implement

Implements work from a spec or tickets using TDD, typechecking, and code review.

1|Updated Jul 23, 2026
One-click install
npx skills add https://github.com/sanjanb/my-agent-harness --skill implement-sanjanb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: implement
Source: https://github.com/sanjanb/my-agent-harness/tree/main/skills/mp-implement
Command: npx skills add https://github.com/sanjanb/my-agent-harness --skill implement-sanjanb

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning a written spec or set of tickets into working, verified code requires discipline: writing tests first, running typechecks, and reviewing the result. This Skill encodes that implementation workflow so the agent follows a consistent engineering process instead of ad-hoc edits. ## Core Features & Use Cases - Spec-Driven Implementation: Builds the work described in a spec or ticket set, using test-driven development at pre-agreed seams. - Continuous Verification: Runs typechecking and single test files regularly during development, then the full test suite once at the end. - Review and Commit: Finishes by running a code review pass over the changes and committing the work to the current branch. - Use Case: You have a ticket describing a new API endpoint. Invoke this Skill to implement it with tests first, verify types and tests pass, review the diff, and commit the finished work. ## Quick Start Implement the work described in the current spec or tickets using TDD, then review and commit it.

Frequently Asked Questions about implement

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I implement a feature from a spec using TDD?

Invoke this Skill with the spec or tickets as input. It writes tests first at pre-agreed seams, implements the code to pass them, runs typechecking and tests regularly, then reviews and commits the finished work.

What workflow does this Skill follow for ticket-based work?

It reads the spec or tickets, applies test-driven development where possible, runs typechecks and individual test files during development, executes the full test suite once at the end, performs a code review, and commits to the current branch.

Does this Skill run the full test suite on every change?

No. It runs typechecking and single test files regularly during implementation, but the full test suite runs only once at the end. This keeps iteration fast while still verifying the complete codebase before finishing.

Can this Skill be invoked automatically by the agent?

No. The Skill sets disable-model-invocation to true and its agent policy disallows implicit invocation, so it only runs when the user explicitly invokes it.

When should I not use a spec-driven implementation workflow?

Avoid it for exploratory spikes or throwaway prototypes where requirements are unclear and tests would be discarded. The workflow assumes a defined spec or tickets and pre-agreed seams for testing.