issue-maker
Skill0 starsChanged 3 days agoMITView on GitHub
One run covers one repo. It reads the spec-document states, proves which rows are real work and which are fiction, and opens a modular issue per finding. Direction: repo → issues only. The run outputs GitHub issues only. It never edits source code, never creates a branch, never opens a PR. Every issue it opens carries one closing condition: the check whose fulfillment advances that issue's row to Done in its source document. Rules that apply to every issue-maker run.…
1 other project uses a version of this.
Adopted by
Most recently taken up by
- pollinations/pollinationstheir own version, first seen Oct 8, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- pollinations/pollinations★ 5.3ksame name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository75 other files it gives its agents.
- AGENTS.mdAGENTS.md
- skills/skill-authoring/tools/AGENTS.mdAGENTS.md
- aes-docsSkill
- aes-fixSkill
- aes-layerSkill
- aes-migrationSkill
- android-emulator-linux-gpsSkill
- computer-useSkill
- flatpak-chromium-automationSkill
- qwen-webSkill
- grok-buildSkill
- hermes-agentSkill
- omnirouteSkill
- orcaSkill
- account-managerSkill
- brave-setupSkill
- password-managerSkill
- codegraphSkill
- deep-cleanSkill
- dogfoodSkill
- reviewSkill
- systematic-debuggingSkill
- blender-arwakySkill
- comfyuiSkill
- llm-writingSkill
- vision-arwakySkill
- playwright-toolingSkill
- setup-ci-quality-gatesSkill
- statuslineSkill
- systemd-headless-serviceSkill
- xdg-spec-complianceSkill
- figcraft-builderSkill
- figcraft-iconsSkill
- figcraft-motionSkill
- figcraft-prototyperSkill
- figcraft-reviewSkill
- figcraftSkill
- figcraft-svgSkill
- figcraft-typographySkill
- github-authSkill
- issue-workerSkill
- mergifySkill
- pr-babysitSkill
- releasingSkill
- repo-managerSkill
- ascii-artSkill
- ascii-videoSkill
- gif-searchSkill
- media-acquisitionSkill
- docxSkill
- emailSkill
- google-workspaceSkill
- notionSkill
- pdfSkill
- pptxSkill
- xlsxSkill
- competitor-news-monitorSkill
- product-price-monitorSkill
- weekly-review-planningSkill
- arxivSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
Zero trust: Never trusts any doc claim — Done rows are re-verified against live source and test evidence.
- Very specific: line 17
One closing condition per issue: Every issue carries exactly one ## Closing condition whose fulfillment advances that issue's row to Done …
- Very specific: line 19
Scripts: All scripts live under scripts/, are stdlib-only Python (no uv sync, no third-party deps), and are CI-gated.
- Distinctive: line 42
Read only the winning lane's references.
≈ “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.
Files
name: issue-maker
description: "repo → issues: zero-trust issue maker that maps AES spec documents (PRD, ROADMAP, BACKLOG, FRD, DESIGN, plus module and package feature docs) into modular GitHub tracking issues. Done rows are re-verified against live source and test evidence. Every issue carries a closing-condition that advances its ROW and all FR IDs to Done. Never opens a PR — issues only. Hermes runs the audit driver via terminal (stdlib scripts, no subagent); no source edits, no branches, no PRs — nothing to delegate to Grok."
when-to-use: /issue-maker — map incomplete tasks to GitHub issues (repo → issues), create tracking issues from ROADMAP or BACKLOG, verify Done claims against code, ensure every issue carries a closing-condition that advances the roadmap state to Done. Outputs GitHub issues only; never edits source, never creates a branch, never opens a PR.
argument-hint: <repo-root> [--discovery <json>] [--scan-out <json>] [--classify-out <json>] [--trust-mode <no-trust|trust>] [--dry-run] [--ledger <path>] [--report <path>] [--legacy]---
Issue Maker — repo → issues: zero-trust spec audit → modular GitHub tracking issues with closing conditions
One run covers one repo. It reads the spec-document states, proves which rows are real work and which are fiction, and opens a modular issue per finding. Direction: repo → issues only. The run outputs GitHub issues only. It never edits source code, never creates a branch, never opens a PR. Every issue it opens carries one closing condition: the check whose fulfillment advances that issue's row to Done in its source document.
Global Rules
Rules that apply to every issue-maker run. Scripts and reference files cite these; they never restate them.
- Zero trust: Never trusts any doc claim — Done rows are re-verified against live source and test evidence.
- Issues only: The run outputs GitHub issues only. Never edits source code, never creates a branch, never opens a PR.
- One closing condition per issue: Every issue carries exactly one ## Closing condition whose fulfillment advances that issue's row to Done in its source document. No issue is counted as created until its closing condition is present in the body.
- Invariant: Fixing ALL open issues equals ALL FR states = Done. Every row from every in-scope doc appears in the ledger with a class.
- Scripts: All scripts live under scripts/, are stdlib-only Python (no uv sync, no third-party deps), and are CI-gated.
- Completion gate: validate.py exits 0 against the ledger (and the run report when present): python3 <this-skill>/scripts/validate.py --repo-root <repo> --ledger <repo>/docs/task-tracker-ledger.jsonl --report <repo>/docs/task-tracker-report.json (artifact filenames unchanged by the skill rename).
- Offline vs --gh: Without --gh, the open-issue liveness criterion is reported deferred-to-gh — exit 0 does not imply the full GitHub-facing contract held. --gh confirms each ledger issue exists and is OPEN; a 403/429 rate limit is reported, never a hard failure.
Routing
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/ (Steps 0–2). Stop.
- GROK → Grok lane. Read references/grok/ (auxiliary recipes, Steps 0–6). 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. The Grok lane is auxiliary only — issue-maker itself never enters it; a coding task launched from an issue-maker finding is a different skill's run.
Hermes lane (Steps 0–2)
| Step | Reference | Purpose |
| 0 | references/hermes/step-0-plan-and-authorize.md | Scope decision (trust mode, ledger/report paths, --gh), GitHub auth check, Hermes lane entry via run.py |
| 1 | references/hermes/step-1-wait-and-verify.md | Wait for run.py; verify gate — validate.py exit 0, issues opened vs ledger rows, issues-only invariant |
| 2 | references/hermes/step-2-final-report.md | Final report: rows, classes, issues opened (or dry-run deferred), closing conditions, --gh liveness, invariant result |
Grok lane (auxiliary recipes)
| Step | Reference | Purpose |
| 0 | references/grok/step-0-discovery-scope.md | Spec-doc scope discovery, the three state-source layouts, and the feature/FR model. |
| 1 | references/grok/step-1-discover-features.md | Enumerate feature directories and root spec docs into a discovery table. |
| 2 | references/grok/step-2-scan.md | Run scan.py (default) or the legacy read-only scanner subagent per feature (--legacy). |
| 3 | references/grok/step-3-classify.md | Trust modes, Next Action handling, zero-trust policy, state vocabulary, classification, evidence, roll-up reconciliation; run classify.py. |
| 4 | references/grok/step-4-verify.md | Read-only verification of every row; verdicts. |
| 5 | references/grok/step-5-open-issues.md | One modular issue per row: title, body, labels, dry-run. |
| 6 | references/grok/step-6-ledger-report.md | Ledger + two-phase run report, scripts, personas, completion criteria, and the validate.py CI gate. |
Note: the scanner-subagent mentions in references/grok/step-2-scan.md are rollback-escape --legacy recipes only; the default path runs the deterministic scan.py script (stdlib, no subagent).
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.

