| name | tracker |
| description | File-based issue tracker conventions — how specs, tickets, and small tasks live in agent/tasks/ and how to publish, fetch, claim, and retire them. Use when publishing or fetching a spec/ticket/task, picking work from the frontier, or when another skill says "publish to the issue tracker". |
Tracker
Issues for this repo live as markdown files in agent/tasks/. These conventions are the default — if the project's CLAUDE.md declares a different tracker (e.g. GitHub Issues), follow that instead.
Layout
- Feature: one directory per feature,
agent/tasks/<feature-slug>/
spec.md — the work order for the whole feature (written by /mx:to-spec)NN-<slug>.md — tickets, numbered from 01 (written by /mx:to-tickets)
- Small standalone task: a single file
agent/tasks/<slug>.md — ticket-shaped, no spec needed
Ticket state
Frontmatter:
status: open | claimed | done
blocked-by: [01, 02]
- A ticket is unblocked when every ticket in
blocked-by is done.
- The frontier: open, unblocked, unclaimed tickets — what can be started right now.
claimed marks a ticket a session is actively working. Set it before any work. When agents run in parallel, a single orchestrating agent oversees them and is the sole claim-writer — no cross-checkout coordination needed. (With a single agent in a single checkout, claiming is optional.)- Notes and follow-up conversation append under a
## Comments heading at the bottom of the file.
Publish / fetch
- "Publish to the issue tracker" → create the files above (creating the feature directory if needed).
- "Fetch the ticket" → read the ticket file and the feature's
spec.md — tickets don't repeat the feature context, the spec carries it.
Retire
Set status: done when a ticket completes. When the whole feature has shipped, git rm -r agent/tasks/<feature-slug>/ — git history preserves it (git log --diff-filter=D -- agent/tasks finds retired work). If the repo doesn't track agent/tasks/, plain delete.
