> ## Documentation Index > Fetch the complete documentation index at: https://support.profilebehavior.com/llms.txt > Use this file to discover all available pages before exploring further. # Generate AI Debrief > Returns a Playbook AI generated debrief for the individual profile specified. If one has been generated with the current available data already, it simply returns the expected debrief. If new data or a new AI model are available it generates a new debrief with a flag indicating that it is new. ## OpenAPI ````yaml POST /playbook/ai/debrief/{profileId} openapi: 3.1.0 info: title: Profile Behavior API description: > Unlock actionable behavioral insights for leadership, teamwork, recruiting, and team culture. Profile delivers powerful assessments, analytics, and reporting for businesses, athletic teams, and higher education. For full product information, code samples, and API reference, see our [support documentation](https://support.profilebehavior.com). > **LLMs and AI agents**: See [llms.txt](https://support.profilebehavior.com/llms.txt) for a machine-readable index of all documentation pages. > **MCP connector:** LLM clients (ChatGPT apps, Claude connectors) speak to `/mcp` over streamable HTTP. They discover OAuth automatically via [/.well-known/oauth-protected-resource](https://vanguard.profilebehavior.com/.well-known/oauth-protected-resource), returned in the `WWW-Authenticate` header on any 401. > **Legacy manifests:** - [/.well-known/ai-plugin.json](https://vanguard.profilebehavior.com/.well-known/ai-plugin.json) - [/.well-known/mcp.json](https://vanguard.profilebehavior.com/.well-known/mcp.json) version: 3.0.0 termsOfService: https://profilebehavior.com/legal/terms contact: name: Support url: https://profilebehavior.com/contact email: support@profilebehavior.com x-llm-sitemap: https://support.profilebehavior.com/llms.txt x-documentation-site: https://support.profilebehavior.com x-ai-plugin-manifest: https://vanguard.profilebehavior.com/.well-known/ai-plugin.json x-mcp-manifest: https://vanguard.profilebehavior.com/.well-known/mcp.json x-mcp-endpoint: https://vanguard.profilebehavior.com/mcp x-oauth-protected-resource: https://vanguard.profilebehavior.com/.well-known/oauth-protected-resource servers: - url: https://vanguard.profilebehavior.com/api/v4 security: - M2M: [] - OAuth2: [] tags: - name: test description: Endpoints for testing the API. - name: assessment_invites description: Information on sending invitations to take the assessment. - name: assessments description: Behavioral, leadership, and communication assessment endpoints. - name: accounts description: >- Endpoints for retrieving and managing account and organization information. - name: fixtures description: Endpoints for retrieving fixture data content such as report language. - name: playbook_ai description: Endpoints for generating and retrieving Playbook AI related content. - name: profiles description: Endpoints for individual behavioral profile management. - name: resource_aliases description: >- Endpoints for managing resource alias records that link Profile data to custom defined keys. - name: tags description: >- Endpoints for individual profile tags, categorization, and group management. paths: /playbook/ai/debrief/{profileId}: post: tags: - playbook_ai summary: Generate an AI debrief for an individual profile description: > Returns a Playbook AI generated debrief for the individual profile specified. If one has been generated with the current available data already, it simply returns the expected debrief. If new data or a new AI model are available it generates a new debrief with a flag indicating that it is new. operationId: GenerateAIDebrief parameters: - name: profileId in: path description: | The ID of the profile to generate and retrieve debrief content for. required: true schema: type: string responses: '200': description: Success content: application/json: schema: allOf: - $ref: '#/components/schemas/Success' - type: object properties: id: type: string description: The ID of the debrief content document. content: type: string description: > The generated text content debrief meant to help summarize the data of the profile indicated by the `profileId` ratings: type: array description: > An array of user-submitted ratings indicating whether they thought the debrief was good or not, as well as any comments they may have about the given content. items: type: object properties: user: type: string required: true description: The ID of the user that submitted the feedback. rating: type: integer required: true description: >- Whether the content was considered good (+1) by the user or bad (-1). enum: - 1 - -1 comments: type: string description: > Optional feedback written by the user regarding the debrief. example: code: 200 id: 60e9ca73c500a9001534ad84 content: > They lead with Correctness and steady habits: calm, thorough, and careful with commitments, they set a measured pace and expect clean execution. Highly organized yet very flexible, they can pivot when needed without drama, but they still prefer clear plans, few surprises, and time to get the details right. They communicate sparingly and straight to the point, often choosing quiet focus over group chatter, which can make them seem reserved, though teammates learn they’re dependable and consistent. They’re motivated by Freedom to run their lane, Wealth tied to measurable outcomes, and Wisdom gained from digging into how things work; they dislike sloppy work, public criticism, and rushed changes that upend hard-won order. ratings: - user: 60e9ca73c500a9001534ad84 rating: 1 comments: This was pretty helpful to summarize their report. status: success '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' components: schemas: Success: type: object description: > Standard response object for successful single-resource requests in the Profile Behavior API. Contains an HTTP status code, the returned resource data, and a status indicator. required: - code properties: code: type: integer format: int32 description: HTTP status code of the successful response. status: type: string description: Always "success" for successful responses. enum: - success Error: type: object description: > Standard error response object returned by the Profile Behavior API. Contains a machine-readable error code, a human-readable message, and detailed technical information for debugging or support purposes. required: - code - error - status properties: code: type: integer format: int32 description: HTTP status code associated with the error. error: type: object required: - details - timestamp properties: message: type: string description: A human-readable summary describing the error. timestamp: type: string format: date-time description: ISO 8601 timestamp when the error occurred. status: type: string description: Always "error" for error responses. enum: - error responses: BadRequest: description: Invalid request body content: application/json: schema: $ref: '#/components/schemas/Error' example: code: 400 error: message: Invalid request body timestamp: '2025-01-01T09:42:52.329056-05:00' status: error Unauthorized: description: Unauthorized - authentication required. content: application/json: schema: $ref: '#/components/schemas/Error' example: code: 401 error: message: Invalid token timestamp: '2025-01-01T09:42:52.329056-05:00' status: error securitySchemes: M2M: type: http scheme: bearer OAuth2: type: oauth2 flows: authorizationCode: authorizationUrl: https://login.profilebehavior.com/oauth/authorize tokenUrl: https://login.profilebehavior.com/oauth/token scopes: profiles:read: Read profiles and behavioral summaries (scores, AI debriefs) profiles:write: Create, update, delete profiles and send assessment invites tags:manage: Create, update, delete, and assign tags account:read: Read account details, assessments, and resource aliases account:write: Update account settings and resource aliases description: > OAuth 2.1 authorization code flow with PKCE. Use this whenever your integration acts on behalf of a specific Profile user: third-party apps, AI agents and MCP servers, and any browser-based or CLI tool a user signs in to. Tokens issued under this scheme are scoped to one Profile account (the one the user consented from) and carry OAuth scopes that narrow what the integration can do. Discovery is served at [/.well-known/oauth-protected-resource](/.well-known/oauth-protected-resource). The authorization server is Stytch Connected Apps, which supports Dynamic Client Registration for on-the-fly clients (e.g. MCP agents) in addition to pre-registered integrations. ````