","enum":[false],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.activateAfterIdle.oneOf.1","typeLabel":"enum\u003cboolean>"},{"type":"integer","description":"Activate buffered observations once the conversation has been idle this long, before the threshold. Milliseconds, `\"5m\"`, `\"auto\"` or `false`.","maximum":604800000,"minimum":0,"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.activateAfterIdle.oneOf.2","typeLabel":"integer"},{"type":"string","description":"Activate buffered observations once the conversation has been idle this long, before the threshold. Milliseconds, `\"5m\"`, `\"auto\"` or `false`.","pattern":"^\\d+(?:\\.\\d+)?\\s*(ms|msec|msecs|millisecond|milliseconds|s|sec|secs|second|seconds|m|min|mins|minute|minutes|h|hr|hrs|hour|hours)$","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.activateAfterIdle.oneOf.3","typeLabel":"string"}],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.activateAfterIdle","typeLabel":"enum\u003cstring> | enum\u003cboolean> | integer | string"},"retrieval":{"oneOf":[{"type":"boolean","description":"Let the agent read the raw messages behind its observations through a `recall` tool.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.retrieval.oneOf.0","typeLabel":"boolean"},{"type":"object","description":"Let the agent read the raw messages behind its observations through a `recall` tool.","properties":{"vector":{"type":"boolean","description":"Also search the raw messages by meaning (uses the platform embeddings). Default false.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.retrieval.oneOf.1.vector","typeLabel":"boolean"},"scope":{"type":"string","enum":["thread","resource"],"description":"`resource` (default): the agent may browse the user's other conversations; `thread`: this conversation only.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.retrieval.oneOf.1.scope","typeLabel":"enum\u003cstring>"},"instructions":{"type":"string","minLength":1,"maxLength":4000,"description":"Guidance appended to the recall tool's built-in instructions.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.retrieval.oneOf.1.instructions","typeLabel":"string"}},"additionalProperties":false,"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.retrieval.oneOf.1","typeLabel":"object"}],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.retrieval","typeLabel":"boolean | object"},"observation":{"type":"object","properties":{"model":{"type":"string","minLength":1,"maxLength":200,"description":"Observer model. Exclusive with the top-level `model`.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.model","typeLabel":"string"},"messageTokens":{"type":"integer","minimum":2000,"maximum":1000000,"description":"Message tokens kept before observations replace raw history. Default 30000.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.messageTokens","typeLabel":"integer"},"bufferTokens":{"oneOf":[{"type":"boolean","description":"How often the observer runs in the background: a fraction of `messageTokens` (default 0.15) or a token count; `false` makes every observation synchronous (not recommended).","enum":[false],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.bufferTokens.oneOf.0","typeLabel":"enum\u003cboolean>"},{"type":"number","description":"How often the observer runs in the background: a fraction of `messageTokens` (default 0.15) or a token count; `false` makes every observation synchronous (not recommended).","maximum":1,"minimum":0,"exclusiveMaximum":true,"exclusiveMinimum":true,"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.bufferTokens.oneOf.1","typeLabel":"number"},{"type":"integer","description":"How often the observer runs in the background: a fraction of `messageTokens` (default 0.15) or a token count; `false` makes every observation synchronous (not recommended).","minimum":500,"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.bufferTokens.oneOf.2","typeLabel":"integer"}],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.bufferTokens","typeLabel":"enum\u003cboolean> | number | integer"},"bufferActivation":{"type":"number","exclusiveMinimum":0,"maximum":1,"description":"Share of buffered observations activated at the threshold (0-1). Default 0.8.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.bufferActivation","typeLabel":"number"},"blockAfter":{"type":"number","minimum":1,"exclusiveMaximum":2,"description":"Multiplier of `messageTokens` past which an observation is forced synchronously. Default 1.9.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.blockAfter","typeLabel":"number"},"bufferOnIdle":{"type":"boolean","description":"Also observe at the end of every turn, however short. Default false. Required for the observer to keep working memory and titles current on single-turn threads (automation runs).","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.bufferOnIdle","typeLabel":"boolean"},"manageWorkingMemory":{"type":"boolean","description":"The observer maintains the working memory in the background; the agent no longer has to. Needs `workingMemory.enabled`. Default false.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.manageWorkingMemory","typeLabel":"boolean"},"threadTitle":{"type":"boolean","description":"The observer rewrites the thread title when the topic meaningfully changes. Default false.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.threadTitle","typeLabel":"boolean"},"instruction":{"type":"string","minLength":1,"maxLength":4000,"description":"Extra guidance appended to the observer's prompt (what to focus on).","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.instruction","typeLabel":"string"},"activateAfterIdle":{"oneOf":[{"type":"string","description":"Overrides the top-level `activateAfterIdle` for observations; `false` disables it.","enum":["auto"],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.activateAfterIdle.oneOf.0","typeLabel":"enum\u003cstring>"},{"type":"boolean","description":"Overrides the top-level `activateAfterIdle` for observations; `false` disables it.","enum":[false],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.activateAfterIdle.oneOf.1","typeLabel":"enum\u003cboolean>"},{"type":"integer","description":"Overrides the top-level `activateAfterIdle` for observations; `false` disables it.","maximum":604800000,"minimum":0,"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.activateAfterIdle.oneOf.2","typeLabel":"integer"},{"type":"string","description":"Overrides the top-level `activateAfterIdle` for observations; `false` disables it.","pattern":"^\\d+(?:\\.\\d+)?\\s*(ms|msec|msecs|millisecond|milliseconds|s|sec|secs|second|seconds|m|min|mins|minute|minutes|h|hr|hrs|hour|hours)$","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.activateAfterIdle.oneOf.3","typeLabel":"string"}],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation.activateAfterIdle","typeLabel":"enum\u003cstring> | enum\u003cboolean> | integer | string"}},"additionalProperties":false,"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.observation","typeLabel":"object"},"reflection":{"type":"object","properties":{"model":{"type":"string","minLength":1,"maxLength":200,"description":"Reflector model. Exclusive with the top-level `model`.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.reflection.model","typeLabel":"string"},"observationTokens":{"type":"integer","minimum":2000,"maximum":1000000,"description":"Observation-log tokens before the reflector condenses it. Default 40000.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.reflection.observationTokens","typeLabel":"integer"},"bufferActivation":{"type":"number","exclusiveMinimum":0,"maximum":1,"description":"Share of the threshold at which background reflection starts (0-1). Default 0.5.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.reflection.bufferActivation","typeLabel":"number"},"blockAfter":{"type":"number","minimum":1,"exclusiveMaximum":2,"description":"Multiplier of `observationTokens` past which a reflection is forced synchronously. Default 1.9.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.reflection.blockAfter","typeLabel":"number"},"activateAfterIdle":{"oneOf":[{"type":"string","description":"Idle activation for reflections (not inherited from the top level).","enum":["auto"],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.reflection.activateAfterIdle.oneOf.0","typeLabel":"enum\u003cstring>"},{"type":"boolean","description":"Idle activation for reflections (not inherited from the top level).","enum":[false],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.reflection.activateAfterIdle.oneOf.1","typeLabel":"enum\u003cboolean>"},{"type":"integer","description":"Idle activation for reflections (not inherited from the top level).","maximum":604800000,"minimum":0,"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.reflection.activateAfterIdle.oneOf.2","typeLabel":"integer"},{"type":"string","description":"Idle activation for reflections (not inherited from the top level).","pattern":"^\\d+(?:\\.\\d+)?\\s*(ms|msec|msecs|millisecond|milliseconds|s|sec|secs|second|seconds|m|min|mins|minute|minutes|h|hr|hrs|hour|hours)$","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.reflection.activateAfterIdle.oneOf.3","typeLabel":"string"}],"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.reflection.activateAfterIdle","typeLabel":"enum\u003cstring> | enum\u003cboolean> | integer | string"},"instruction":{"type":"string","minLength":1,"maxLength":4000,"description":"Extra guidance appended to the reflector's prompt.","uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.reflection.instruction","typeLabel":"string"}},"additionalProperties":false,"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1.reflection","typeLabel":"object"}},"additionalProperties":false,"uniqueKey":"memory.oneOf.0.options.observationalMemory.oneOf.1","typeLabel":"object"}],"uniqueKey":"memory.oneOf.0.options.observationalMemory","typeLabel":"boolean | object"}},"additionalProperties":false,"uniqueKey":"memory.oneOf.0.options","typeLabel":"object"}},"additionalProperties":false,"description":"Memory settings of this version, as `{ options }` — see the memory guide. `options`: `lastMessages` (recent messages read back, default 10), `generateTitle` (first title from the first message; `true`, `false` or `{ model, instructions }`), `semanticRecall` (`true`, `false` or `{ topK, messageRange, scope }`), `workingMemory` (`true`, `false` or `{ enabled, scope, template | schema, agentManaged }` — `scope: \"resource\"`, the default, is one record per user shared by every conversation and automation run), `observationalMemory` (`true`, `false` or `{ model, activateAfterIdle, retrieval, observation: { messageTokens, bufferTokens, bufferActivation, blockAfter, bufferOnIdle, manageWorkingMemory, threadTitle, instruction, activateAfterIdle }, reflection: { observationTokens, bufferActivation, blockAfter, activateAfterIdle, instruction } }`; its `scope` is `thread` only). Unknown keys and out-of-range values answer 400 naming the path. Null = platform defaults. Model slugs bill the organization's key.","example":{"options":{"lastMessages":20,"workingMemory":{"enabled":true,"scope":"resource"},"observationalMemory":{"enabled":true,"observation":{"manageWorkingMemory":true,"bufferOnIdle":true,"threadTitle":true}}}},"uniqueKey":"memory.oneOf.0","typeLabel":"object"},{"type":"null","properties":{"options":{"type":"object","properties":{"lastMessages":{"oneOf":[{"type":"boolean","description":"Recent messages read back on every turn. Default 10; `false` or 0 reads none.","enum":[false]},{"type":"integer","description":"Recent messages read back on every turn. Default 10; `false` or 0 reads none.","maximum":200,"minimum":0}]},"readOnly":{"type":"boolean","description":"Read memory but save nothing. Default false."},"generateTitle":{"oneOf":[{"type":"boolean","description":"Name the thread from its first message. Default on, with the platform title model."},{"type":"object","description":"Name the thread from its first message. Default on, with the platform title model.","properties":{"model":{"type":"string","minLength":1,"maxLength":200,"description":"Model slug for the title. Default: the platform title model."},"instructions":{"type":"string","minLength":1,"maxLength":2000,"description":"How to write the title."}},"additionalProperties":false}]},"semanticRecall":{"oneOf":[{"type":"boolean","description":"Recall similar past messages by meaning. Default off."},{"type":"object","description":"Recall similar past messages by meaning. Default off.","properties":{"topK":{"type":"integer","minimum":1,"maximum":50,"description":"How many similar past messages to recall. Default 5."},"messageRange":{"oneOf":[{"type":"integer","description":"Context messages around each match, one number or `{ before, after }`. Default 2.","maximum":20,"minimum":0},{"type":"object","description":"Context messages around each match, one number or `{ before, after }`. Default 2.","required":["before","after"],"properties":{"before":{"type":"integer","minimum":0,"maximum":20},"after":{"type":"integer","minimum":0,"maximum":20}},"additionalProperties":false}]},"scope":{"type":"string","enum":["thread","resource"],"description":"`resource` (default): across all of the user's conversations; `thread`: this conversation only."}},"additionalProperties":false}]},"workingMemory":{"oneOf":[{"type":"boolean","description":"The agent's standing notes about the user. Default off."},{"type":"object","description":"The agent's standing notes about the user. Default off.","properties":{"enabled":{"type":"boolean","description":"Default true when this object is sent."},"scope":{"type":"string","enum":["thread","resource"],"description":"`resource` (default): one record per user, shared by every conversation and automation run; `thread`: one per conversation."},"template":{"type":"string","minLength":1,"maxLength":20000,"description":"Markdown template of the notes. Exclusive with `schema`."},"schema":{"type":"object","additionalProperties":{"oneOf":[{}]},"description":"JSON Schema of the notes, kept as a JSON object. Exclusive with `template`."},"agentManaged":{"type":"boolean","description":"Whether the agent itself gets the tool and instructions to update its notes. Default true; default false when the observer manages them (`observationalMemory.observation.manageWorkingMemory`)."}},"additionalProperties":false}]},"observationalMemory":{"oneOf":[{"type":"boolean","description":"The background observer. Default off."},{"type":"object","description":"The background observer. Default off.","properties":{"enabled":{"type":"boolean","description":"Default true when this object is sent."},"model":{"type":"string","minLength":1,"maxLength":200,"description":"Observer and reflector model. Default: the platform observer model. Exclusive with `observation.model` / `reflection.model`."},"scope":{"type":"string","enum":["thread","resource"],"description":"`thread` only."},"activateAfterIdle":{"oneOf":[{"type":"string","description":"Activate buffered observations once the conversation has been idle this long, before the threshold. Milliseconds, `\"5m\"`, `\"auto\"` or `false`.","enum":["auto"]},{"type":"boolean","description":"Activate buffered observations once the conversation has been idle this long, before the threshold. Milliseconds, `\"5m\"`, `\"auto\"` or `false`.","enum":[false]},{"type":"integer","description":"Activate buffered observations once the conversation has been idle this long, before the threshold. Milliseconds, `\"5m\"`, `\"auto\"` or `false`.","maximum":604800000,"minimum":0},{"type":"string","description":"Activate buffered observations once the conversation has been idle this long, before the threshold. Milliseconds, `\"5m\"`, `\"auto\"` or `false`.","pattern":"^\\d+(?:\\.\\d+)?\\s*(ms|msec|msecs|millisecond|milliseconds|s|sec|secs|second|seconds|m|min|mins|minute|minutes|h|hr|hrs|hour|hours)$"}]},"retrieval":{"oneOf":[{"type":"boolean","description":"Let the agent read the raw messages behind its observations through a `recall` tool."},{"type":"object","description":"Let the agent read the raw messages behind its observations through a `recall` tool.","properties":{"vector":{"type":"boolean","description":"Also search the raw messages by meaning (uses the platform embeddings). Default false."},"scope":{"type":"string","enum":["thread","resource"],"description":"`resource` (default): the agent may browse the user's other conversations; `thread`: this conversation only."},"instructions":{"type":"string","minLength":1,"maxLength":4000,"description":"Guidance appended to the recall tool's built-in instructions."}},"additionalProperties":false}]},"observation":{"type":"object","properties":{"model":{"type":"string","minLength":1,"maxLength":200,"description":"Observer model. Exclusive with the top-level `model`."},"messageTokens":{"type":"integer","minimum":2000,"maximum":1000000,"description":"Message tokens kept before observations replace raw history. Default 30000."},"bufferTokens":{"oneOf":[{"type":"boolean","description":"How often the observer runs in the background: a fraction of `messageTokens` (default 0.15) or a token count; `false` makes every observation synchronous (not recommended).","enum":[false]},{"type":"number","description":"How often the observer runs in the background: a fraction of `messageTokens` (default 0.15) or a token count; `false` makes every observation synchronous (not recommended).","maximum":1,"minimum":0,"exclusiveMaximum":true,"exclusiveMinimum":true},{"type":"integer","description":"How often the observer runs in the background: a fraction of `messageTokens` (default 0.15) or a token count; `false` makes every observation synchronous (not recommended).","minimum":500}]},"bufferActivation":{"type":"number","exclusiveMinimum":0,"maximum":1,"description":"Share of buffered observations activated at the threshold (0-1). Default 0.8."},"blockAfter":{"type":"number","minimum":1,"exclusiveMaximum":2,"description":"Multiplier of `messageTokens` past which an observation is forced synchronously. Default 1.9."},"bufferOnIdle":{"type":"boolean","description":"Also observe at the end of every turn, however short. Default false. Required for the observer to keep working memory and titles current on single-turn threads (automation runs)."},"manageWorkingMemory":{"type":"boolean","description":"The observer maintains the working memory in the background; the agent no longer has to. Needs `workingMemory.enabled`. Default false."},"threadTitle":{"type":"boolean","description":"The observer rewrites the thread title when the topic meaningfully changes. Default false."},"instruction":{"type":"string","minLength":1,"maxLength":4000,"description":"Extra guidance appended to the observer's prompt (what to focus on)."},"activateAfterIdle":{"oneOf":[{"type":"string","description":"Overrides the top-level `activateAfterIdle` for observations; `false` disables it.","enum":["auto"]},{"type":"boolean","description":"Overrides the top-level `activateAfterIdle` for observations; `false` disables it.","enum":[false]},{"type":"integer","description":"Overrides the top-level `activateAfterIdle` for observations; `false` disables it.","maximum":604800000,"minimum":0},{"type":"string","description":"Overrides the top-level `activateAfterIdle` for observations; `false` disables it.","pattern":"^\\d+(?:\\.\\d+)?\\s*(ms|msec|msecs|millisecond|milliseconds|s|sec|secs|second|seconds|m|min|mins|minute|minutes|h|hr|hrs|hour|hours)$"}]}},"additionalProperties":false},"reflection":{"type":"object","properties":{"model":{"type":"string","minLength":1,"maxLength":200,"description":"Reflector model. Exclusive with the top-level `model`."},"observationTokens":{"type":"integer","minimum":2000,"maximum":1000000,"description":"Observation-log tokens before the reflector condenses it. Default 40000."},"bufferActivation":{"type":"number","exclusiveMinimum":0,"maximum":1,"description":"Share of the threshold at which background reflection starts (0-1). Default 0.5."},"blockAfter":{"type":"number","minimum":1,"exclusiveMaximum":2,"description":"Multiplier of `observationTokens` past which a reflection is forced synchronously. Default 1.9."},"activateAfterIdle":{"oneOf":[{"type":"string","description":"Idle activation for reflections (not inherited from the top level).","enum":["auto"]},{"type":"boolean","description":"Idle activation for reflections (not inherited from the top level).","enum":[false]},{"type":"integer","description":"Idle activation for reflections (not inherited from the top level).","maximum":604800000,"minimum":0},{"type":"string","description":"Idle activation for reflections (not inherited from the top level).","pattern":"^\\d+(?:\\.\\d+)?\\s*(ms|msec|msecs|millisecond|milliseconds|s|sec|secs|second|seconds|m|min|mins|minute|minutes|h|hr|hrs|hour|hours)$"}]},"instruction":{"type":"string","minLength":1,"maxLength":4000,"description":"Extra guidance appended to the reflector's prompt."}},"additionalProperties":false}},"additionalProperties":false}]}},"additionalProperties":false}},"additionalProperties":false,"description":"Memory settings of this version, as `{ options }` — see the memory guide. `options`: `lastMessages` (recent messages read back, default 10), `generateTitle` (first title from the first message; `true`, `false` or `{ model, instructions }`), `semanticRecall` (`true`, `false` or `{ topK, messageRange, scope }`), `workingMemory` (`true`, `false` or `{ enabled, scope, template | schema, agentManaged }` — `scope: \"resource\"`, the default, is one record per user shared by every conversation and automation run), `observationalMemory` (`true`, `false` or `{ model, activateAfterIdle, retrieval, observation: { messageTokens, bufferTokens, bufferActivation, blockAfter, bufferOnIdle, manageWorkingMemory, threadTitle, instruction, activateAfterIdle }, reflection: { observationTokens, bufferActivation, blockAfter, activateAfterIdle, instruction } }`; its `scope` is `thread` only). Unknown keys and out-of-range values answer 400 naming the path. Null = platform defaults. Model slugs bill the organization's key.","example":{"options":{"lastMessages":20,"workingMemory":{"enabled":true,"scope":"resource"},"observationalMemory":{"enabled":true,"observation":{"manageWorkingMemory":true,"bufferOnIdle":true,"threadTitle":true}}}},"uniqueKey":"memory.oneOf.1","typeLabel":"null"}],"uniqueKey":"memory","typeLabel":"object | null"}},"uniqueKey":"","typeLabel":"object"},"examples":{"Example":{"title":"Example","value":{"instructions":[{"type":"prompt_block_ref","id":"d41a8f6b-9c27-4b13-a5e8-2f60c1d7b394"},{"type":"text","content":"Answer in {{locale || 'English'}}."}],"model_slug":"anthropic/claude-sonnet-4-5","policy_slug":"policy/fastest-sonnet","model_options":{"reasoningEffort":"medium"},"model_settings":{"temperature":0.2},"max_steps":8,"browser":{"enabled":true,"viewport":{"width":1280,"height":800}},"tool_search":{"threshold":40,"providers":["8d3b1a75-6c02-4e59-b84f-27a9d5e10c63"],"always_loaded":["BackyouApi_request_hotel_classics"]},"guardrails":[{"kind":"prompt_injection","mode":"block"},{"kind":"pii","mode":"warn","options":{"strategy":"redact","detection_types":["email","phone","credit-card"]}},{"kind":"moderation","mode":"block","options":{"categories":["hate","harassment","violence"],"threshold":0.7}}],"workflows":[{"workflow_id":"4c2f9e18-7a63-4d05-b1e8-93a7c0f2d641"}],"skills":[{"skill_id":"9a4e2c71-5b08-4f3d-a6c2-1e7d8b0f4c23"}],"agentTools":[{"tool_id":"c07f2b95-8d41-4e63-a029-5f7b1c8d3e40","provider_id":"8d3b1a75-6c02-4e59-b84f-27a9d5e10c63"}],"name":"Support agent","description":"Answers billing and account questions","scorers":[{"name":"faithfulness","scorerType":"faithfulness","model":"openai/gpt-4o-mini","sampling_rate":0.2,"filter":{"op":"and","args":[{"op":"eq","left":{"path":"requestContext.environment"},"right":{"literal":"production"}},{"op":"in","value":{"path":"requestContext.groupId"},"set":["grp_vip","grp_enterprise"]}]}}],"agents":[{"agentId":"6f1c0f4e-2b7a-4a51-9a2f-0c9d1b3e5a10"}],"memory":{"options":{"lastMessages":20,"workingMemory":{"enabled":true,"scope":"resource"},"observationalMemory":{"enabled":true,"observation":{"manageWorkingMemory":true,"bufferOnIdle":true,"threadTitle":true}}}}}}},"exampleType":"generated"}},"exampleType":"generated"},"responses":{"0df57298-0ace-408a-8f6f-aa15ea4f85f0":{"description":"The saved draft, and the agent record it belongs to","content":{"application/json":{"schema":{"type":"object","properties":{"agent":{"type":"object","description":"The agent after the save — `name`/`description` reflect this call, and the version summary says whether the draft now differs from what is published.","properties":{"id":{"type":"string","format":"uuid","uniqueKey":"agent.id","typeLabel":"string\u003cuuid>"},"org_id":{"type":"string","format":"uuid","uniqueKey":"agent.org_id","typeLabel":"string\u003cuuid>"},"name":{"type":"string","uniqueKey":"agent.name","typeLabel":"string"},"description":{"oneOf":[{"type":"string","uniqueKey":"agent.description.oneOf.0","typeLabel":"string"},{"type":"null","uniqueKey":"agent.description.oneOf.1","typeLabel":"null"}],"uniqueKey":"agent.description","typeLabel":"string | null"},"active_version_id":{"oneOf":[{"type":"string","format":"uuid","description":"Null until a published version is activated.","uniqueKey":"agent.active_version_id.oneOf.0","typeLabel":"string\u003cuuid>"},{"type":"null","format":"uuid","description":"Null until a published version is activated.","uniqueKey":"agent.active_version_id.oneOf.1","typeLabel":"null"}],"uniqueKey":"agent.active_version_id","typeLabel":"string\u003cuuid> | null"},"created_at":{"type":"string","format":"date-time","uniqueKey":"agent.created_at","typeLabel":"string\u003cdate-time>"},"updated_at":{"type":"string","format":"date-time","uniqueKey":"agent.updated_at","typeLabel":"string\u003cdate-time>"},"activeVersionNumber":{"oneOf":[{"type":"integer","description":"Version number currently serving traffic. Null until a published version is activated.","uniqueKey":"agent.activeVersionNumber.oneOf.0","typeLabel":"integer"},{"type":"null","description":"Version number currently serving traffic. Null until a published version is activated.","uniqueKey":"agent.activeVersionNumber.oneOf.1","typeLabel":"null"}],"uniqueKey":"agent.activeVersionNumber","typeLabel":"integer | null"},"latestPublishedVersionNumber":{"oneOf":[{"type":"integer","description":"Newest published version number. Higher than `activeVersionNumber` when a newer version is published but not activated. Null before the first publish.","uniqueKey":"agent.latestPublishedVersionNumber.oneOf.0","typeLabel":"integer"},{"type":"null","description":"Newest published version number. Higher than `activeVersionNumber` when a newer version is published but not activated. Null before the first publish.","uniqueKey":"agent.latestPublishedVersionNumber.oneOf.1","typeLabel":"null"}],"uniqueKey":"agent.latestPublishedVersionNumber","typeLabel":"integer | null"},"draftDirty":{"type":"boolean","description":"True when the draft's config differs from the latest published version — i.e. there are unpublished edits. Fetch them with `?version=draft`.","uniqueKey":"agent.draftDirty","typeLabel":"boolean"}},"uniqueKey":"agent","typeLabel":"object"},"draft":{"type":"object","properties":{"id":{"type":"string","format":"uuid","uniqueKey":"draft.id","typeLabel":"string\u003cuuid>"},"agent_id":{"type":"string","format":"uuid","uniqueKey":"draft.agent_id","typeLabel":"string\u003cuuid>"},"version_number":{"type":"integer","description":"0 is the mutable draft; every published version is 1 or above.","uniqueKey":"draft.version_number","typeLabel":"integer"},"rev":{"type":"integer","description":"Server-owned change counter, bumped on every save of the draft. Read-only — useful to tell whether a draft moved since you last read it.","uniqueKey":"draft.rev","typeLabel":"integer"},"instructions":{"type":"array","description":"The system prompt, as ordered blocks. Entries are `{type:\"text\"}`, `{type:\"prompt_block\"}` or `{type:\"prompt_block_ref\"}`. On a published version each ref also carries the `versionId` it was frozen to at publish time; on the draft that pin is absent.","items":{"type":"object","uniqueKey":"draft.instructions.items.[INDEX]","typeLabel":"object"},"uniqueKey":"draft.instructions","typeLabel":"object[]"},"model_slug":{"oneOf":[{"type":"string","description":"The concrete model this version runs. Ignored when `policy_slug` is set.","uniqueKey":"draft.model_slug.oneOf.0","typeLabel":"string"},{"type":"null","description":"The concrete model this version runs. Ignored when `policy_slug` is set.","uniqueKey":"draft.model_slug.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.model_slug","typeLabel":"string | null"},"policy_slug":{"oneOf":[{"type":"string","description":"Routing policy to run instead of a fixed model. Overrides `model_slug`.","uniqueKey":"draft.policy_slug.oneOf.0","typeLabel":"string"},{"type":"null","description":"Routing policy to run instead of a fixed model. Overrides `model_slug`.","uniqueKey":"draft.policy_slug.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.policy_slug","typeLabel":"string | null"},"model_options":{"oneOf":[{"type":"object","description":"Model-level options passed to the router (reasoning effort, native options).","uniqueKey":"draft.model_options.oneOf.0","typeLabel":"object"},{"type":"null","description":"Model-level options passed to the router (reasoning effort, native options).","uniqueKey":"draft.model_options.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.model_options","typeLabel":"object | null"},"model_settings":{"oneOf":[{"type":"object","description":"Generation settings: temperature, maxTokens, etc.","uniqueKey":"draft.model_settings.oneOf.0","typeLabel":"object"},{"type":"null","description":"Generation settings: temperature, maxTokens, etc.","uniqueKey":"draft.model_settings.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.model_settings","typeLabel":"object | null"},"max_steps":{"oneOf":[{"type":"integer","uniqueKey":"draft.max_steps.oneOf.0","typeLabel":"integer"},{"type":"null","uniqueKey":"draft.max_steps.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.max_steps","typeLabel":"integer | null"},"tool_search":{"oneOf":[{"type":"object","description":"Tool search for large rosters: `{ enabled?, threshold?, providers?, always_loaded?, top_k?, show_catalog? }`. Null = the platform default (on above 40 tools). `enabled: false` turns it off. See the draft save for the keys.","uniqueKey":"draft.tool_search.oneOf.0","typeLabel":"object"},{"type":"null","description":"Tool search for large rosters: `{ enabled?, threshold?, providers?, always_loaded?, top_k?, show_catalog? }`. Null = the platform default (on above 40 tools). `enabled: false` turns it off. See the draft save for the keys.","uniqueKey":"draft.tool_search.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.tool_search","typeLabel":"object | null"},"browser":{"oneOf":[{"type":"object","description":"The built-in browser: `{ enabled, viewport?: { width, height } }`. Null or `enabled: false` = the agent has no browser. See the draft save.","uniqueKey":"draft.browser.oneOf.0","typeLabel":"object"},{"type":"null","description":"The built-in browser: `{ enabled, viewport?: { width, height } }`. Null or `enabled: false` = the agent has no browser. See the draft save.","uniqueKey":"draft.browser.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.browser","typeLabel":"object | null"},"guardrails":{"oneOf":[{"type":"array","description":"Guardrails on this version: `[{ kind, mode?, enabled?, options? }]`. `kind` is one of `pii`, `moderation`, `prompt_injection`, `unicode`, `language`, `system_prompt_scrub`; `mode` is `warn` (default) or `block`. Null = none. See the draft save for the options.","items":{"type":"object","uniqueKey":"draft.guardrails.oneOf.0.items.[INDEX]","typeLabel":"object"},"uniqueKey":"draft.guardrails.oneOf.0","typeLabel":"object[]"},{"type":"null","description":"Guardrails on this version: `[{ kind, mode?, enabled?, options? }]`. `kind` is one of `pii`, `moderation`, `prompt_injection`, `unicode`, `language`, `system_prompt_scrub`; `mode` is `warn` (default) or `block`. Null = none. See the draft save for the options.","items":{"type":"object"},"uniqueKey":"draft.guardrails.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.guardrails","typeLabel":"object[] | null"},"agentTools":{"type":"array","description":"Tool references this version can call, each as stored (`tool_id`, `provider_id`, `config`, flags — see the draft save) plus two read-only names resolved on the way out: `tool_name` and `provider_name`. Either is null when the reference no longer resolves (the tool or provider was deleted). The names are stripped again when the entry is sent back on a save, so an entry can be round-tripped unchanged.","items":{"type":"object","properties":{"tool_id":{"type":"string","description":"The stored reference: a tool id, an MCP tool name, a platform tool key or a connected-app action.","uniqueKey":"draft.agentTools.items.[INDEX].tool_id","typeLabel":"string"},"provider_id":{"oneOf":[{"type":"string","description":"The provider the reference belongs to (an id, or a platform provider key).","uniqueKey":"draft.agentTools.items.[INDEX].provider_id.oneOf.0","typeLabel":"string"},{"type":"null","description":"The provider the reference belongs to (an id, or a platform provider key).","uniqueKey":"draft.agentTools.items.[INDEX].provider_id.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.agentTools.items.[INDEX].provider_id","typeLabel":"string | null"},"tool_name":{"oneOf":[{"type":"string","description":"The tool's current name. Read-only.","uniqueKey":"draft.agentTools.items.[INDEX].tool_name.oneOf.0","typeLabel":"string"},{"type":"null","description":"The tool's current name. Read-only.","uniqueKey":"draft.agentTools.items.[INDEX].tool_name.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.agentTools.items.[INDEX].tool_name","typeLabel":"string | null"},"provider_name":{"oneOf":[{"type":"string","description":"The provider's current name. Read-only.","uniqueKey":"draft.agentTools.items.[INDEX].provider_name.oneOf.0","typeLabel":"string"},{"type":"null","description":"The provider's current name. Read-only.","uniqueKey":"draft.agentTools.items.[INDEX].provider_name.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.agentTools.items.[INDEX].provider_name","typeLabel":"string | null"},"config":{"oneOf":[{"type":"object","uniqueKey":"draft.agentTools.items.[INDEX].config.oneOf.0","typeLabel":"object"},{"type":"null","uniqueKey":"draft.agentTools.items.[INDEX].config.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.agentTools.items.[INDEX].config","typeLabel":"object | null"},"require_approval":{"type":"boolean","uniqueKey":"draft.agentTools.items.[INDEX].require_approval","typeLabel":"boolean"},"loop_responses":{"type":"boolean","uniqueKey":"draft.agentTools.items.[INDEX].loop_responses","typeLabel":"boolean"}},"uniqueKey":"draft.agentTools.items.[INDEX]","typeLabel":"object"},"uniqueKey":"draft.agentTools","typeLabel":"object[]"},"workflows":{"type":"array","description":"Attached workflows.","items":{"type":"object","uniqueKey":"draft.workflows.items.[INDEX]","typeLabel":"object"},"uniqueKey":"draft.workflows","typeLabel":"object[]"},"skills":{"oneOf":[{"type":"array","description":"Organization skills this version references, as `[{ skill_id }]` (see `/v1/skills`). The agent loads a skill's active version on demand; null = none.","items":{"type":"object","uniqueKey":"draft.skills.oneOf.0.items.[INDEX]","typeLabel":"object"},"uniqueKey":"draft.skills.oneOf.0","typeLabel":"object[]"},{"type":"null","description":"Organization skills this version references, as `[{ skill_id }]` (see `/v1/skills`). The agent loads a skill's active version on demand; null = none.","items":{"type":"object"},"uniqueKey":"draft.skills.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.skills","typeLabel":"object[] | null"},"agents":{"type":"array","description":"Subagents this version can delegate to, as `[{ agentId }]`.","items":{"type":"object","uniqueKey":"draft.agents.items.[INDEX]","typeLabel":"object"},"uniqueKey":"draft.agents","typeLabel":"object[]"},"scorers":{"type":"array","description":"Scorers attached to this version: `{ name, scorerType, model?, sampling_rate?, options?, filter? }`. `filter` is an eligibility predicate evaluated before sampling (see the experiments guide).","items":{"type":"object","uniqueKey":"draft.scorers.items.[INDEX]","typeLabel":"object"},"uniqueKey":"draft.scorers","typeLabel":"object[]"},"memory":{"oneOf":[{"type":"object","description":"Memory settings of this version, as `{ options }` — the exact shape the draft save accepts (every option is listed there, and explained in the memory guide). Null = platform defaults.","example":{"options":{"lastMessages":20,"workingMemory":{"enabled":true,"scope":"resource"},"observationalMemory":{"enabled":true,"observation":{"manageWorkingMemory":true,"bufferOnIdle":true,"threadTitle":true}}}},"uniqueKey":"draft.memory.oneOf.0","typeLabel":"object"},{"type":"null","description":"Memory settings of this version, as `{ options }` — the exact shape the draft save accepts (every option is listed there, and explained in the memory guide). Null = platform defaults.","example":{"options":{"lastMessages":20,"workingMemory":{"enabled":true,"scope":"resource"},"observationalMemory":{"enabled":true,"observation":{"manageWorkingMemory":true,"bufferOnIdle":true,"threadTitle":true}}}},"uniqueKey":"draft.memory.oneOf.1","typeLabel":"null"}],"uniqueKey":"draft.memory","typeLabel":"object | null"},"created_at":{"type":"string","format":"date-time","uniqueKey":"draft.created_at","typeLabel":"string\u003cdate-time>"},"updated_at":{"type":"string","format":"date-time","uniqueKey":"draft.updated_at","typeLabel":"string\u003cdate-time>"}},"description":"The whole draft after the merge, not just the fields you sent. `rev` has been bumped by the DB.","uniqueKey":"draft","typeLabel":"object"}},"uniqueKey":"","typeLabel":"object"},"examples":{"Example":{"title":"Example","value":{"agent":{"id":"3c90c3cc-0d44-4b50-8888-8dd25736052a","org_id":"3c90c3cc-0d44-4b50-8888-8dd25736052a","name":"\u003cstring>","description":"\u003cstring>","active_version_id":"3c90c3cc-0d44-4b50-8888-8dd25736052a","created_at":"2023-11-07T05:31:56Z","updated_at":"2023-11-07T05:31:56Z","activeVersionNumber":123,"latestPublishedVersionNumber":123,"draftDirty":true},"draft":{"id":"3c90c3cc-0d44-4b50-8888-8dd25736052a","agent_id":"3c90c3cc-0d44-4b50-8888-8dd25736052a","version_number":123,"rev":123,"instructions":[{}],"model_slug":"\u003cstring>","policy_slug":"\u003cstring>","model_options":{},"model_settings":{},"max_steps":123,"tool_search":{},"browser":{},"guardrails":[{}],"agentTools":[{"tool_id":"\u003cstring>","provider_id":"\u003cstring>","tool_name":"\u003cstring>","provider_name":"\u003cstring>","config":{},"require_approval":true,"loop_responses":true}],"workflows":[{}],"skills":[{}],"agents":[{}],"scorers":[{}],"memory":{"options":{"lastMessages":20,"workingMemory":{"enabled":true,"scope":"resource"},"observationalMemory":{"enabled":true,"observation":{"manageWorkingMemory":true,"bufferOnIdle":true,"threadTitle":true}}}},"created_at":"2023-11-07T05:31:56Z","updated_at":"2023-11-07T05:31:56Z"}}}},"exampleType":"generated"}},"exampleType":"generated"},"913bfefa-1d42-4af0-908b-b78127fe9800":{"description":"Invalid body","content":{"_mintlify/placeholder":{"schema":{"type":"object","description":""}}}},"54dea7b5-ddc3-4d3f-8fd4-9f2ed66c5dbe":{"description":"Not found","content":{"_mintlify/placeholder":{"schema":{"type":"object","description":""}}}}},"security":{"b98d762b-7a65-4629-83ec-12fce5b08580":{"bearerAuth":{"type":"http","scheme":"bearer","description":"Organization API key, generated in studio.sidenet.ai. Backend only — never in a browser.","scopes":[],"schemeName":"bearerAuth"}}},"servers":{"5a9ef6cd-d896-47ca-98f0-fbb8d6f1aa96":{"url":"https://api.sidenet.ai"}},"processedSecurityOptions":[{"schemes":[{"scheme":{"type":"http","scheme":"bearer","description":"Organization API key, generated in studio.sidenet.ai. Backend only — never in a browser.","scopes":[],"schemeName":"bearerAuth"},"schema":{"title":"Bearer Auth","type":"string","isRequired":true,"description":"Organization API key, generated in studio.sidenet.ai. Backend only — never in a browser.","format":"bearer","typeLabel":"string\u003cbearer>","deprecated":false},"schemeName":"bearerAuth","name":"Authorization"}],"title":"bearerAuth"}]},"originalFileLocation":"api-reference/openapi.json"},"subdomain":"sidenetai","siteName":"Sidenet API","title":"Save agent draft - Sidenet API","navMaps":{"firstHrefInVersion":{},"firstHrefInLanguage":{}},"lastModified":"2026-11-15T17:51:29.199Z","originalFileLocation":"api-reference/agents/save-agent-draft.mdx"},"headHtml":"\u003cscript data-mint-head>(function(){var d=document.documentElement,p=null;if(!p){try{p=localStorage.getItem(\"isDarkMode\")}catch(e){}if(p==='true')p='dark';else if(p==='false')p='light';if(p!=='light'\u0026\u0026p!=='dark'\u0026\u0026p!=='system')p=\"system\"}var r=p==='system'?(matchMedia(\"(prefers-color-scheme: dark)\").matches?'dark':'light'):p;d.classList.remove(r==='dark'?'light':'dark');d.classList.add(r);d.setAttribute(\"data-theme-preference\",p)})();\u003c/script>\u003cmeta data-mint-head name=\"description\" content=\"Partial save of the draft agent config (instructions, model fields, agentTools, tool_search, browser, guardrails, workflows, memory, scorers, subagents, etc). Only the fields you send are written. Subagents are [{ agentId }]. The model fields are model_slug, policy_slug, model_options and model_settings; the older requesty_model_slug, requesty_policy_slug and requesty_options spellings are still accepted and mapped onto them (the canonical name wins if you send both). prompt, model_name, model_provider and model_openrouter_slug no longer exist and are ignored — the system prompt is instructions. name and description are the agent\u0026#x27;s own, not the draft\u0026#x27;s: they are saved here for convenience but apply immediately to every version, published ones included, and need no publish. The response carries the updated agent alongside the draft.\">\u003cmeta data-mint-head name=\"application-name\" content=\"Sidenet API\">\u003cmeta data-mint-head name=\"generator\" content=\"Mintlify\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"192x192\" href=\"/docs/mintlify-assets/_mintlify/favicons/sidenetai/jpMqiy2x92lioK32/_generated/favicon/android-chrome-192x192.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"apple-touch-icon\" type=\"image/png\" sizes=\"180x180\" href=\"/docs/mintlify-assets/_mintlify/favicons/sidenetai/jpMqiy2x92lioK32/_generated/favicon/apple-touch-icon.png\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"16x16\" href=\"/docs/mintlify-assets/_mintlify/favicons/sidenetai/jpMqiy2x92lioK32/_generated/favicon/favicon-16x16.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"32x32\" href=\"/docs/mintlify-assets/_mintlify/favicons/sidenetai/jpMqiy2x92lioK32/_generated/favicon/favicon-32x32.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"shortcut icon\" type=\"image/x-icon\" href=\"/docs/mintlify-assets/_mintlify/favicons/sidenetai/jpMqiy2x92lioK32/_generated/favicon/favicon.ico\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"192x192\" href=\"/docs/mintlify-assets/_mintlify/favicons/sidenetai/jpMqiy2x92lioK32/_generated/favicon-dark/android-chrome-192x192.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"16x16\" href=\"/docs/mintlify-assets/_mintlify/favicons/sidenetai/jpMqiy2x92lioK32/_generated/favicon-dark/favicon-16x16.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"32x32\" href=\"/docs/mintlify-assets/_mintlify/favicons/sidenetai/jpMqiy2x92lioK32/_generated/favicon-dark/favicon-32x32.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"shortcut icon\" type=\"image/x-icon\" href=\"/docs/mintlify-assets/_mintlify/favicons/sidenetai/jpMqiy2x92lioK32/_generated/favicon-dark/favicon.ico\" media=\"(prefers-color-scheme: dark)\">\u003cmeta data-mint-head property=\"og:title\" content=\"Save agent draft - Sidenet API\">\u003cmeta data-mint-head property=\"og:description\" content=\"Partial save of the draft agent config (instructions, model fields, agentTools, tool_search, browser, guardrails, workflows, memory, scorers, subagents, etc). Only the fields you send are written. Subagents are [{ agentId }]. The model fields are model_slug, policy_slug, model_options and model_settings; the older requesty_model_slug, requesty_policy_slug and requesty_options spellings are still accepted and mapped onto them (the canonical name wins if you send both). prompt, model_name, model_provider and model_openrouter_slug no longer exist and are ignored — the system prompt is instructions. name and description are the agent\u0026#x27;s own, not the draft\u0026#x27;s: they are saved here for convenience but apply immediately to every version, published ones included, and need no publish. The response carries the updated agent alongside the draft.\">\u003cmeta data-mint-head property=\"og:site_name\" content=\"Sidenet API\">\u003cmeta data-mint-head property=\"og:url\" content=\"https://www.sidenet.ai/api-reference/agents/save-agent-draft\">\u003cmeta data-mint-head property=\"og:type\" content=\"website\">\u003cmeta data-mint-head property=\"og:image\" content=\"https://sidenetai.mintlify.site/docs/_mintlify/api/og/v2?division=Agents\u0026amp;title=Save+agent+draft\u0026amp;description=Partial+save+of+the+draft+agent+config+%28instructions%2C+model+fields%2C+agentTools%2C+tool_search%2C+browser%2C+guardrails%2C+workflows%2C+memory%2C+scorers%2C+subagents%2C+etc%29.+O\u0026amp;theme=a1e7f929b234304c081618fd\u0026amp;signature=fa1a6a0fb2dd70b3fbd5e087f9875562c96e972290b73e416bbd551084cde717\">\u003cmeta data-mint-head property=\"og:image:width\" content=\"1200\">\u003cmeta data-mint-head property=\"og:image:height\" content=\"630\">\u003cmeta data-mint-head name=\"twitter:card\" content=\"summary_large_image\">\u003cmeta data-mint-head name=\"twitter:title\" content=\"Save agent draft - Sidenet API\">\u003cmeta data-mint-head name=\"twitter:description\" content=\"Partial save of the draft agent config (instructions, model fields, agentTools, tool_search, browser, guardrails, workflows, memory, scorers, subagents, etc). Only the fields you send are written. Subagents are [{ agentId }]. The model fields are model_slug, policy_slug, model_options and model_settings; the older requesty_model_slug, requesty_policy_slug and requesty_options spellings are still accepted and mapped onto them (the canonical name wins if you send both). prompt, model_name, model_provider and model_openrouter_slug no longer exist and are ignored — the system prompt is instructions. name and description are the agent\u0026#x27;s own, not the draft\u0026#x27;s: they are saved here for convenience but apply immediately to every version, published ones included, and need no publish. The response carries the updated agent alongside the draft.\">\u003cmeta data-mint-head name=\"twitter:image\" content=\"https://sidenetai.mintlify.site/docs/_mintlify/api/og/v2?division=Agents\u0026amp;title=Save+agent+draft\u0026amp;description=Partial+save+of+the+draft+agent+config+%28instructions%2C+model+fields%2C+agentTools%2C+tool_search%2C+browser%2C+guardrails%2C+workflows%2C+memory%2C+scorers%2C+subagents%2C+etc%29.+O\u0026amp;theme=a1e7f929b234304c081618fd\u0026amp;signature=fa1a6a0fb2dd70b3fbd5e087f9875562c96e972290b73e416bbd551084cde717\">\u003cmeta data-mint-head name=\"twitter:image:width\" content=\"1200\">\u003cmeta data-mint-head name=\"twitter:image:height\" content=\"630\">\u003cmeta data-mint-head name=\"msapplication-config\" content=\"/docs/mintlify-assets/_mintlify/favicons/sidenetai/jpMqiy2x92lioK32/_generated/favicon/browserconfig.xml\">\u003cmeta data-mint-head name=\"apple-mobile-web-app-title\" content=\"Sidenet API\">\u003cmeta data-mint-head name=\"msapplication-TileColor\" content=\"#4F46E5\">\u003clink data-mint-head rel=\"canonical\" href=\"https://www.sidenet.ai/api-reference/agents/save-agent-draft\">\u003clink data-mint-head rel=\"alternate\" type=\"application/xml\" href=\"/docs/sitemap.xml\">\u003clink data-mint-head rel=\"alternate\" type=\"text/markdown\" href=\"/docs/api-reference/agents/save-agent-draft.md\">\u003cscript data-mint-head type=\"application/ld+json\">{\"@context\":\"https://schema.org\",\"@graph\":[{\"@type\":\"Organization\",\"@id\":\"https://www.sidenet.ai/#organization\",\"name\":\"Sidenet API\",\"url\":\"https://www.sidenet.ai\",\"logo\":{\"@type\":\"ImageObject\",\"url\":\"https://mintcdn.com/sidenetai/IsfHUeiSCKNFr_6U/images/logo-light.svg?fit=max\u0026auto=format\u0026n=IsfHUeiSCKNFr_6U\u0026q=85\u0026s=206c6acd674e00cc951ff99d54600daa\"}},{\"@type\":\"WebSite\",\"@id\":\"https://www.sidenet.ai#website\",\"name\":\"Sidenet API\",\"url\":\"https://www.sidenet.ai\",\"publisher\":{\"@id\":\"https://www.sidenet.ai/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https://www.sidenet.ai/api-reference/agents/save-agent-draft#webpage\",\"url\":\"https://www.sidenet.ai/api-reference/agents/save-agent-draft\",\"name\":\"Save agent draft\",\"description\":\"Partial save of the draft agent config (instructions, model fields, agentTools, tool_search, browser, guardrails, workflows, memory, scorers, subagents, etc). Only the fields you send are written. Subagents are [{ agentId }]. The model fields are model_slug, policy_slug, model_options and model_settings; the older requesty_model_slug, requesty_policy_slug and requesty_options spellings are still accepted and mapped onto them (the canonical name wins if you send both). prompt, model_name, model_provider and model_openrouter_slug no longer exist and are ignored — the system prompt is instructions. name and description are the agent's own, not the draft's: they are saved here for convenience but apply immediately to every version, published ones included, and need no publish. The response carries the updated agent alongside the draft.\",\"dateModified\":\"2026-11-15T17:51:29.199Z\",\"isPartOf\":{\"@id\":\"https://www.sidenet.ai#website\"},\"breadcrumb\":{\"@id\":\"https://www.sidenet.ai/api-reference/agents/save-agent-draft#breadcrumb\"}},{\"@type\":\"BreadcrumbList\",\"@id\":\"https://www.sidenet.ai/api-reference/agents/save-agent-draft#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Agents\",\"item\":\"https://www.sidenet.ai/api-reference/agents/list-agents\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Save agent draft\",\"item\":\"https://www.sidenet.ai/api-reference/agents/save-agent-draft\"}]},{\"@type\":[\"TechArticle\",\"APIReference\"],\"@id\":\"https://www.sidenet.ai/api-reference/agents/save-agent-draft#apireference\",\"headline\":\"Save agent draft\",\"name\":\"Save agent draft\",\"description\":\"Partial save of the draft agent config (instructions, model fields, agentTools, tool_search, browser, guardrails, workflows, memory, scorers, subagents, etc). Only the fields you send are written. Subagents are [{ agentId }]. The model fields are model_slug, policy_slug, model_options and model_settings; the older requesty_model_slug, requesty_policy_slug and requesty_options spellings are still accepted and mapped onto them (the canonical name wins if you send both). prompt, model_name, model_provider and model_openrouter_slug no longer exist and are ignored — the system prompt is instructions. name and description are the agent's own, not the draft's: they are saved here for convenience but apply immediately to every version, published ones included, and need no publish. The response carries the updated agent alongside the draft.\",\"url\":\"https://www.sidenet.ai/api-reference/agents/save-agent-draft\",\"mainEntityOfPage\":{\"@id\":\"https://www.sidenet.ai/api-reference/agents/save-agent-draft#webpage\"},\"dateModified\":\"2026-11-15T17:51:29.199Z\",\"publisher\":{\"@id\":\"https://www.sidenet.ai/#organization\"},\"isPartOf\":{\"@id\":\"https://www.sidenet.ai#website\"}}]}\u003c/script>\u003cscript data-mint-head type=\"application/ld+json\">{\"@context\":\"https://schema.org\",\"@type\":\"WebSite\",\"name\":\"Sidenet API\",\"publisher\":{\"@type\":\"Organization\",\"name\":\"Sidenet API\"},\"provider\":{\"@type\":\"Organization\",\"name\":\"Mintlify\",\"url\":\"https://mintlify.com\"}}\u003c/script>","agentPointerHref":"/docs/api-reference/agents/save-agent-draft.md","siteName":"Sidenet API","nav":{"global":null,"tabs":[{"tab":"Guides","pages":[{"title":"Guides","href":"/concepts"}]},{"tab":"API Reference","groups":[{"group":"Overview","root":{"title":"Overview","href":"/api-reference/overview"},"pages":[]},{"group":"Authentication","pages":[{"title":"Mint session token","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/token","href":"/api-reference/authentication/mint-session-token"},{"title":"Refresh session token","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/token/refresh","href":"/api-reference/authentication/refresh-session-token"},{"title":"List runtime auth providers","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/tools-auth","href":"/api-reference/authentication/list-runtime-auth-providers"}]},{"group":"Users \u0026 Groups","pages":[{"title":"List users","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/users","href":"/api-reference/users-\u0026-groups/list-users"},{"title":"Create user","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/users","href":"/api-reference/users-\u0026-groups/create-user"},{"title":"Update user","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/users/{userId}","href":"/api-reference/users-\u0026-groups/update-user"},{"title":"List groups","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/groups","href":"/api-reference/users-\u0026-groups/list-groups"},{"title":"Create group","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/groups","href":"/api-reference/users-\u0026-groups/create-group"},{"title":"Update group","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/groups/{groupId}","href":"/api-reference/users-\u0026-groups/update-group"}]},{"group":"Organizations","pages":[{"title":"Create organization","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/orgs","href":"/api-reference/organizations/create-organization"},{"title":"List organizations","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/orgs","href":"/api-reference/organizations/list-organizations"},{"title":"Suspend organization","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/orgs/{id}/suspend","href":"/api-reference/organizations/suspend-organization"},{"title":"Resume organization","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/orgs/{id}/resume","href":"/api-reference/organizations/resume-organization"}]},{"group":"Billing","pages":[{"title":"Get credit balance","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/billing/balance","href":"/api-reference/billing/get-credit-balance"},{"title":"List credit transactions","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/billing/transactions","href":"/api-reference/billing/list-credit-transactions"},{"title":"List usage invoices","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/billing/invoices","href":"/api-reference/billing/list-usage-invoices"}]},{"group":"Chat \u0026 Messaging","pages":[{"title":"Stream chat","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/chat","href":"/api-reference/chat/stream-chat"},{"title":"Run agent","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/agents/{id}/run","href":"/api-reference/chat/run-agent"},{"title":"Run network","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/copilots/{id}/run","href":"/api-reference/chat/run-network"},{"title":"Resume stream","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/chat/stream","href":"/api-reference/chat/resume-stream"},{"title":"Cancel stream","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/chat/stream","href":"/api-reference/chat/cancel-stream"},{"title":"List threads","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/threads","href":"/api-reference/chat/list-threads"},{"title":"List messages","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/messages","href":"/api-reference/chat/list-messages"},{"title":"Vote on message","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/vote","href":"/api-reference/chat/vote-on-message"},{"title":"Mark thread read","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/threads/{id}/read","href":"/api-reference/chat/mark-thread-read"},{"title":"Append thread messages","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/threads/{id}/messages","href":"/api-reference/chat/append-thread-messages"},{"title":"Delete thread","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/threads/{id}","href":"/api-reference/chat/delete-thread"},{"title":"Transcribe audio","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/audio/transcriptions","href":"/api-reference/chat/transcribe-audio"}]},{"group":"Agents","pages":[{"title":"List agents","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/agents","href":"/api-reference/agents/list-agents"},{"title":"Create agent","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/agents","href":"/api-reference/agents/create-agent"},{"title":"Get agent","description":"Returns the agent and exactly ONE of its versions. Which one is up to `version`; by default it is the active (published) version, falling back to the draft when nothing has been activated yet. The `agent` object always reports `activeVersionNumber`, `latestPublishedVersionNumber` and `draftDirty`, so a client can tell what else exists — unpublished edits, or a newer published version than the one serving traffic — and ask for it by number in a second call. `referencedBlocks` and `variables` describe the version that was returned. `variables` is the {{placeholders}} it and its subagents consume, as a nested object mirroring the dot paths — `{{instance.store_count}}` comes back as `{\"instance\":{\"store_count\":null}}`. Each leaf is the default it renders without a value, or null when it has none (an unsupplied value stays visible in the prompt). Fill the nulls in and send the same object as a chat body's or an experiment's `variables`; leaves left null are ignored.","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/agents/{id}","href":"/api-reference/agents/get-agent"},{"title":"Save agent draft","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/agents/{id}/draft","href":"/api-reference/agents/save-agent-draft"},{"title":"Preview agent prompt","description":"Resolves the agent's prompt blocks and renders them with the supplied `variables`, returning the final prompt plus a per-block breakdown: whether each block was included (and why not), what it rendered to, and which of its `{{variables}}` had no value. `version` selects which instructions to render — \"draft\" (default), a published version number, or a version id. Unresolvable block references are an error here (422) rather than being skipped, so you see problems before publishing. Variables rejected by the size/type limits come back in `droppedVariables`.","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/agents/{id}/instructions/preview","href":"/api-reference/agents/preview-agent-prompt"},{"title":"Publish agent","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/agents/{id}/publish","href":"/api-reference/agents/publish-agent"},{"title":"Activate agent version","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/agents/{id}/activate","href":"/api-reference/agents/activate-agent-version"},{"title":"List agent versions","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/agents/{id}/versions","href":"/api-reference/agents/list-agent-versions"},{"title":"Delete agent","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/agents/{id}","href":"/api-reference/agents/delete-agent"}]},{"group":"Prompt Blocks","pages":[{"title":"List prompt blocks","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/prompt-blocks","href":"/api-reference/prompt-blocks/list-prompt-blocks"},{"title":"Create prompt block","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/prompt-blocks","href":"/api-reference/prompt-blocks/create-prompt-block"},{"title":"Get prompt block","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/prompt-blocks/{id}","href":"/api-reference/prompt-blocks/get-prompt-block"},{"title":"Save block draft","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/prompt-blocks/{id}/draft","href":"/api-reference/prompt-blocks/save-block-draft"},{"title":"Publish prompt block","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/prompt-blocks/{id}/publish","href":"/api-reference/prompt-blocks/publish-prompt-block"},{"title":"Activate block version","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/prompt-blocks/{id}/activate","href":"/api-reference/prompt-blocks/activate-block-version"},{"title":"List block versions","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/prompt-blocks/{id}/versions","href":"/api-reference/prompt-blocks/list-block-versions"},{"title":"List block references","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/prompt-blocks/{id}/references","href":"/api-reference/prompt-blocks/list-block-references"},{"title":"Roll out block version","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/prompt-blocks/{id}/rollout","href":"/api-reference/prompt-blocks/roll-out-block-version"},{"title":"Preview block content","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/prompt-blocks/preview","href":"/api-reference/prompt-blocks/preview-block-content"},{"title":"Delete prompt block","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/prompt-blocks/{id}","href":"/api-reference/prompt-blocks/delete-prompt-block"}]},{"group":"Skills","pages":[{"title":"List skills","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/skills","href":"/api-reference/skills/list-skills"},{"title":"Create skill","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/skills","href":"/api-reference/skills/create-skill"},{"title":"Get skill","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/skills/{id}","href":"/api-reference/skills/get-skill"},{"title":"Save skill","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/skills/{id}","href":"/api-reference/skills/save-skill"},{"title":"Publish skill","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/skills/{id}/publish","href":"/api-reference/skills/publish-skill"},{"title":"Activate skill version","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/skills/{id}/activate","href":"/api-reference/skills/activate-skill-version"},{"title":"List skill versions","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/skills/{id}/versions","href":"/api-reference/skills/list-skill-versions"},{"title":"List skill references","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/skills/{id}/references","href":"/api-reference/skills/list-skill-references"},{"title":"Delete skill","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/skills/{id}","href":"/api-reference/skills/delete-skill"}]},{"group":"Workflows","pages":[{"title":"List workflows","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/workflows","href":"/api-reference/workflows/list-workflows"},{"title":"Create workflow","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/workflows","href":"/api-reference/workflows/create-workflow"},{"title":"Get workflow","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/workflows/{id}","href":"/api-reference/workflows/get-workflow"},{"title":"Save workflow draft","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/workflows/{id}/draft","href":"/api-reference/workflows/save-workflow-draft"},{"title":"Publish workflow","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/workflows/{id}/publish","href":"/api-reference/workflows/publish-workflow"},{"title":"Activate workflow version","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/workflows/{id}/activate","href":"/api-reference/workflows/activate-workflow-version"},{"title":"List workflow versions","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/workflows/{id}/versions","href":"/api-reference/workflows/list-workflow-versions"},{"title":"Run workflow","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/workflows/{id}/run","href":"/api-reference/workflows/run-workflow"},{"title":"Get workflow schema","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/workflows/definition-schema","href":"/api-reference/workflows/get-workflow-schema"},{"title":"Delete workflow","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/workflows/{id}","href":"/api-reference/workflows/delete-workflow"}]},{"group":"Approvals","pages":[{"title":"List my approval requests","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/approval-requests","href":"/api-reference/approvals/list-my-approval-requests"},{"title":"Get approval request","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/approval-requests/{ref}","href":"/api-reference/approvals/get-approval-request"},{"title":"Decide approval","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/approval-requests/{ref}/decide","href":"/api-reference/approvals/decide-approval"}]},{"group":"Workflow Schedules","pages":[{"title":"List workflow schedules","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/workflow-schedules","href":"/api-reference/workflows/list-workflow-schedules"},{"title":"Create workflow schedule","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/workflow-schedules","href":"/api-reference/workflows/create-workflow-schedule"},{"title":"Preview schedule timing","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/workflow-schedules/preview","href":"/api-reference/workflows/preview-schedule-timing"},{"title":"Get workflow schedule","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/workflow-schedules/{id}","href":"/api-reference/workflows/get-workflow-schedule"},{"title":"Update workflow schedule","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/workflow-schedules/{id}","href":"/api-reference/workflows/update-workflow-schedule"},{"title":"Delete workflow schedule","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/workflow-schedules/{id}","href":"/api-reference/workflows/delete-workflow-schedule"},{"title":"Pause workflow schedule","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/workflow-schedules/{id}/pause","href":"/api-reference/workflows/pause-workflow-schedule"},{"title":"Resume workflow schedule","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/workflow-schedules/{id}/resume","href":"/api-reference/workflows/resume-workflow-schedule"},{"title":"Run workflow schedule now","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/workflow-schedules/{id}/run-now","href":"/api-reference/workflows/run-workflow-schedule-now"},{"title":"List schedule runs","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/workflow-schedules/{id}/runs","href":"/api-reference/workflows/list-schedule-runs"},{"title":"List workflow runs","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/workflow-runs","href":"/api-reference/workflows/list-workflow-runs"},{"title":"Get workflow run","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/workflow-runs/{id}","href":"/api-reference/workflows/get-workflow-run"}]},{"group":"Dashboards","pages":[{"title":"List dashboards","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/dashboards","href":"/api-reference/dashboards/list-dashboards"},{"title":"Create dashboard from spec","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/dashboards","href":"/api-reference/dashboards/create-dashboard-from-spec"},{"title":"List dashboard templates","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/dashboards/templates","href":"/api-reference/dashboards/list-dashboard-templates"},{"title":"Get dashboard","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/dashboards/{id}","href":"/api-reference/dashboards/get-dashboard"},{"title":"Refresh dashboard","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/dashboards/{id}/refresh","href":"/api-reference/dashboards/refresh-dashboard"},{"title":"Get dashboard refresh","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/dashboards/{id}/runs/{runId}","href":"/api-reference/dashboards/get-dashboard-refresh"},{"title":"Run dashboard view","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/dashboards/{id}/views/{viewId}/run","href":"/api-reference/dashboards/run-dashboard-view"},{"title":"List dashboard versions","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/dashboards/{id}/versions","href":"/api-reference/dashboards/list-dashboard-versions"},{"title":"Activate dashboard version","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/dashboards/{id}/activate","href":"/api-reference/dashboards/activate-dashboard-version"},{"title":"Rename dashboard","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/dashboards/{id}","href":"/api-reference/dashboards/rename-dashboard"},{"title":"Delete dashboard","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/dashboards/{id}","href":"/api-reference/dashboards/delete-dashboard"}]},{"group":"Dashboard templates","pages":[{"title":"List stored dashboard templates","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/dashboard-templates","href":"/api-reference/dashboard-templates/list-stored-dashboard-templates"},{"title":"Get dashboard template","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/dashboard-templates/{key}","href":"/api-reference/dashboard-templates/get-dashboard-template"},{"title":"Save dashboard template","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json put /v1/dashboard-templates/{key}","href":"/api-reference/dashboard-templates/save-dashboard-template"},{"title":"Update dashboard template","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/dashboard-templates/{key}","href":"/api-reference/dashboard-templates/update-dashboard-template"},{"title":"Delete dashboard template","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/dashboard-templates/{key}","href":"/api-reference/dashboard-templates/delete-dashboard-template"},{"title":"Publish dashboard templates","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/dashboard-templates/publish","href":"/api-reference/dashboard-templates/publish-dashboard-templates"},{"title":"Get dashboard template publish state","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/dashboard-templates/state","href":"/api-reference/dashboard-templates/get-dashboard-template-publish-state"}]},{"group":"Browser","pages":[{"title":"Open a browser session","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/browser-sessions","href":"/api-reference/browser/open-a-browser-session"},{"title":"List browser sessions","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/browser-sessions","href":"/api-reference/browser/list-browser-sessions"},{"title":"Get a browser session","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/browser-sessions/{id}","href":"/api-reference/browser/get-a-browser-session"},{"title":"Get the live view of a browser session","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/browser-sessions/{id}/live-view","href":"/api-reference/browser/get-the-live-view-of-a-browser-session"},{"title":"Get a still of a browser session","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/browser-sessions/{id}/still","href":"/api-reference/browser/get-a-still-of-a-browser-session"},{"title":"Get the replay of a browser session","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/browser-sessions/{id}/replay","href":"/api-reference/browser/get-the-replay-of-a-browser-session"},{"title":"Get one page of a replay","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/browser-sessions/{id}/replay/{page}","href":"/api-reference/browser/get-one-page-of-a-replay"},{"title":"Open a tab","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/browser-sessions/{id}/tabs","href":"/api-reference/browser/open-a-tab"},{"title":"Close a tab","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/browser-sessions/{id}/tabs/{index}","href":"/api-reference/browser/close-a-tab"},{"title":"Take over a browser session","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/browser-sessions/{id}/take-over","href":"/api-reference/browser/take-over-a-browser-session"},{"title":"Hand a browser session back to the agent","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/browser-sessions/{id}/hand-back","href":"/api-reference/browser/hand-a-browser-session-back-to-the-agent"},{"title":"Pause the agent in a browser session","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/browser-sessions/{id}/pause","href":"/api-reference/browser/pause-the-agent-in-a-browser-session"},{"title":"Resume the agent in a browser session","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/browser-sessions/{id}/resume","href":"/api-reference/browser/resume-the-agent-in-a-browser-session"},{"title":"Stop a browser session","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/browser-sessions/{id}/stop","href":"/api-reference/browser/stop-a-browser-session"},{"title":"Hand a browser session to an agent","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/browser-sessions/{id}/attach","href":"/api-reference/browser/hand-a-browser-session-to-an-agent"},{"title":"Keep the sign-in of a browser session","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/browser-sessions/{id}/keep-sign-in","href":"/api-reference/browser/keep-the-sign-in-of-a-browser-session"},{"title":"Delete a saved login","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/browser-saved-logins/{ref}","href":"/api-reference/browser/delete-a-saved-login"}]},{"group":"Networks","pages":[{"title":"List networks","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/networks","href":"/api-reference/networks/list-networks"},{"title":"Create network","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/networks","href":"/api-reference/networks/create-network"},{"title":"Get network copilots","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/networks/{id}","href":"/api-reference/networks/get-network-copilots"},{"title":"Rename network","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/networks/{id}","href":"/api-reference/networks/rename-network"},{"title":"Get network","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/copilots/{id}","href":"/api-reference/networks/get-network"},{"title":"Update network","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/copilots/{id}","href":"/api-reference/networks/update-network"},{"title":"Attach agent","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/copilots/{id}/agents","href":"/api-reference/networks/attach-agent"},{"title":"Detach agent","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/copilots/{id}/agents/{agentId}","href":"/api-reference/networks/detach-agent"},{"title":"Attach workflow","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/copilots/{id}/workflows","href":"/api-reference/networks/attach-workflow"},{"title":"Detach workflow","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/copilots/{id}/workflows/{workflowId}","href":"/api-reference/networks/detach-workflow"},{"title":"Deploy network","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/copilots/{id}/deploy","href":"/api-reference/networks/deploy-network"}]},{"group":"Tools","pages":[{"title":"List tools","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/tools","href":"/api-reference/tools/list-tools"},{"title":"Execute tool","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/tools/{toolId}/execute","href":"/api-reference/tools/execute-tool"},{"title":"Create tool","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/tools","href":"/api-reference/tools/create-tool"},{"title":"Update tool","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/tools/{id}","href":"/api-reference/tools/update-tool"},{"title":"Delete tool","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/tools/{id}","href":"/api-reference/tools/delete-tool"}]},{"group":"Tool Providers","pages":[{"title":"List tool providers","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/tool-providers","href":"/api-reference/tool-providers/list-tool-providers"},{"title":"Create tool provider","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/tool-providers","href":"/api-reference/tool-providers/create-tool-provider"},{"title":"Get tool provider","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/tool-providers/{id}","href":"/api-reference/tool-providers/get-tool-provider"},{"title":"Update tool provider","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/tool-providers/{id}","href":"/api-reference/tool-providers/update-tool-provider"},{"title":"Save provider tools","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/tool-providers/{id}/tools","href":"/api-reference/tool-providers/save-provider-tools"},{"title":"Delete tool provider","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/tool-providers/{id}","href":"/api-reference/tool-providers/delete-tool-provider"},{"title":"Grant tool provider","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/tool-providers/{id}/grants","href":"/api-reference/tool-providers/grant-tool-provider"},{"title":"Revoke tool provider grant","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/tool-providers/{id}/grants/{org_id}","href":"/api-reference/tool-providers/revoke-tool-provider-grant"}]},{"group":"Datasets","pages":[{"title":"List datasets","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/datasets","href":"/api-reference/datasets/list-datasets"},{"title":"Create dataset","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/datasets","href":"/api-reference/datasets/create-dataset"},{"title":"Get dataset","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/datasets/{id}","href":"/api-reference/datasets/get-dataset"},{"title":"Update dataset","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/datasets/{id}","href":"/api-reference/datasets/update-dataset"},{"title":"Delete dataset","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/datasets/{id}","href":"/api-reference/datasets/delete-dataset"},{"title":"List dataset items","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/datasets/{id}/items","href":"/api-reference/datasets/list-dataset-items"},{"title":"Add dataset items","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/datasets/{id}/items","href":"/api-reference/datasets/add-dataset-items"},{"title":"Update dataset item","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json patch /v1/datasets/{id}/items/{itemId}","href":"/api-reference/datasets/update-dataset-item"},{"title":"Delete dataset item","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/datasets/{id}/items/{itemId}","href":"/api-reference/datasets/delete-dataset-item"}]},{"group":"Scoring \u0026 Experiments","pages":[{"title":"Run experiment","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/experiments","href":"/api-reference/experiments/run-experiment"},{"title":"Get experiment","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/experiments/{experimentId}","href":"/api-reference/experiments/get-experiment"},{"title":"Delete experiment","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json delete /v1/experiments/{experimentId}","href":"/api-reference/experiments/delete-experiment"},{"title":"List experiments","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/datasets/{id}/experiments","href":"/api-reference/experiments/list-experiments"},{"title":"Compare experiments","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/datasets/{id}/compare","href":"/api-reference/experiments/compare-experiments"},{"title":"Score traces","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/score-trace","href":"/api-reference/experiments/score-traces"},{"title":"List trace scores","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/traces/{traceId}/scores","href":"/api-reference/experiments/list-trace-scores"}]},{"group":"Observability","pages":[{"title":"List requests","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/requests","href":"/api-reference/observability/list-requests"},{"title":"Get a request","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/requests/{id}","href":"/api-reference/observability/get-a-request"},{"title":"Request statistics","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/requests/stats","href":"/api-reference/observability/request-statistics"},{"title":"Request filter options","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json get /v1/requests/filters","href":"/api-reference/observability/request-filter-options"}]},{"group":"System","pages":[{"title":"Rebuild agents","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /v1/rebuild","href":"/api-reference/system/rebuild-agents"},{"title":"Rebuild org tools","deprecated":null,"version":null,"openapi":"/api-reference/openapi.json post /rebuild-org-tools","href":"/api-reference/system/rebuild-org-tools"}]}]},{"tab":"Chat SDK","pages":[{"title":"Chat SDK","href":"/sdk/overview"}]}]},"basePath":"","title":"Save agent draft - Sidenet API","buildId":"dpl_Aw98ji4XvBsgtZwL1VrkUF9ehFqz","deploymentId":"dpl_Aw98ji4XvBsgtZwL1VrkUF9ehFqz","dynamics":["c2aa737661","2b790eda92","b0f84f53ff","a9bb46feb3"],"chunks":["OpenApiReferenceProvider-CQGbdtm4.js","utils-D88A1f-Z.js","cloneDeep-izs3Hzxb.js","_Uint8Array-BVIqJ90l.js","isObjectLike-C8vmzPEN.js","isArray-Dcu1dQ5t.js","utils-ChqWdz1A.js","toString-BX-iyh8G.js","isSymbol-CSH9U2Gm.js","addUniqueKeysToSchema-yBrxdh_6.js","snippetPresets-BgqDrFJC.js","style-to-object-BY3JEAdk.js","v4-DIgz8zy6.js","conditional-requirements-CFM0id0Q.js","ApiExamples-Duxc2rZw.js","isEqual-BfaiBbcl.js","usePlaygroundInputsStore-C7jidOil.js","shallow-xkjpHm-U.js","getApiPlaygroundDisplay-fE3rVpvL.js","CodeSnippet-LgxljKed.js","CodeSnippets-D-pXWDHq.js","Page-C2-BijcO.js","getPastedJson-CLRcu-uo.js","icon-Cxirpjgo.js","MethodPill-D49eqY5e.js","DuplicateIcon-CivCz5Rp.js","createIcon-D71qL3f5.js","DynamicMarkdown-BcGRTrw6.js","SectionHeading-DzaqmmSD.js","axios-vUo_bRtp.js","Markdown-CZqUYmMp.js","clsx-B6IZIRsp.js","lib-DuNMoZ05.js","lib-B8HbPgcC.js","html-void-elements-mCSPHezU.js","dist-B0ZV6X5T.js","katex-YMM-FMeD.js","marked.esm-CZWythHk.js","AssistantPageLink-MqLyPtsW.js"]}