Install
$ agentstack add skill-britt-claude-code-skills-working-on-an-issue ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
Working on an Issue
Announce at start: "I'm using the working-on-an-issue skill to implement this GitHub issue."
Overview
A structured workflow for implementing GitHub issues with verification.
Core Principle: Understand → Plan → Verify → Implement → Verify Again
When to Use
- Developer asks to work on a GitHub issue
- Developer asks to implement an issue
- Developer provides an issue URL or number
Pre-flight Checklist
Before starting:
- [ ] Issue URL or number obtained
- [ ] Repository cloned and on correct branch
- [ ] CLAUDE.md exists (or run
setting-up-a-projectfirst) - [ ] Developer available for questions (or async mode agreed)
The Process
1. Read and Understand the Issue
Get the issue content via:
- GitHub CLI:
gh issue view - Ask user to paste the issue content
- Use GitHub MCP tools if available
Clarify before proceeding:
- If requirements are ambiguous, ask specific questions
- If acceptance criteria are missing, propose them
- If scope is unclear, confirm boundaries
Do not assume or invent requirements not in the issue.
2. Write an Implementation Plan
Use the superpowers:writing-plans skill (if available) or create a brief plan covering:
- What changes are needed
- Which files will be modified/created
- Order of implementation
- Risks or unknowns
Save to docs/plans/issue--plan.md and get developer approval before proceeding.
3. Write a Verification Plan
Use the writing-verification-plans skill to create acceptance tests for the issue.
4. Implement
- Follow TDD practices if
TDD.rules.mdis present (a project-level TDD rules file, if the repo defines one) - Commit after each logical change
- Pause and ask if you hit unexpected complexity
5. Execute Verification
Run the verification plan. Report results using the verification log format from the writing-verification-plans skill.
Absolute Rules
- No assumptions: Ask if anything is unclear
- No scope creep: Only implement what's specified
- Verification required: Task is incomplete until verification passes or developer confirms manual verification
- Blocked = Stop: If blocked, stop and ask the developer — do not guess.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: britt
- Source: britt/claude-code-skills
- License: MIT
- Homepage: https://britt.github.io/claude-code-skills/
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.