pm
Skill4 starsChanged 3 months agoMITView on GitHub
Run the outer agentic loop once. Triage the backlog, pick the most valuable ready item, shape its spec, and hand it to the inner loop. Use when the user asks "what's next", "groom the backlog", "run the PM", or on a schedule. Pass --dispatch to label the chosen issue for CI instead of building locally.
In this repository5 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- buildSkill
- commitSkill
- planSkill
- setup-stackSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 23
If nothing on the backlog is actionable, say so, list what would make the top items ready, and stop.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/pm: the outer loop (orchestrator)
Input ($ARGUMENTS): optionally --dispatch.
1. Triage
Spawn the pm agent for a triage. Its deliverables: backlog health (stale items, blocked items, items missing acceptance criteria), the ONE most valuable ready-or-readyable item, and a build-ready spec for it.
2. Report
Relay to the user: the pick and why, the spec, plus a short health note (e.g. "3 items lack acceptance criteria: #12, #15, #18").
3. Hand off
- Without
--dispatch(interactive): ask the user whether to run/buildon the spec now. If yes, invoke the build skill with the spec. - With
--dispatch(autonomous/CI): post the spec as a comment on the chosen issue, then add theagentlabel, which triggers.github/workflows/agent-task.ymlto run the inner loop headlessly. Independent tasks run in parallel: after dispatching, repeat the triage while ready items remain that are independent of everything now in flight (each pm pass picks knowing what the previous one dispatched). Two limits are real, and only these two: a dispatch must not depend on or touch the same code as in-flight work, and the human merge gate paces everything: when unreviewed agent PRs are piling up, stop dispatching rather than adding more.
If nothing on the backlog is actionable, say so, list what would make the top items ready, and stop. An empty dispatch is a fine outcome.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

