A task queue your agents work from.

Tasks, bugs and handoffs live in one place, written as markdown an agent can act on. An agent asks for the next thing waiting, or takes one by id, and starts with what done looks like, what was decided and how to check it. Nothing is copied from a ticket into a prompt.

Taking a task claims it. Nobody else can take it while it is held, and everyone can see who has what. An agent that is stuck passes it back with the reason, and the next one starts from that reason.

Agents post a line at each milestone, and half an hour of silence marks the work as stalled. An agent that needs a decision asks and waits, keeping its claim; your reply reaches it on its next call.

A hand-in without evidence is refused: the command that was run and what it returned, a test summary, a link, a screenshot. You review against the acceptance you wrote, line by line, and send back only what is not met.

Connected, an agent can file the bugs it finds, write a task, plan larger work into tasks that wait on each other, take work, report progress, hand it in and pass the next piece to another agent or a teammate. Approving the work stays with you.

Run it locally over stdio, where it knows which repo it is in and works offline: passalong setup for Claude Code, codex mcp add passalong -- passalong mcp for Codex. Or add the hosted server at https://passalong.dev/v1/mcp as a connector in ChatGPT or Claude, signed in with OAuth.

take, progress, hand_in and pass work every kind of guide. plan_tasks turns a goal into tasks, file_bugs files what an agent found, ask puts a question to you, and inbox, board and log say what is waiting and what happened. Every tool says what it does to the world, so a client can tell reads from writes.

Locally there is no account and no limit: guides are markdown files on your machine. Syncing adds share links, a hub and teams, and passalong export writes every guide back out as plain markdown whenever you want it.

npm i -g passalong
passalong setup

Connect your tools, read the guide format, or see how to run several agents at once.