2.6k
GitHub Stars
3
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill openclaw/skills --skill issue-prioritizer- _meta.json644 B
- README.md498 B
- SKILL.md18.2 KB
Overview
This skill prioritizes GitHub issues by computing an Adjusted Score that combines ROI with solution sanity (Tripping Scale), architectural impact, and actionability. It is read-only: it analyzes issues and recommends quick wins, risky proposals, and actionable bugs without changing repositories. Use it during triage to rank work and match issues to contributor skill levels.
How this skill works
The skill fetches open issues (with optional topic/search/label filters) using the gh CLI, optionally excludes issues already addressed by open PRs, and scores each issue on Difficulty, Importance, Tripping Scale, Architectural Impact, and Actionability. It derives ROI and applies multipliers to produce an Adjusted Score, then categorizes issues (Quick Wins, Critical Bugs, Tripping, Over-Engineered, Not Actionable) and sorts by score. It reports results in human-readable or JSON/Markdown output.
When to use it
- Triaging or ranking repository issues to pick the best work
- Finding quick wins for new contributors or sprints
- Filtering out non-actionable items (questions, duplicates)
- Detecting over-engineered proposals that need simplification
- Matching tasks to contributor skill levels (beginner/intermediate/advanced)
Best practices
- Provide a repository in owner/repo format when invoking the skill
- Use --topic, --search, or --label to focus the fetch to relevant issues
- Run gh auth login beforehand; skill is read-only and requires gh CLI access
- Include --include-with-prs only when you explicitly want issues that already have PRs
- Review Tripping and Architectural flags carefully before approving heavy changes
Example use cases
- Sprint planning: surface high-ROI, low-difficulty issues for the next sprint
- New contributor onboarding: list beginner-level PR-ready tasks
- Maintainer triage: detect feature proposals that propose unnecessary architectural rewrites
- Bug response: prioritize critical bugs by Importance × Difficulty
- Community triage events: filter and publish quick wins to the community
FAQ
No. It is read-only and only analyzes and ranks issues. All changes must be made by users.
What happens if gh is not authenticated or rate limited?
The skill will report auth errors and suggest running gh auth login, or advise reducing the --limit if rate limited.