Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/drjkl-issue-triage
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Shell scripts
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/issue-triage
What's inside
- Skill name
issue-triage- Description
- Triages GitHub issue backlogs systematically. Categorizes, verifies relevance, requests context, escalates, and suggests repo improvements. Use when processing issues, cleaning up backlogs, or asked to triage.
- License
- MIT
- Instructions
- 154 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (14)
- SKILL.md
- categories.md · reference/
- decision-tree.md · reference/
- relevance-checks.md · reference/
- skip-patterns.md · reference/
- transfer-targets.md · reference/
- triage-save.sh · scripts/
- triage-status.sh · scripts/
- escalation-format.md · templates/
- responses.md · templates/
- cross-machine-state.md · workflows/
- escalation.md · workflows/
- self-evolution.md · workflows/
- triage-session.md · workflows/
SKILL.md (first lines)
# Issue Triage Systematically process GitHub issue backlogs with intelligent categorization, relevance verification, and self-evolving improvements. ## Quick Start ``` 1. Load repo context (labels, templates, docs) 2. Check cross-machine state (if available) 3. Fetch batch of oldest untriaged issues (skip bots, skip in-flight PRs) 4. For each issue: verify → categorize → act → log 5. Report summary + escalations 6. Save state and suggest systemic improvements ``` ## Contents | Document | Purpose | | ---------------------------------------------------------------------- | --------------------------------------------- | | [SKILL.md](./SKILL.md) | This file - overview and quick reference | | [reference/decision-tree.md](./reference/decision-tree.md) | Full decision flowchart for issue handling | | [reference/categories.md](./reference/categories.md) | Issue categories, labels, and signals | | [reference/relevance-checks.md](./reference/relevance-checks.md) | How to verify if issues are still valid | | [reference/transfer-targets.md](./reference/transfer-targets.md) | Where to transfer misplaced issues | | [reference/skip-patterns.md](./reference/skip-patterns.md) | Issues to skip during triage | | [workflows/triage-session.md](./workflows/triage-session.md) | Step-by-step triage session workflow | | [workflows/self-evolution.md](./workflows/self-evolution.md) | Pattern detection and improvement suggestions | | [workflows/escalation.md](./workflows/escalation.md) | When and how to escalate to humans | | [workflows/cross-machine-state.md](./workflows/cross-machine-state.md) | Track triage progress across sessions | | [templates/responses.md](./templates/responses.md) | Canned responses for common situations | | [templates/escalation-format.md](./templates/escalation-format.md) | Format for escalation reports | ## Core Principles 1. **Process oldest to newest** — Start with the oldest untriaged issues 2. **Verify before acting** — Check if code/feature still exists 3. **Use existing semantics** — Use repo's existing labels; verify before using 4. **Transfer, don't close** — Misplaced issues go to the right repo 5. **Escalate uncertainty** — Low confidence → ask human 6. **Learn from patterns** — Suggest docs/template/label improvements 7. **Be idempotent** — Skip already-processed issues, don't duplicate comments ## Issue Decision Quick Reference ``` Security issue? → Escalate immediately, no public comment Bot-generated? → Skip (CodeRabbit, Dosu auto-issues) Has in-flight PR? → Skip (note PR exists) Wrong repo? → Transfer (don't close) Old version reported? → Request version verification Incomprehensible? → Request clarification Code doesn't exist? → Close with explanation Duplicate? → Close with reference Already fixed? → Close with reference to fix Working as designed? → Close as wontfix with explanation Workaround exists? → Keep open if underlying bug valid Clear category? → Apply label Enough info? → Ready for prioritization Uncertain? → Escalate ``` ## Skip Patterns Skip these issues entirely: - **CodeRabbit auto-issues**: author `coderabbitai[bot]` - **Issues with in-flight PRs**: Note the PR, skip triage - **Recently active with maintainers**: Being worked on - **Already assigned**: Someone owns it ## Version Staleness Request verification when reported version is >3 minor versions old: - Use comment patterns: "Version check:" and "This was reported on v" - Reference specific PRs that may have fixed the issue ## Transfer vs Close **Transfer** when issue belongs elsewhere: ```bash gh issue transfer <number> <target-repo> ``` **Close as wontfix** when: - Working as designed - User misunderstanding - Workaround IS the permanent solution **Keep open** when: - Workaround is temporary - Underlying bug still exists - API improvement still needed ## Confidence Thresholds | Confidence | Action | | ---------- | ------------------------------- | | >80% | Act automatically | | 50-80% | Act but flag for review | | <50% | Escalate, suggest but don't act | ## Dosu Bot Integration If Dosu has commented: 1. Parse Dosu's analysis as starting context 2. Use linked PRs/issues for cross-referencing 3. Don't duplicate questions Dosu already asked 4. Validate Dosu's code suggestions against current codebase (can be outdated) ## Key Commands Before starting, gather context: ```bash
From DrJKL/ticket-to-pr-pipeline, MIT.
More from DrJKL/ticket-to-pr-pipeline
- 0.414score
quality-checks
DrJKL
Runs quality gates (format, lint, typecheck) via single script.
SkillDFreshness: DClaude Code +9 - 0.409score
addressing-review-comments
DrJKL
Checks out a PR branch via worktree, implements code review feedback as individual commits linking to each comment URL, runs quality gates, and hands off to…
SkillDFreshness: DClaude Code +9 - 0.409score
notion-personal
DrJKL
Access personal/life Notion workspace shared with girlfriend.
SkillDFreshness: DClaude Code +9 - 0.409score
resolving-review-comments
DrJKL
Resolves GitHub PR review threads after commits address feedback.
SkillDFreshness: DClaude Code +9 - 0.409score
reviewing-external-prs
DrJKL
Review external GitHub PRs with human approval for each comment.
SkillDFreshness: DClaude Code +9 - 0.409score
slack-context-fetcher
DrJKL
Fetches Slack thread content using a Slack bot for ticket context enrichment.
SkillDFreshness: DClaude Code +9 - 0.409score
coderabbit-intake
DrJKL
Fast-track CodeRabbit issues to PRs.
SkillDFreshness: DClaude Code +9 - 0.409score
ci-checker
DrJKL
Checks GitHub CI status and guides fixing failures.
SkillDFreshness: DClaude Code +9 - 0.408score
review-orchestrator
DrJKL
Dispatch review subagents, compile feedback, triage for human decision.
SkillDFreshness: DClaude Code +9 - 0.406score
coderabbit-reviewer
DrJKL
Integrates CodeRabbit for automated PR code review.
SkillDFreshness: DClaude Code +9 - 0.406score
pr-split-advisor
DrJKL
Analyzes implementation plans and recommends PR splitting strategy.
SkillDFreshness: DClaude Code +9 - 0.406score
metrics-dashboard
DrJKL
Tracks pipeline velocity, success rates, and cycle times.
SkillDFreshness: DClaude Code +9
Signals
- Stars
- 0
- Forks
- 1
- Open issues
- 0
- Last commit (this folder)
- Feb 3, 2026 (8 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/DrJKL/ticket-to-pr-pipeline
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from DrJKL
- 0.414score
quality-checks
DrJKL
Runs quality gates (format, lint, typecheck) via single script.
SkillDFreshness: DClaude Code +9 - 0.409score
reviewing-external-prs
DrJKL
Review external GitHub PRs with human approval for each comment.
SkillDFreshness: DClaude Code +9 - 0.409score
ci-checker
DrJKL
Checks GitHub CI status and guides fixing failures.
SkillDFreshness: DClaude Code +9 - 0.409score
notion-personal
DrJKL
Access personal/life Notion workspace shared with girlfriend.
SkillDFreshness: DClaude Code +9 - 0.409score
resolving-review-comments
DrJKL
Resolves GitHub PR review threads after commits address feedback.
SkillDFreshness: DClaude Code +9 - 0.409score
slack-context-fetcher
DrJKL
Fetches Slack thread content using a Slack bot for ticket context enrichment.
SkillDFreshness: DClaude Code +9
Cite this page
issue-triage skill. Agent Mart. https://agentmart.dev/skills/drjkl-issue-triage (last updated 2026-10-24)