skills.rest
fx
fxCOMMUNITY
Explore all skills by this author →
GitHub

coder

Implements code changes with tests following a supplied scope and project conventions.

Updated Nov 2, 2025
Search Tags:#testing#pull request#bug fix#github#sdlc#code implementation#scope discipline
One-click install
npx skills add https://github.com/fx/skills --skill coder-fx
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: coder
Source: https://github.com/fx/skills/tree/main/skills/coder
Command: npx skills add https://github.com/fx/skills --skill coder-fx

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? It turns a defined scope or issue into committed, tested code changes while preventing scope creep, so implementation stays aligned with what was actually requested. ## Core Features & Use Cases - Scoped Implementation: Implements features and bug fixes strictly within a Scope Brief, stopping to report when work exceeds the request's framing. - Dual Invocation Modes: Runs implementation-only when called by the SDLC workflow, or owns the full lifecycle (review, PR creation, feedback resolution) when invoked directly. - Test Discipline: Enforces a no-skipped-tests policy, requiring failing tests to be fixed, replaced, refactored, or removed rather than bypassed. - Use Case: A user says "just fix the login bug" — the skill writes a failing test first, implements the minimal fix, runs the test suite, and commits, without expanding into unrelated refactors. ## Quick Start Ask the agent to use the coder skill to implement the described feature or bug fix with tests and commit the changes.

Frequently Asked Questions about coder

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

FAQPage Schema
How do I implement a bug fix with tests using an AI coding agent?▼

Invoke the coder skill with a description of the bug. It writes a failing test first, implements the minimal fix within the request's scope, runs the test suite, and commits the changes following project conventions.

What is the difference between SDLC invocation and direct invocation of the coder skill?▼

Under the SDLC workflow, the coder is implementation-only: it writes code, runs tests, and commits, stopping before PR creation. When invoked directly, it owns the full lifecycle including local review, PR creation, reviewer settlement, and feedback resolution.

Does the coder skill create pull requests automatically?▼

Only when invoked directly by the user. When called as a sub-agent by the SDLC workflow, it stops after implementing and committing, and the orchestrator delegates PR creation to the pr-preparer skill.

Can I skip failing tests that are hard to fix?▼

No. The skill forbids test.skip, it.skip, and describe.skip. Failing tests must be fixed, replaced, refactored, or removed, and missing test infrastructure such as fixtures or mocks must be set up instead.

What happens when the work exceeds the original request scope?▼

The coder stops and reports what it found, why it exceeds the request, and the cheapest path forward, offering the narrow option first. It still delivers everything unambiguously in scope before stopping.

Related Skills

Explore More Software Engineering Skills→
avibebuilderavibebuilder
113

give-plan

Create a reviewable implementation plan from repository files before coding.

Official
Intermediate
QredenceQredence
2

babysit-pr

Poll GitHub PR CI checks, reviews, and mergeability until merge or close.

Official
Advanced
fxfx
8

resolve-pr-feedback

Detect unresolved Copilot and CodeRabbit threads on GitHub pull requests and invoke resolver skills.

Community
Advanced
ahostbrahostbr

max-parallel

Decompose large tasks into dependent parallel waves using subagents.

Community
Advanced