Skip to content

Implement From Spec

Builds a spec or ticket end to end, using TDD, typechecks, tests, a review pass, and a commit.

Stars
309,658
Licence
MIT
Files
2 files

What it does

When you ask the agent to implement a spec or a set of tickets, it follows a fixed workflow: if you give a ticket ID it pulls the ticket from your issue tracker and confirms the title, then it hands work to a test-driven development skill, runs typechecks and tests as it goes, calls a code-review skill at the end, and commits the result to your current git branch. Without this skill, the agent would just start coding without this structure.

Great for

  • Developers who already write specs or tickets and want a repeatable build-test-review-commit loop
  • Teams using a connected issue tracker who want ticket references resolved automatically

Skip it if

  • Exploratory prototyping or spikes where you do not want tests, reviews, or commits yet (the workflow forces all of them)
  • Repositories where committing directly to the current branch is unsafe, such as shared main branches with no review process (it commits without asking)

Try asking

  • “Implement ticket ENG-123 from our issue tracker”
  • “Implement the spec in docs/payments-spec.md and commit when the tests pass”
  • “Work through these three tickets in order and commit the result to this branch”
Step-by-step guide: how to use Implement From SpecInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    implement.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip implement.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip implement.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Give a clear spec or ticket ID so the agent does not need to ask for clarification
  • Make sure the 'tdd' and 'code-review' skills are installed, otherwise those steps will not run as described
  • Review the commit on your branch before pushing, since the skill commits without a confirmation step

What's inside

  • agents/openai.yaml
  • SKILL.md

Redistributed unmodified from mattpocock/skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.