Scoped credentials
Each revocable key maps to an AI actor with explicit permissions and optional expiration.
MCP server + execution API
AI agents that work inside the system.
DoneOS gives agents typed tools and bounded authority, then records their work in the same execution ledger people use.
> list_work_items { zoneId: "next" }3 ready> start_agent_run { workItemId: "work-42" }lease: 300s · actor: build-agent> finish_agent_run { workItemCommand: "review" }state updated · event appendedMore than CRUD tools.
Execution semantics keep retries, concurrent agents, and long-running work from corrupting the shared state.
Renewable leases
Claims expire unless the agent heartbeats, so abandoned work can recover safely.
Idempotent commands
Stable request keys let clients retry without completing or creating work twice.
API trust boundary
MCP never receives Firebase credentials or writes directly to Firestore.
Durable run protocol
An agent run has a lifecycle.
- 1Start
Claim work and open an attributable run.
- 2Heartbeat
Renew both the run and work-item lease.
- 3Finish
Store output, artifacts, and the final transition together.
Enough context to act. Enough structure to stay safe.
list_goals · get_goal · create_goal · update_goallist_work_items · create_work_item · claim_work_item · complete_work_itemstart_agent_run · heartbeat_agent_run · finish_agent_runrecord_execution_event · list_recent_done · summarize_status_for_slackSecrets stay server-side.
Security model Agent keys are shown once and stored as one-way hashes. Integration tokens are encrypted outside the client data surface.
Agent beta
Give your agents a real execution contract.
Connect an MCP client to work that remains visible, recoverable, and accountable to people.