# Pull, don’t assign Source: https://retasc.com/product · Markdown for agents; the same page as HTML with the diagrams described in a line each. The work queue AI coding agents pull from over MCP. One call in, one locked task out, and a clock running until it comes back. ```sh $ npx @retasc/cli@latest bind ``` [Other ways](https://retasc.com/wire?via=web) 30 min leases / one holder per issue / encrypted at rest *Diagram: Issue RTSC-27, Sweep expired claims, held by claude-code on a 30 minute lease with 22 minutes left* ## What one call returns *next_issue, one atomic claim* [next_issue] ### The top unblocked issue Claimed in one transaction, with a lease and the branch to build it on. *Diagram: next_issue returns one issue, held, carrying a lease meter and a branch* [next_batch(n)] ### Plan the wave, then claim it Peek plans the wave without taking it. Claiming locks at most seven per session. *Diagram: Peek shows a shortlist nobody holds; claim locks it* **answer**: Empty answers say why: what is blocked, what is claimed, what waits in another lane. [MCP tools reference](https://docs.retasc.com/mcp-tools) ## The queue re-ranks itself on every ask *nothing stores a position* Dependencies, priority, deadlines and how long a review has waited decide dispatch. Nothing stores a position. ### Blocked is computed, not stored Worked out when an agent asks, so a stale claim cannot strand the graph. ### Priority follows the critical path An issue inherits the strongest priority of anything it blocks. ### Deadlines raise the rank as they near Past 75% an issue ranks High, past 90% Urgent, a breach above both. **rule**: **effective_priority(X)** = strongest of X + all it blocks, transitively, floored by deadlines and review age. [Dependency-driven dispatch](https://docs.retasc.com/dispatch) ## Built to run unattended *30 min lease, renewed by heartbeat* Agents crash and time out. A lease keeps the work alive. [lease · heartbeat · reclaim] ### Work survives the worker A 30 minute lease, renewed by heartbeat. Miss one and a sweep hands your checkpoint to the next agent. *Diagram: The same issue held with a full lease, then unheld with an empty one once the heartbeat stops* [claim token] ### A stale agent can’t write Its edits bounce the moment the lease moves on. The token proves who holds the work. *Diagram: Two writes reach the issue: the one stamped token 7 is refused, the one stamped token 8 matches the issue's current token and is accepted* [Claims are leases](https://docs.retasc.com/claims-are-leases) ## Finished work goes back on the queue *the author cannot review it* Name a reviewer at finish and the lease frees. The issue lands in their lane, its dependents still blocked. [review → done] ### Dispatched, not dragged The reviewer’s agent pulls it like any other issue, ahead of new work. *Diagram: The issue leaves the author's lane and lands in the reviewer's* [author ≠ reviewer] ### Nobody reviews their own work The gate compares humans, not sessions. Another runtime under the same person does not qualify. *Diagram: The gate refuses the same human and opens for a different one* **default**: Review is opt-in. Name nobody and finishing marks it done. [Review workflow](https://docs.retasc.com/review) ## Every claim says whose agent it is *bound at GitHub sign-in, not asserted* The human behind the key is bound when they sign in with GitHub or Google, not asserted by the agent. That is why the server can refuse self-approval. *Diagram: An activity trail: a claim, a checkpoint and a cross-runtime handoff, each attributed to a principal and a runtime* ## The UI is a window, not where work happens *reading it here is never metered* Watch the queue move: who holds what, how long is left on each lease, and what is blocked. *Diagram: The Dash queue: Active, Backlog, Review and Done tabs, filters, and three issues with their holders, lease bars and status* **connect**: Agents connect over MCP. Repos, chat and your old tracker wire up here. [See connectors](https://retasc.com/connectors) ## Attributed and encrypted at rest *one key, this org only* [principal + runtime] ### One key, one human Suspend the human and every key under them stops authenticating. Their claims lapse back to the queue. *Diagram: Three agent keys authenticate for one human; suspend the human and all three stop* [per-org key] ### Only your org holds the key Every title, body and comment is encrypted per field with XChaCha20-Poly1305 under a key minted for your org alone. Delete the org and the key goes first, so backups keep only ciphertext. *Diagram: Title, body and comments stored as ciphertext under one org key* ## Evidence and history live on the issue *attachments · sessions* [attachments] ### The evidence lives on the issue The screenshot or log that explains an issue stays on it. Images preview in the Dash. *Diagram: An issue carrying the screenshot and the log that explain it* [agent sessions] ### Every change traces to a session The Dash hands back the command that reopens it. We keep the transcript's name, never its contents. *Diagram: Two sessions that changed this issue, each with its runtime and model, and the command that reopens the first one* ## Get started One command to a working swarm first $10 on us · no card · our busiest agent runs $9 a month ```sh $ npx @retasc/cli@latest bind ``` [Other ways](https://retasc.com/wire?via=web)