implementing
/implementing
Implement a piece of work, such as a task, test-first where the project has tests, then review and commit it. Use when the user wants work built, or when another skill needs work built.
- Author
- fynnhaupt ↗
- Repository
- skills ↗
- Licence
- MIT
- Last repository change
- 07.11.2026
Assessment and security
- Files in this package
- 2
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 07.11.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Implement a piece of work, such as a task, test-first where the project has tests, then review and commit it. Use when the user wants work built, or when another skill needs work built.
Original author description
Implement a piece of work, such as a task, test-first where the project has tests, then review and commit it. Use when the user wants work built, or when another skill needs work built.
How it works
- Steps
SKILL.md preview
--- name: implementing description: Implement a piece of work, such as a task, test-first where the project has tests, then review and commit it. Use when the user wants work built, or when another skill needs work built. --- # Implementing Build the work in the conversation, such as a **task**, and leave it committed and reviewed on the current branch. Build only what the work asks; anything else you notice goes to the user as a note, not into the code. Ask the user every question, including a confirmation, through your harness's interactive question tool when it has one (e.g. `AskUserQuestion` in Claude Code), with the likely replies as options; show anything they must review, such as an issue body, as text first. Only without such a tool, type the question as text. ## Steps 1. **Read the work.** For a task issue, read it and its parent spec. Then read the code it touches. Done when you can name the modules it changes and the seams to test. 2. **Build it.** If the project has a test runner, drive the code with the `testing` skill; otherwise build it directly. Run the affected tests and the typecheck, where the project has them, as you go. When the work leaves a choice open that changes behaviour, ask the user instead of guessing. Done when everything the work asks is met, such as every acceptance criterion of a task. 3. **Check it.** Run the full test suite and every check the project uses, such as the typecheck, linter, and formatter. A check that also fails on the base was failing before: report it, don't fix it. Done when every other check passes. 4. **Commit** to the current branch, following the project's commit conventions. Done when `git status` is clean. 5. **Review it** with the `reviewing` skill against the branch's base. Fix every broken standard and every Spec finding, and the smells you agree with; list the smells you leave and why. Check and commit the fixes as in steps 3 and 4. Done when every finding is fixed or listed for the user.
Files in this package (2)
- LICENSE 1 KB
- SKILL.md 2 KB
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · fynnhaupt/skills | 90 | MIT |
| azat-io/ai-config | 90 | MIT |
| choam2426/geas | 90 | Apache-2.0 |
Database snapshot: 13.11.2026 · package version 4e7a7d695fa0 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 90task-implementation
Implement the next unfinished task from PLAN.md using strict test-driven development. Writes tests first, then writes just enough code to make them pass, and prepares the result for human review and commit. Operates at the individual task level — one task per invocation. Trigger… - 100kit-create-pr
Commit changes and create a pull request with proper formatting. Use when ready to publish work for review. - 90dev
Build a development task end to end with the dev graph - plan (the advisor approves it), build tests-first, prove every acceptance criterion, then the QA graph - and commit the result, unattended after any up-front questions. Use when the user says "/dev <task>", "build this… - 95implement
Implement a piece of work based on a spec or set of tickets. Use when an agreed spec or ticket set should now be built. Triggers on: implement, build this spec, work these tickets. - 82implement-spec
Execute an approved spec's tasks.md test-first, one task per commit, with review subagents. Use when the user asks to implement or continue a spec. - 81implement
Implements a piece of work from a spec or ticket set. Use when the user says \"implement this spec\", \"work the tickets\", or wants code written from an existing plan.