Reduces a bug found in someone else's open-source tool to a minimal reproduction in the ryoppippi-bug-repro org and files the upstream issue. Use when a dependency misbehaves, a drop-in swap breaks an app, or the user asks for a repro or an issue.
Guides execute-inspect-adjust development and t-wada Red-Green-Refactor TDD. Use when the user asks for TDD, a failing test or a regression test, or when a change has an obvious cheap test target in an existing suite. Skip when the test path is unclear,…
Write or review Vitest tests with clear behaviour-focused assertions, appropriate globals/imports, file snapshots for terminal output, and readable fixture setup. Use when adding tests, fixing Vitest failures, or improving TypeScript test quality.
Runs the full PR workflow — creates a feature branch, commits, pushes, and opens the pull request. Use when the user asks to create or open a PR ("create a PR", "push this up and open a PR").
Resolves missing CLI tools. Use when a command is unavailable, a shell reports command not found, or a tool must be run without installing it globally.
Gets an independent second opinion from another coding agent CLI — Codex, Claude Code, Grok, or Cursor — or hands one a task the user names it for. Use when the user asks for one of them, or before a significant approach is settled — not for routine searches,…
Reviews uncommitted changes, a commit, or a branch with another vendor's model or harness — Codex from Claude Code, Claude Code from Codex. Use when the user asks for a code review, or before opening a PR.
Creates a stack of dependent PRs with gh-stack. Use when the user asks to create or open a stacked PR ("create a stacked PR", "open these as a stack").
Delegates a task to another provider or model as a T3 Code child task through `delegate_task`. Use inside T3 Code whenever a skill would otherwise shell out to `codex`, `claude -p`, `grok -p`, or `cursor-agent`.
Guides authoring agent-facing instructions following Anthropic's SKILL.md best practices. Use when adding or editing skills under `agents/skills/`, the global instructions in `agents/AGENTS.md`, or the `agents/shared/` fragments — frontmatter, references, and…
Administrateurs de réseaux et de systèmes informatiques
description
Write idiomatic, functional Nushell — the default language for shell scripts here. Use when writing or reviewing any new script, .nu files, `shell: nu {0}` CI steps, or when porting bash/POSIX shell to Nushell.
Audits all local skills in this repo for skill-creator and Anthropic best-practice compliance. Use to review the whole skill collection — checking descriptions, SKILL.md length, cross-skill duplication, and documentation references — not to author a single…
Reviews React Server/Client Component boundaries against Next.js and React docs. Use when auditing `'use client'` placement or splitting components for proper RSC behaviour.
Review React code for unnecessary useEffect usage based on react.dev guidelines. Analyses code for useEffect anti-patterns and optionally applies fixes. Use when reviewing React code quality or refactoring effects.
Creates atomic git commits and pushes to remote. Use when you want to commit changes and push without creating a PR (e.g., "commit and push my changes" or "push this up").
Manages worktree lifecycle operations through git-wt without replacing an existing linked worktree. Use when creating, switching, listing, renaming, or deleting worktrees.
Spawns parallel task agents to explore a codebase area. Use when researching unfamiliar code, auditing a subsystem, or gathering diverse perspectives before a design decision.
Diagnoses and fixes CI failures with the gh CLI. Use when GitHub Actions checks are red and the failing logs need to be fetched, analysed, and resolved.
Apply TypeScript style rules that prefer checked literal typing with satisfies and as const satisfies over unsafe assertions. Use when writing or reviewing TypeScript or TSX code, test fixtures, typed object literals, mocks, config objects, or table-driven…