Install
$ agentstack add skill-twanahc-claude-linear-skill-linear-implement ✓ 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 Used
- ✓ 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
Linear Implement
Overview
Given an implementation plan for a Linear issue, execute it step by step: write code, run tests/checks, and update the Linear issue with progress. Create follow-up issues for any bugs or tech debt discovered along the way.
API Script
bun ~/.claude/skills/linear/scripts/linear-api.ts
Process
1. Pull Latest & Create Worktree
This is mandatory. Never work outside a worktree.
# Pull latest from main
git checkout main && git pull origin main
# Create a worktree for this feature
git worktree add .claude/worktrees/ -b
# Symlink .env.local from the main repo root (required for API keys / Firebase)
ln -s /.env.local .claude/worktrees//.env.local
# Switch into the worktree
cd .claude/worktrees/
Replace ` with the issue identifier in lowercase (e.g., blue-42). Replace ` with the absolute path to the main repository root.
2. Understand the Plan
Read the provided implementation plan carefully. Identify:
- Files to modify or create
- Order of operations (dependencies between steps)
- Tests to write or update
- The issue's team key (from the identifier, e.g., "BLU" from "BLU-42")
3. Use Superpowers Skills (if available)
If superpowers skills are available in the environment, use them to guide implementation:
- brainstorming — before creative work (new features, new components, behavioral changes)
- test-driven-development — before writing implementation code
- systematic-debugging — if you encounter bugs or unexpected behavior
- verification-before-completion — before claiming work is done
- writing-plans — if the plan needs further decomposition
Invoke the relevant skill before proceeding. If no superpowers skills are available, continue normally.
4. Implement
Execute the plan step by step:
- Write code changes using Edit/Write tools
- Follow existing code patterns and conventions in the project
- Keep changes minimal and focused on the issue
- Write tests for new functionality
5. Run Checks
After implementation, run the project's checks. Check package.json for available scripts:
# Common patterns — use what the project has:
bun run check # if available (preferred)
bun run lint # linting
bun run typecheck # type checking
bun test # tests
Fix any errors before proceeding. If checks fail:
- Read the error output carefully
- Fix the root cause (not the symptom)
- Re-run checks until they pass
6. Handle Discoveries
If you find bugs, tech debt, or edge cases while implementing that are outside the scope of the current issue:
source ~/.bashrc && bun ~/.claude/skills/linear/scripts/linear-api.ts create-issue \
--title "Found during : " \
--team \
--priority 3 \
--description "Discovered while implementing . "
Track all created follow-up issues in your summary.
7. Update Linear
Add an implementation summary comment:
source ~/.bashrc && bun ~/.claude/skills/linear/scripts/linear-api.ts add-comment \
"Implementation complete. Changes: . Tests: . Follow-ups: ."
Do NOT change the issue status to Done. The orchestrator or user will decide when to close it. Only update status if explicitly asked.
8. Return Summary
Return to the orchestrator:
- What was implemented: files changed, approach taken
- Tests: what was written, whether they pass
- Checks: whether all project checks pass
- Follow-up issues: any new issues created (with identifiers)
- Blockers: anything that couldn't be completed and why
Key Principles
- Scope discipline: Only implement what the plan says. Don't refactor nearby code, add features, or "improve" things outside the plan.
- Check before claiming done: Always run the project's check/test commands and confirm they pass.
- File follow-ups, don't fix everything: If you find unrelated issues, create Linear issues for them instead of fixing them in this PR.
- Be transparent: If something doesn't work as expected, report it honestly in the summary.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: twanahc
- Source: twanahc/claude-linear-skill
- License: MIT
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.