agentleFS
Sign inStart free
rakaarwakyrakaarwaky/agents-arwaky/skills/git/issue-worker/SKILL.md

issue-worker

Skill0 starsChanged 3 days agoMITView on GitHub

issue worker: analyze open GitHub issues and take each one from backlog to a PR opened to main — Hermes verifies at HEAD and runs the decision quiz, then 1 issue → 1 Orca worktree → Grok Build implements → Hermes verifies → Hermes pushes branch + opens a (draft) PR to main → close/reopen the issue per result. Never merges. Takes EXISTING issues to PRs; does not create issues (that is /issue-maker's domain). CRUD recipes stay as reference only; the spine is verify → decide → isolate → delegate → verify → PR.

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
  1. Distinctive:

    Read only the winning lane's references.

    line 31

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Issue Worker — issue → PR

End-to-end skill for GitHub issue work: analyze issues to create PRs. The spine is verify at HEAD → decide → 1 issue → 1 Orca worktree → Grok Build implements → Hermes verifies → Hermes pushes branch + opens a (draft) PR to main → close/reopen the issue per result. This file is the lane identification step only; the working method lives in references/hermes/ and references/grok/.

Global Rules

The lane is determined from a physical signal, not by self-identification. Each agent's launcher stamps a fixed attribution env var into every command it runs; the model cannot set these.

if [[ -n "${HERMES_AGENT:-}" ]]; then echo HERMES
elif [[ -n "${GROK_AGENT:-}" ]]; then echo GROK
else echo UNSET; fi
  • HERMES → Hermes lane. Read references/hermes/. Stop.
  • GROK → Grok lane. Read references/grok/. Stop.
  • UNSET → stop and report. Do not guess the lane.

Marker provenance:

  • Hermes re-exports AI_AGENT=hermes-agent and HERMES_AGENT=true on every command.
  • Grok Build runs with GROK_AGENT=1.

Read only the winning lane's references. Do not read the other lane's.

Routing

Hermes lane (Steps 0–3)

StepReferencePurpose
0references/hermes/step-0-triage-and-spawn.mdVerify claims at HEAD per issue; ask the decision quiz; spawn one Grok worktree per still-broken issue and drop the brief
1references/hermes/step-1-wait-and-verify.mdWait for each Grok session; verify commits, code, and re-run the symptom repro in that worktree
2references/hermes/step-2-ship-and-close.mdPush each verified branch, open one draft PR per issue; close verified-on-main issues with evidence; rebase still-open branches
3references/hermes/step-3-final-report.mdFinal report: verdicts, evidence, PRs, closed issues, failed items, worktree teardown

Grok lane (Steps 4–10, run by Grok Build per issue)

StepReferencePurpose
4references/grok/step-4-verify-triage.mdLoad and triage: verify every claim at HEAD before closing.
5references/grok/step-5-validate.mdValidate issue state before deciding — rejects stale/fake/duplicate/already-fixed issues.
6references/grok/step-6-decision-quiz.mdDecision quiz before implementation: collect all open questions, ask in one interactive quiz with concrete recommended options first.
7references/grok/step-7-fanout.mdFan-out setup: 1 issue → 1 Orca worktree → 1 owner; child brief contract.
8references/grok/step-8-build-isolation.mdBuild isolation per worktree.
9references/grok/step-9-waves-verify.mdRun waves, verify children (report files changed, mechanism, verification output, commit SHA).
10references/grok/step-10-rebase.mdRebase still-open worktree branches onto the default tip after PRs land.
1references/grok/step-1-viewing.mdAuxiliary: list, search, view issues (gh + curl).
2references/grok/step-2-managing.mdAuxiliary: label, assign, comment, close/reopen, link to PRs, branch from issue.
3references/grok/step-3-triage-bulk.mdAuxiliary: issue triage workflow and bulk label/close operations.
0references/grok/step-0-setup.mdPrerequisites, auth detection, gh-env.sh helper sourcing.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free