do-work

Plan, implement, validate with typecheck and tests, and commit code changes.

1|Updated Mar 15, 2026
One-click install
npx skills add https://github.com/mappa-ai/skills --skill do-work-mappa-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: do-work
Source: https://github.com/mappa-ai/skills/tree/main/utils-skills/do-work
Command: npx skills add https://github.com/mappa-ai/skills --skill do-work-mappa-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Execute a complete unit of work end-to-end: plan, implement, validate with typecheck and tests, and commit to the repository, enabling teams to ship scoped changes efficiently.

Core Features & Use Cases

  • End-to-end workflow: from understanding to commit, with optional planning and tracer-bullet testing for backend tasks.
  • Dual-path implementation: structured backend TDD style and straightforward frontend changes.
  • Validation loop: iterative typechecking and testing until green, followed by a formal commit.

Quick Start

Give a clear task or plan reference and let the agent plan, implement, validate with typecheck and tests, and commit the change.

Frequently Asked Questions about do-work

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

FAQPage Schema
How do I plan, implement, and commit a complete feature end-to-end?

To plan, implement, and commit a feature end-to-end, you provide a clear task or plan reference. The workflow then understands the task, optionally plans, implements with tracer-bullet tests, runs validations, and performs a final commit.

What is the best way to validate code changes before committing them to the repository?

The best way to validate code changes before committing is running an iterative validation loop. This process enforces typechecking and testing until all checks pass green, ensuring code integrity before the formal commit.

Can I use this workflow for both frontend and backend codebases?

Yes, you can use this workflow for both frontend and backend codebases. It supports dual-path implementation: structured backend test-driven development with tracer-bullet tests and straightforward direct changes for frontend tasks.

Does end-to-end work execution require a separate planning phase?

End-to-end work execution does not strictly require a separate planning phase. The workflow includes optional planning, allowing you to either provide a direct task reference or let the agent generate a plan before implementation and validation.

How do tracer-bullet tests work during feature implementation?

Tracer-bullet tests work during feature implementation by providing a structured test-driven development path for backend tasks. They validate the implementation iteratively alongside typechecks until all tests pass before committing.

Why do I need to run typechecks and tests iteratively until green before committing?

You need to run typechecks and tests iteratively until green to enforce a disciplined validation loop. This ensures the implemented unit of work is fully deterministic and error-free before performing the final tracked commit.