# Linear Implement > Execute an implementation plan for a single Linear issue — write code, run checks, and update Linear with results. - **Type:** Skill - **Install:** `agentstack add skill-twanahc-claude-linear-skill-linear-implement` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [twanahc](https://agentstack.voostack.com/s/twanahc) - **Installs:** 0 - **Category:** [Productivity](https://agentstack.voostack.com/c/productivity) - **Latest version:** 0.1.0 - **License:** MIT - **Upstream author:** [twanahc](https://github.com/twanahc) - **Source:** https://github.com/twanahc/claude-linear-skill/tree/master/linear-implement ## Install ```sh agentstack add skill-twanahc-claude-linear-skill-linear-implement ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## 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 ```bash bun ~/.claude/skills/linear/scripts/linear-api.ts ``` ## Process ### 1. Pull Latest & Create Worktree **This is mandatory. Never work outside a worktree.** ```bash # 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: ```bash # 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: ```bash 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: ```bash 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](https://github.com/twanahc) - **Source:** [twanahc/claude-linear-skill](https://github.com/twanahc/claude-linear-skill) - **License:** MIT Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** yes - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/skill-twanahc-claude-linear-skill-linear-implement - Seller: https://agentstack.voostack.com/s/twanahc - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.