> ## Documentation Index > Fetch the complete documentation index at: https://docs.getcliq.io/llms.txt > Use this file to discover all available pages before exploring further. # Build a team with AI > Builds a team with AI. Choose what to do with `action`: - `generate` (`intent`): start drafting a whole team from a description. Returns a `job_id` at once; poll with `status`. - `status` (`job_id`): the job's progress (`stage`), then the generated team and its validation when `status` is `done`, or `error`. - `improve_role` (`role_name`, `role_content`, optional `instruction`): a rewritten role prompt. - `suggest` (`team_name`, optional `description`, `phases`, `roles`): suggestions to improve a team. - `validate` (`team`): check a team snapshot; returns `valid`, `errors`, `warnings`. - `chat` (`team`, `message`, `history`): one turn of a conversation about a team; returns a `reply` and proposed `actions`. Requires a signed-in user token (`Authorization: Bearer cliq_tok_…`). Results are limited to what that user can see. ## OpenAPI ````yaml /openapi/hub-by-tag/hub-teams.yaml post /v1/teams/build openapi: 3.0.3 info: title: 'CliqHub API: Teams' version: 1.0.0 description: >- The CliqHub team registry. A team is published under a scope (`@acme/code-review`) as numbered versions. Create a draft, change it (each change can mint the next version), publish it to make it installable, then install it on daemons. List teams with a `realm_id` to see a realm's roster and where each team is installed, or with a `daemon_id` to see what one daemon has. Build a team with AI drafts a team from a description. Changing a team (update, rename, publish, unpublish, delete) needs its author, or the publish permission in the organization that owns its scope. servers: - url: https://{host} variables: host: default: cliqhub.io description: CliqHub host security: [] tags: - name: Teams description: >- The CliqHub team registry. A team is published under a scope (`@acme/code-review`) as numbered versions. Create a draft, change it (each change can mint the next version), publish it to make it installable, then install it on daemons. List teams with a `realm_id` to see a realm's roster and where each team is installed, or with a `daemon_id` to see what one daemon has. Build a team with AI drafts a team from a description. Changing a team (update, rename, publish, unpublish, delete) needs its author, or the publish permission in the organization that owns its scope. paths: /v1/teams/build: post: tags: - Teams summary: Build a team with AI description: >- Builds a team with AI. Choose what to do with `action`: - `generate` (`intent`): start drafting a whole team from a description. Returns a `job_id` at once; poll with `status`. - `status` (`job_id`): the job's progress (`stage`), then the generated team and its validation when `status` is `done`, or `error`. - `improve_role` (`role_name`, `role_content`, optional `instruction`): a rewritten role prompt. - `suggest` (`team_name`, optional `description`, `phases`, `roles`): suggestions to improve a team. - `validate` (`team`): check a team snapshot; returns `valid`, `errors`, `warnings`. - `chat` (`team`, `message`, `history`): one turn of a conversation about a team; returns a `reply` and proposed `actions`. Requires a signed-in user token (`Authorization: Bearer cliq_tok_…`). Results are limited to what that user can see. operationId: POST_v1_teams_build requestBody: required: true content: application/json: schema: type: object properties: action: type: string enum: - generate - status - improve_role - suggest - validate - chat description: >- `generate`, `status`, `improve_role`, `suggest`, `validate` or `chat`. intent: type: string maxLength: 5000 description: 'generate: what the team should do (up to 5000 characters).' job_id: type: string description: 'status: the job from `generate`.' role_name: type: string description: 'improve_role: the phase whose role to rewrite.' role_content: type: string maxLength: 50000 description: >- improve_role: the current role prompt (up to 50000 characters). team_name: type: string description: 'improve_role, suggest: the team''s name.' team_description: type: string description: 'improve_role: the team''s description.' phases: type: array items: type: object additionalProperties: true description: 'improve_role, suggest: the team''s phase names.' instruction: type: string description: 'improve_role: what to focus on.' description: type: string description: 'suggest: the team''s description.' roles: type: array items: type: object additionalProperties: true description: 'suggest: the team''s roles.' team: type: object additionalProperties: true description: >- validate, chat: the team snapshot (as the console's builder holds it). message: type: string maxLength: 5000 description: 'chat: your message (up to 5000 characters).' history: type: array items: type: object properties: role: type: string enum: - user - assistant content: type: string required: - role - content description: >- chat: earlier turns, each `{ role: user | assistant, content }`. required: - action examples: generate: value: action: generate intent: >- Review each pull request for security issues and open a Jira ticket for every finding. status: value: action: status job_id: 2b3c4d5e-6f70-4812-9a3b-4c5d6e7f8091 responses: '200': description: Depends on `action`. content: application/json: schema: type: object required: - ok - data properties: ok: type: boolean enum: - true description: Always `true` on success. data: type: object additionalProperties: true properties: job_id: type: string description: generate, status. status: type: string enum: - queued - running - done - error description: generate, status. stage: type: string description: 'generate, status: the step running now.' team: type: object additionalProperties: true description: 'status (done), chat: the generated or updated team.' validation: type: object additionalProperties: true description: 'status (done): the generated team''s validation.' error: type: object properties: code: type: string message: type: string description: status (error). improved_content: type: string description: improve_role. changes_summary: type: string description: improve_role. suggestions: type: array items: type: object additionalProperties: true description: suggest. valid: type: boolean description: validate. errors: type: array items: type: object additionalProperties: true description: validate. warnings: type: array items: type: object additionalProperties: true description: validate. reply: type: string description: chat. actions: type: array items: type: object additionalProperties: true description: 'chat: proposed changes to the team.' examples: generate: summary: generate value: ok: true data: job_id: 2b3c4d5e-6f70-4812-9a3b-4c5d6e7f8091 status: queued stage: queued validate: summary: validate value: ok: true data: valid: false errors: - path: phases.1.agent message: Unknown agent 'jira-syncc' warnings: [] '401': description: >- `unauthorized`: No `Authorization` header, or the token is malformed, expired or revoked. content: application/json: schema: $ref: '#/components/schemas/HubErr' examples: unauthorized: summary: unauthorized value: ok: false error: code: unauthorized message: Authentication required '403': description: '`forbidden`: A daemon token was used; this route needs a user token.' content: application/json: schema: $ref: '#/components/schemas/HubErr' examples: forbidden: summary: forbidden value: ok: false error: code: forbidden message: You do not have access to do this '404': description: '`not_found`: `status`: the job does not exist or has expired.' content: application/json: schema: $ref: '#/components/schemas/HubErr' examples: not_found: summary: not_found value: ok: false error: >- Generate job '2b3c4d5e-6f70-4812-9a3b-4c5d6e7f8091' not found or expired code: not_found '422': description: >- `invalid_params`: `action` is unknown or a field it needs is missing or too long (for example `intent` for `generate`). content: application/json: schema: $ref: '#/components/schemas/HubErr' examples: invalid_params: summary: invalid_params value: ok: false error: intent is required for generate action code: invalid_params '500': description: >- `internal_error`: An unexpected server fault. Retry later; if it persists, report it with the time of the request. `generation_failed`: The AI model failed or returned something unusable. Try again. content: application/json: schema: $ref: '#/components/schemas/HubErr' examples: internal_error: summary: internal_error value: ok: false error: Internal server error code: internal_error generation_failed: summary: generation_failed value: ok: false error: 'Failed to generate roles for phases: review' code: generation_failed security: - BearerAuth: [] components: schemas: HubErr: type: object required: - ok - error description: >- Every failure. Route-policy denials (401/403/404, and 400 when neither realm_id nor org_id was sent) answer `error` as `{ code, message }`; errors raised by the endpoint answer `error` as the message with the code beside it in `code`. Every error has a code. See [Errors](/hub-api/errors). properties: ok: type: boolean description: Always false on error error: description: The message, or `{ code, message }` for an access denial. oneOf: - type: string - type: object required: - code - message properties: code: type: string message: type: string code: type: string description: Stable error code of an endpoint error. details: type: object additionalProperties: true description: Structured detail, when there is some. securitySchemes: BearerAuth: type: http scheme: bearer bearerFormat: cliq_tok_* | cliq_dt_* description: >- User PAT (`cliq_tok_…`, including session-scoped) or realm token (`cliq_dt_…`). ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.