> ## Documentation Index > Fetch the complete documentation index at: https://docs.octavehq.com/llms.txt > Use this file to discover all available pages before exploring further. # Update Agent > Update an existing agent ## OpenAPI ````yaml post /api/v2/agents/update openapi: 3.0.0 info: version: 1.0.0 title: Octave API description: API for Octave workspace management and AI-powered content generation servers: - url: https://app.octavehq.com security: - ApiKeyAuth: [] paths: /api/v2/agents/update: post: tags: - Agents summary: Update Agent description: Update an existing agent operationId: updateAgent requestBody: description: Agent update input content: application/json: schema: type: object properties: oId: type: string minLength: 1 description: The OId of the agent to update example: agent_123 name: type: string description: The name of the agent description: type: string description: The description of the agent model: type: string enum: - NOTE - PULSE - ECHO - HARMONY - CHORUS - SYMPHONY description: The model to use for the agent productOId: type: string description: The OId of the product/offering to associate with the agent serviceOId: type: string description: >- Alias for productOId — use either field, not both. Maps to the same internal product/offering. playbookOId: type: string description: The OId of the playbook to associate with the agent personaOIds: type: array items: type: string description: The OIds of the personas to associate with the agent useCaseOIds: type: array items: type: string description: The OIds of the use cases to associate with the agent segmentOIds: type: array items: type: string description: The OIds of the segments to associate with the agent competitorOIds: type: array items: type: string description: The OIds of the competitors to associate with the agent referenceOIds: type: array items: type: string description: >- The OIds of the reference customers to associate with the agent proofPointOIds: type: array items: type: string description: The OIds of the proof points to associate with the agent enableBrandVoice: type: boolean description: Whether to enable brand voice for the agent enableWebSearch: type: boolean description: >- Whether to enable web search. Only enable when explicitly requested webSearchInstructions: type: string description: Custom instructions for the web search tool enableCrmActivity: type: boolean description: Whether to enable CRM activity for the agent crmActivityInstructions: type: string description: Custom instructions for the CRM activity tool enableHighEffortMode: type: boolean description: Whether to enable high-effort mode for the agent sequenceSettings: type: object properties: sequenceType: type: string enum: - COLD_OUTBOUND - WARM_OUTBOUND - INBOUND - CUSTOM - CVI_WHY_CHANGE - BARE_METAL description: >- Sequence type (e.g. COLD_OUTBOUND, WARM_OUTBOUND). Changing this does not reset step-level configs. numEmails: type: number nullable: true description: Number of emails in the sequence (default 4) sequenceTypeCustomInstructions: type: string nullable: true description: >- Custom instructions specific to the sequence type (e.g. context about the cold outbound campaign) sequenceCustomInstructions: type: string nullable: true description: >- General custom instructions applied across the entire sequence writingStyleOId: type: string nullable: true description: OId of a saved writing style to apply to the sequence generateUniqueSubjectLines: type: boolean nullable: true description: >- If true, generate a unique subject line per email instead of reusing one tone: type: string enum: - Casual - Neutral - Formal description: >- Writing tone applied to all emails (e.g. Casual, Professional, Formal) readingLevel: type: string enum: - 8th Grade - High School - College - Professional description: >- Reading level for generated text (e.g. High School, College, PhD) applyCtaToAllEmails: type: boolean description: >- If true, apply the same CTA config to every email in the sequence applySubjectToAllEmails: type: boolean description: >- If true, apply the same subject config to every email in the sequence description: >- EMAIL agents only: sequence-level settings (type, numEmails, tone, reading level, instructions, writingStyleOId, etc.) stepPatches: type: object additionalProperties: type: object properties: methodology: type: string enum: - Octave - Octave + Custom - Custom description: >- Sales methodology for this step. Set to Custom or Octave + Custom when providing customInstructions. focus: type: string nullable: true enum: - Pain - Neutral - Benefits description: >- Email focus area for this step (e.g. PAIN_POINT, PRODUCT) customInstructions: type: string description: >- Custom instructions for this step. Only applied when methodology is Custom or Octave + Custom. length: type: object properties: wordCountPreset: type: string enum: - EXTRA_SHORT - SHORT - MEDIUM - LONG - EXTRA_LONG - CUSTOM wordCount: type: object properties: min: type: number max: type: number required: - min - max numSentences: type: number numParagraphs: type: number description: >- Length controls: wordCountPreset (SHORT, MEDIUM, LONG, EXTRA_SHORT) or custom wordCount min/max, numSentences, numParagraphs subject: type: object properties: includePersonName: type: boolean default: false includeCompanyName: type: boolean default: false allLowercase: type: boolean default: false length: type: object properties: minWords: type: number default: 3 maxWords: type: number default: 5 default: minWords: 3 maxWords: 5 customInstructions: type: string examples: type: array items: type: string description: Subject line configuration for this step cta: type: object properties: type: anyOf: - type: string enum: - INITIATE_CONVERSATION - SCHEDULE_MEETING - REQUEST_INTRODUCTION - ASK_FOR_FEEDBACK - FIND_TIME_TO_CONNECT - POLITELY_DISENGAGE - SMART_CTA - type: string enum: - CUSTOM description: >- CTA preset or CUSTOM for free-form CTA instructions customInstructions: type: string description: >- Custom CTA instructions (only used when type is CUSTOM) smartCtaConfig: type: object properties: mode: type: string enum: - NONE - AUTO - SPECIFIC description: >- Smart CTA mode controlling how CTAs are generated ctaOIds: type: array items: type: string description: OIds of smart CTA entities to use description: Smart CTA configuration for AI-generated CTAs description: Call-to-action configuration for this step trainingExamples: type: object properties: mode: type: string enum: - paste - updatePaste - remove description: >- paste: create a new training example; updatePaste: update an existing one in place; remove: delete a training example from the step bodyText: type: string minLength: 1 description: >- The example email body text. Required for paste and updatePaste modes; ignored for remove. flexibility: type: string nullable: true enum: - STRICT - LOOSE description: >- How closely the generated email should follow this example (e.g. STRICT, FLEXIBLE) targetTrainingIndex: type: integer minimum: 1 description: >- For updatePaste/remove: 1-based index into this step's training examples (first example is 1). Omitted means 1. Ignored for paste. required: - mode description: >- Training example management: paste new examples, update existing ones, or remove them skipReferenceCustomer: type: boolean description: If true, skip reference customer mentions in this step skipProofPoint: type: boolean description: If true, skip proof point mentions in this step stepCustomPrompt: type: string description: >- Custom prompt appended to this step's generation instructions description: >- EMAIL agents only: map of 1-based step index to partial email builder config (same numbering as sequence run runtimeContext keys). Deep-merged into writingStyle.emails[i] after update. trainingExamples.targetTrainingIndex is 1-based (default 1). callPrepMethodology: type: string enum: - BANT - SPIN - CHAMP - MEDDIC - SANDLER - CHALLENGER - AIDA - NEAT - SPICED - WHY_CHANGE description: >- Call prep methodology (CALL_PREP agents only). Controls discovery questions and call script framing. instructions: type: string description: Instructions for content generation (CONTENT agents only) examples: type: array items: type: string description: Example outputs for content generation (CONTENT agents only) outputFormat: type: string enum: - JSON - TEXT - HTML - MARKDOWN description: >- The output format for content generation (CONTENT agents only) jsonSchemaInstructions: type: string description: >- JSON schema instructions for content generation (CONTENT agents only, used with JSON output format) data: type: object additionalProperties: nullable: true description: >- Full agent configuration data passthrough. When provided, this will be merged with the simplified fields above, with simplified fields taking precedence. required: - oId responses: '200': description: Agent updated successfully content: application/json: schema: type: object properties: _metadata: $ref: '#/components/schemas/Metadata' data: type: object properties: oId: type: string description: The OId of the agent name: type: string description: The name of the agent description: type: string description: The description of the agent type: type: string enum: - PROSPECTOR - CONTENT - EMAIL - ENRICH_PERSON - ENRICH_COMPANY - QUALIFY_PERSON - QUALIFY_COMPANY - CALL_PREP - CONTEXT description: The type of the agent data: nullable: true description: The full agent configuration data createdAt: type: string nullable: true description: The creation date of the agent updatedAt: type: string nullable: true description: The last updated date of the agent costToRun: type: number description: The cost to run the agent warnings: type: array items: type: string description: >- Warnings about type-mismatched fields that were provided but ignored required: - oId - name - description - type - createdAt - updatedAt - costToRun required: - _metadata - data '400': description: Bad request content: application/json: schema: type: object properties: _metadata: $ref: '#/components/schemas/Metadata' message: type: string required: - _metadata - message '404': description: Agent not found content: application/json: schema: type: object properties: _metadata: $ref: '#/components/schemas/Metadata' message: type: string required: - _metadata - message deprecated: false components: schemas: Metadata: type: object properties: usage: type: number default: 0 description: API usage example: 0 requestId: type: string description: Request ID example: requestId message: type: string description: Message example: message timestamp: type: string description: Timestamp example: '2021-01-01T00:00:00.000Z' required: - requestId - timestamp securitySchemes: ApiKeyAuth: type: apiKey in: header name: api_key ````