development-skills

Compare Claude Code and Codex on the same real code-change task with isolated worktrees, identical gates, transcripts, time, and cost.

Align AGENTS.md, rules, README, plans, and chronicles with the codebase; use --clean to archive obsolete task docs into ATLAS.md.

Research current best practices and turn the evidence into a clear recommendation. Use for state-of-the-art, evidence-based, optimal-approach, comparison, or pros-and-cons questions in any field.

Clarify an ambiguous or consequential change before planning or implementation when more than one sound approach exists.

Add CHANGELOG.md entries, derive them from commits, or cut a Keep a Changelog and SemVer release.

Create a user-requested Conventional Commit from the staged changes.

Define or implement regression proof for a project or business flow. Use for test strategy, black-box or deep-integration tests, business KPIs and thresholds, test audits, missing tests, or test implementation.

Use when the user asks to compare a plugin or skill against HEAD, run regression evals, or check behavioral regressions before a commit.

Use when the user asks to hand off the current session to a new chat, transfer context, or runs /handoff. Creates a self-contained handoff document in the OS temp directory (not the repo) and prints its absolute path.

Record or apply development-skills plugin feedback; runs on /plugin-feedback. Mode via argument: produce (factual report of plugin interactions this conversation) | ingest <report-path> (apply only evidence-backed simplifications).

Resolve an active merge or rebase conflict, including safe renumbering of colliding plans and chronicles.

Roast code with cynical, aggressive humor while preserving a factual staff review. Use for roast, brutally honest critique, or tear-apart-my-code requests; supports --fix for selected CRITICAL and HIGH findings.

Run a factual, staff-level review of a local branch, diff, repo, directory, or file. Use for code review, deep review, review-my-changes, or staff-review requests.

Read at session start to route development work through the direct or full development loop on Claude Code and Codex.

Use when any coding, development, analysis, debugging, or code-related task is detected. Triggers on: implementing features, fixing bugs, refactoring code, reviewing diffs, investigating errors, evaluating approaches, or making architecture decisions.

Use when user wants to reduce noise, verbosity, or redundancy in markdown or text files. Use when user says distill, compress, clean up, tighten, denoise, reduce entropy, improve signal-to-noise ratio, or make text more concise. Accepts a file path, a directory (distills all .md/.txt files in it), or no argument (distills all .md/.txt files in the current directory).

Python development. Use for Python, FastAPI, Pydantic, asyncpg, pytest, pandas, SQLAlchemy.

Use when fixing bugs, investigating errors, debugging failures, or diagnosing unexpected behavior.

Use when user wants to update .pre-commit-config.yaml hooks to their latest versions from GitHub

Use when user wants to update a pip requirements .in file (requirements.in or requirements-dev.in) with the latest PyPI versions while preserving each line's version-specifier pattern.

Frontend development. Use for React, Next.js, Raycast extensions, Vite-based frontend projects.

Java development. Use for Java, Spring Boot, Maven, Gradle, JPA, Hibernate.

Swift development. Use for Swift, SwiftUI, UIKit, Vapor, SPM, XCTest, Combine.

TypeScript development. Use for TypeScript, Node.js, Express, Fastify, Zod, vitest, jest. Backend, CLI, libraries only — no frontend frameworks.