--- title: "Edit draft config (prompt, tools, post-call metrics, voice, etc.)" method: PATCH path: "/agent/{id}/drafts/{draftId}/config" tags: ["Agent Versioning - Drafts", "Post-Call Analytics"] --- # Edit draft config (prompt, tools, post-call metrics, voice, etc.) `PATCH /agent/{id}/drafts/{draftId}/config` Update the configuration of a draft. This single endpoint is how every agent-level config field is changed: prompt, tools, voice, language, **post-call analytics (disposition metrics)**, and more. There is no standalone post-call-analytics endpoint — it lives here as the `postCallAnalyticsConfig` body field. ## Post-Call Analytics Pass a `postCallAnalyticsConfig` object to configure disposition metrics (STRING, BOOLEAN, INTEGER, ENUM, DATETIME) that are automatically extracted from each completed call, along with the `useInternalAnalyticsModel` and `useReasoningModel` flags. See the [Post-Call Metrics guide](/atoms/atoms-platform/features/post-call-metrics) for a full Python walkthrough and disposition metric schema reference. ## Full payload Accepts the full agent-shaped config payload (language, synthesizer, slmModel, defaultVariables, preCallAPI, etc.) plus two draft-specific fields: - `singlePromptConfig` — prompt and tools (end_call, transfer_call, api_call, extract_dynamic_variables, knowledge_base_search). - `postCallAnalyticsConfig` — disposition metrics + analytics/ reasoning model flags. Each PATCH increments the draft's revision counter. Config is not live until the draft is published and activated (see `/drafts/{draftId}/publish` and `/versions/{versionId}/activate`). ## Path parameters - `id` string, required - `draftId` string, required ## Request body - DraftConfigRequest — Config payload for editing a draft via `PATCH /agent/{id}/drafts/{draftId}/config`. Accepts the same shape as the legacy `PATCH /agent/{id}` body plus two versioning-era fields — `singlePromptConfig` (prompt + tools) and `postCallAnalyticsConfig`. All fields are optional; a draft save may update any subset. - `name` string, required - `description` string - `backgroundSound` '' | 'office' | 'cafe' | 'call_center' | 'static' — Ambient background sound during calls. Options: '' (none), 'office', 'cafe', 'call_center', 'static'. Note: this value is currently overridden by the server default on creation; update via PATCH after creation. - `language` object — Language configuration for the agent. Cross-field rule: `default` must be one of the values in `supported`. Tamil (`ta`) cannot be combined with other languages in `supported`. - `default` 'en' | 'hi' | 'mr' | 'gu' | 'ta' | 'es' | 'north_indic' | 'bn' | 'or' — The default language of the agent. Note: `ta` cannot be combined with other languages in `supported`. - `supported` string[] — Languages the agent understands. `default` must be one of these values. Tamil (`ta`) cannot be combined with other languages. - `switching` object — Language switching configuration for the agent. If enabled, the agent will be able to switch between languages based on the user's language. - `isEnabled` boolean — Whether to enable language switching for the agent - `minWordsForDetection` number — Minimum number of words required for language detection - `strongSignalThreshold` number — Threshold for strong language signal detection (0.1 to 0.9) - `weakSignalThreshold` number — Threshold for weak language signal detection (0.1 to 0.9) - `minConsecutiveForWeakThresholdSwitch` number — Minimum consecutive detections required for weak threshold language switch - `synthesizer` object — Synthesizer (TTS) configuration for the agent. Models `waves`, `waves_lightning_large`, `waves_lightning_v2`, and `waves_lightning_v3_1` validate `voiceId` against the Waves API. All other models accept any voiceId. Cloned voices are regular voiceIds — use them with any compatible Waves model. - `voiceConfig` object — Voice configuration for the synthesizer. - `model` 'waves' | 'waves_lightning_large' | 'waves_lightning_v2' | 'waves_lightning_v3_1' | 'waves_lightning_v3' | 'waves_lightning_v2_http' | 'gpt-realtime' | 'gpt-realtime-mini' — The TTS model to use. - `voiceId` string — The voice ID to use. For cloned voices, pass the voiceId from the Waves platform with a compatible model. - `gender` 'male' | 'female' — The gender of the voice. - `speed` number - `consistency` number - `similarity` number - `enhancement` 0 | 1 | 2 - `sampleRate` 8000 | 16000 | 24000 | 44100 — Output audio sample rate in Hz. - `globalKnowledgeBaseId` string — The global knowledge base ID of the agent. You can create a global knowledge base by using the /knowledgebase endpoint and assign it to the agent. The agent will use this knowledge base for its responses. - `slmModel` 'electron' | 'electron-kogta' | 'electron-kogta-v2' | 'gpt-4o' | 'gpt-4.1' | 'gpt-5.2' | 'gpt-realtime' | 'gpt-realtime-mini' — The LLM model to use for the agent. Note: `gpt-5.2`, `electron-kogta`, and `electron-kogta-v2` require org-level access and return 403 if not enabled. `workflowType` must be `single_prompt` to use `gpt-realtime` or `gpt-realtime-mini`. - `defaultVariables` object — The default variables to use for the agent. These variables will be used if no variables are provided when initiating a conversation with the agent. - `preCallAPI` object — Configuration for an API call to be made before the call starts. The response variables can be injected into the agent's prompt. - `isEnabled` boolean — Whether the pre-call API is enabled. - `url` string, uri, required — The URL of the API endpoint to call. - `method` 'GET' | 'POST' | 'PUT' | 'DELETE' | 'PATCH', required — The HTTP method to use for the API call. - `headers` object — Optional HTTP headers to include in the request. - `body` object — Optional request body for POST/PUT/PATCH requests. - `timeout` integer — Timeout in seconds for the API call. - `queryParams` object — Optional query parameters to include in the request URL. - `responseVariables` object[] — List of variables to extract from the API response using JSON path expressions. - `variableName` string, required — The name of the variable to inject into the agent prompt. - `jsonPath` string, required — JSON path expression to extract the value from the API response. - `globalPrompt` string — Set global instructions for your agent's personality, role, and behavior throughout conversations. Note: Only used for workflow_graph agents. Maximum 4000 characters. - `telephonyProductId` string[] — IDs of telephony products (phone numbers) to associate with the agent for inbound/outbound calls. - `workflowType` 'workflow_graph' | 'single_prompt' — The type of workflow configuration. workflow_graph uses a node-based visual workflow, single_prompt uses a simple prompt-based configuration. - `firstMessage` string — The first message the agent sends when a conversation starts. - `muteUserUntilFirstBotResponse` boolean — When true, the user's audio is muted until the agent has finished its first response. - `allowInterruptions` boolean — Whether the user can interrupt the agent while it is speaking. - `waitForUserToSpeakFirst` boolean — When true, the agent waits for the user to speak before sending the first message. - `interruptionBackoffTimer` number — Seconds the agent waits after being interrupted before resuming speech. - `smartTurnConfig` object — Smart turn-detection configuration. When enabled, the agent uses an additional model to decide whether the user has finished a turn. - `isEnabled` boolean - `waitTimeInSecs` number — How long to wait after the user stops speaking before responding. - `voiceDetectionConfig` object — Voice activity detection (VAD) configuration. Controls how the agent decides when speech is present. - `confidence` number — Minimum VAD confidence threshold to register speech. - `minVolume` number — Minimum input volume threshold to register speech. - `triggerTimeInSecs` number — How long sustained speech must be detected before turning the VAD on. - `releaseTimeInSecs` number — How long after silence before the VAD turns off. - `voiceMailDetectionConfig` object — Voicemail-detection configuration. When the call hits a voicemail tone, the agent plays `endText` and ends the call. - `enabled` boolean - `endText` string — Message played before hanging up when voicemail is detected. - `denoisingConfig` object — Background-noise denoising configuration for the agent's input audio. - `isEnabled` boolean - `redactionConfig` object — PII redaction configuration. When enabled, personally identifiable information is redacted from transcripts before storage. - `isEnabled` boolean - `pronunciationDicts` object[] — Pronunciation overrides — words the TTS engine should pronounce differently from its default. - `word` string, required — The word to override. - `pronunciation` string, required — How the word should be pronounced (phonetic spelling). - `llmIdleTimeoutConfig` object — Timeout configuration for the LLM stage of a conversation. Triggers a retry or call termination when the LLM does not respond within the configured window. - `chatTimeoutTimeInSecs` number — LLM idle timeout for chat conversations, in seconds. - `webcallTimeoutTimeInSecs` number — LLM idle timeout for web calls, in seconds. - `telephonyTimeoutTimeInSecs` number — LLM idle timeout for telephony calls, in seconds. - `maxRetries` number — Maximum number of LLM-idle retries before terminating the call. System-defined min/max. - `sessionTimeoutConfig` object — Maximum duration of a conversation session. The call ends after this elapsed time even if active. - `timeoutTimeInSecs` number — Maximum session duration in seconds (max 1 hour). Defaults to 1800 (30 minutes). - `timezone` object — Timezone applied to scheduled actions and timestamps the agent reports to the user. - `label` string — IANA timezone label (e.g. `America/New_York`). - `offset` number — UTC offset in minutes (e.g. -300 for EST). - `callDispositionConfig` string — Configuration string for call disposition tracking. - `allowInboundCall` boolean — Whether the agent accepts inbound calls. - `enableStyleGuide` boolean — Whether style guide enforcement is applied to agent responses. - `speechFormatting` boolean — Whether speech formatting is applied to the agent's responses. - `singlePromptConfig` SinglePromptConfig — Configuration for single prompt workflow type - `prompt` string, required — The main prompt that defines the agent's behavior and responses - `tools` Tool[] — Array of tools/functions available to the agent during conversations. Five tool types are supported: `end_call`, `transfer_call`, `api_call`, `extract_dynamic_variables`, and `knowledge_base_search`. Each type has its own required fields — see `Tool` schema. - `type` 'end_call' | 'transfer_call' | 'api_call' | 'extract_dynamic_variables' | 'knowledge_base_search', required — The type of function/tool - `name` string, required — Unique name for the function (no spaces) - `description` string, required — Description of what the function does - `enabled` boolean — Whether the tool is enabled - `transferNumber` string — Required for transfer_call type. Phone number to transfer the call to (E.164 format) - `transferOption` object — Required for transfer_call type. Controls cold vs warm transfer behavior. - `type` 'cold_transfer' | 'warm_transfer' — Transfer mode. `cold_transfer` hands off immediately; `warm_transfer` briefs the receiving party first. - `privateHandoffOption` object, nullable — Private briefing delivered to the transfer target before the caller is connected. Only used when `type = warm_transfer`. - `type` 'prompt' | 'static' — `prompt` generates briefing from the LLM; `static` plays fixed text. - `prompt` string — The prompt or static text for the private handoff. - `publicHandoffOption` object, nullable — Message played to the caller while the transfer is being set up. Only used when `type = warm_transfer`. - `type` 'prompt' | 'static' - `prompt` string - `onHoldMusic` 'ringtone' | 'relaxing_sound' | 'uplifting_beats' | 'none' — Optional for transfer_call type. Audio played to the caller while the transfer is in progress. - `transferOnlyIfHuman` boolean — Optional for transfer_call type. If true, the call is only transferred when a human is detected on the receiving end (voicemail/IVR skipped). - `detectionTimeout` integer — Optional for transfer_call type. Seconds to wait for human detection before giving up (5–60). - `url` string, uri — Required for api_call type. The URL to make the HTTP request to. - `method` 'GET' | 'POST' | 'PUT' | 'DELETE' | 'PATCH' — Required for api_call type. HTTP method to use. - `timeout` integer — Optional for api_call type. Request timeout in milliseconds (1000–30000). - `headers` object — Optional for api_call type. Static HTTP headers as a key/value map. - `headersArray` object[] — Optional for api_call type. Headers as an array of key/value objects (alternative to `headers` map). - `key` string, required - `value` string, required - `queryParams` object[] — Optional for api_call type. Query parameters to include in the request URL. Values support variable templating like `{{order_id}}`. - `key` string, required - `value` string, required - `requestBody` string — Optional for api_call type. Raw request body as a JSON string. Supports variable templating. - `llmParameters` object[] — Optional for api_call type. Parameters the LLM can supply dynamically at runtime. - `name` string, required — Parameter name - `description` string, required — What the parameter represents - `type` 'text' | 'number' | 'boolean' | 'enum', required - `values` string[] — Required when type is `enum`. Allowed values. - `required` boolean - `responseVariables` object[] — Optional for api_call type. Variables to extract from the API response into the agent's variable store. - `variableName` string, required — Name to store the extracted value under - `jsonPath` string, required — JSON path to extract the value from the response - `variablesExtractionSchema` object[] — Required for extract_dynamic_variables type. Schema defining variables to extract from the conversation. - `name` string, required — Name of the variable to extract - `description` string, required — What this variable represents - `type` 'text' | 'number' | 'boolean' | 'enum', required - `values` string[] — Required when type is `enum`. List of possible values. - `knowledgeBaseId` string — Required for knowledge_base_search type. ID of the knowledge base to search. - `fillerPhrases` string[] — Optional for knowledge_base_search type. Phrases spoken while searching. - `postCallAnalyticsConfig` PostCallAnalyticsConfig — Per-agent post-call analytics configuration. Evaluated after each call ends and surfaced in call logs under the `postCallAnalytics` field. - `dispositionMetrics` DispositionMetric[] — Structured metrics extracted from each completed call. - `identifier` string, required — Stable machine identifier. Lowercase letters, digits, and underscores only. - `dispositionMetricPrompt` string, required — Natural-language question evaluated against the transcript after the call ends. - `dispositionMetricType` 'STRING' | 'BOOLEAN' | 'INTEGER' | 'ENUM' | 'DATETIME', required — Data type returned by the metric. - `choices` string[] — Required when `dispositionMetricType = ENUM`. Allowed values. - `successMetrics` object[] — **Deprecated** — will be removed in a future version. Use `dispositionMetrics` instead. Kept here because the backend still accepts it on writes and returns it on reads. - `identifier` string, required - `successMetricPrompt` string, required - `successMetricType` 'NUMERIC_SCALE' | 'PERCENTAGE_SCALE' | 'PASS_FAIL' | 'DESCRIPTIVE_SCALE', required - `summaryPrompt` string — **Deprecated** — no longer used in post-call analysis and will be removed in a future version. Kept here because the backend still accepts it on writes and returns it on reads. - `useInternalAnalyticsModel` boolean — Use the internal analytics model. When false, falls back to the agent's own LLM. - `useReasoningModel` boolean — Route analytics evaluation through the reasoning model for higher-quality results at a latency/cost tradeoff. ## Response `200` Draft config updated successfully - object - `status` boolean - `data` AgentVersion — Represents either a draft revision or a published version of an agent's configuration. - `_id` string — Unique identifier - `agent` string — The agent this version belongs to - `status` 'published' | 'draft' | 'archived' — Current status of the version record - `versionNumber` integer, nullable — Auto-incremented version number (published versions only) - `label` string, nullable — Human-readable label for the version - `description` string, nullable — Description of what changed in this version - `isPinned` boolean — Whether the version is pinned for quick access - `publishedBy` string, nullable — User ID of who published this version - `draftId` string, nullable — Unique draft identifier (drafts only) - `draftName` string, nullable — Human-readable draft name - `draftRevision` integer, nullable — Revision number within the draft (drafts only) - `sourceVersionId` string, nullable — The published version this draft was branched from - `blocks` object — References to the 13 config section blocks - `workflow_prompt` string - `workflow_tools` string - `workflow_graph` string - `llm` string - `voice` string - `language` string - `call_handling` string - `detection` string - `analytics` string - `timeouts` string - `audio` string - `privacy` string - `widget` string - `workflowType` 'workflow_graph' | 'single_prompt' — The type of workflow configuration. workflow_graph uses a node-based visual workflow, single_prompt uses a simple prompt-based configuration. - `parentVersion` string, nullable — The version this was derived from - `isActive` boolean — Whether this is the currently active version for the agent - `createdBy` string — User ID of who created this record - `createdAt` string, date-time - `updatedAt` string, date-time ## Other responses - `400` — Invalid input - `401` — Unauthorized access - `500` — Internal server error ## Changes - **2026-05-26** `7131fe74f3cb` — 3 breaking, 4 warning, 16 info - removed `subschema #1, subschema #2` from the `allOf[#/components/schemas/CreateAgentRequest]/synthesizer/voiceConfig` request property `oneOf` list - the `allOf[#/components/schemas/CreateAgentRequest]/synthesizer/voiceConfig` request property type/format changed from ``/`` to `object`/`` - the `allOf[#/components/schemas/CreateAgentRequest]/telephonyProductId` request property type/format changed from `string`/`` to `array`/`` - the `allOf[#/components/schemas/CreateAgentRequest]/globalPrompt` request property's maxLength was set to `4000` - …19 more - **2026-05-06** `98bb2336fc95` — 15 info - added the new optional request property `allOf[#/components/schemas/CreateAgentRequest]/allowInterruptions` - added the new optional request property `allOf[#/components/schemas/CreateAgentRequest]/denoisingConfig` - added the new optional request property `allOf[#/components/schemas/CreateAgentRequest]/firstMessage` - added the new optional request property `allOf[#/components/schemas/CreateAgentRequest]/interruptionBackoffTimer` - …11 more - …earlier changes not shown [Full history](https://skmtc.dev/smallest-inc/apis/agent-management-api/changes/agent/:id/drafts/:draftId/config/patch.md) --- [API](https://skmtc.dev/smallest-inc/apis/agent-management-api.md) · [All operations](https://skmtc.dev/smallest-inc/apis/agent-management-api/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/smallest-inc/agent-management-api/revisions/2bfbd089ccb3/schema)