The MCP server is in public beta. It is open to use and still being refined, so tools and behavior may change.
Tool
ask_gcai · Behavior Open worldfile_ids), playbooks (playbook_ids), or skills (skill_ids). Returns a job envelope, not a final answer: while status is pending or running, keep polling ask_gcai_status with the returned job_id until status is succeeded (then result holds the answer). Pass chat_id to continue a prior conversation, sending only the new message (conversation state is kept server-side).
Input parameters
string
required
The user’s message or prompt
string (uuid)[]
Optional uploaded file IDs to attach as context for this completion. Upload files first via
POST /files.string (uuid)[]
Optional playbook IDs to ground the completion in (up to 20). The model uses each playbook’s checks and guidance to structure its review of the attached files. Discover playbooks via
GET /playbooks.Org-scoped keys can use org-visible and official playbooks.string
Deprecated and no longer accepted. Use
playbook_ids (an array) instead. Requests that include this field are rejected with a 400 so the playbook is never silently dropped.string (uuid)[]
Optional skill IDs to run on the first turn. Each skill’s instructions are injected as context for this completion. Discover skills via
GET /skills.Only valid when starting a new chat; omit them when continuing with chat_id.string (uuid)
Optional chat ID to continue an existing conversation, from the
chat_id of a prior completion. Omit to start a new chat.Conversation state is held server-side, so send only your new message (plus any new file_ids); do not re-send prior turns or previously returned documents/emails/diagrams. Only one turn may be in flight per chat at a time. See Multi-turn Conversations.string (uuid)
Optional project to file the chat into, so it is browsable under
GET /projects/{id}/chats. Requires write access to the project. The chat becomes visible in the project once you materialize it. Discover projects via GET /projects.string (uuid)
Optional company profile to ground the completion in, so the model has that company’s context (industry, jurisdiction, regulations, risk posture). Discover company profiles via
GET /company-profiles.The company is fixed for the life of a chat. On a new chat, supply this to target a specific company (the only way to ground against a particular one in a multi-company organization); when omitted, a company is auto-resolved: user-scoped keys use the caller’s active (default) company, and org-scoped keys use the organization’s sole company, or none when there are several.When continuing a chat with chat_id, omit this to reuse the chat’s company. You may echo the same company_id, but a different one is rejected with 409 (start a new chat to use a different company).Response
string
Job identifier. Pass it to the matching
*_status tool to poll.'pending' | 'running' | 'succeeded' | 'failed' | 'canceled'
succeeded means result is populated; running/pending means poll again; failed/canceled are terminal.object | null
Typed result when
status is succeeded; null until then.string
The exact tool to call to poll this job. Call it yourself with
job_id; do not ask the user to poll.number | null
While
status is pending/running, wait this many seconds, then call status_tool with job_id. null once the job is terminal.string
Plain-language next step. Follow it before responding.