Template that can be copied into any project with some base subagents and skills to execute on work
This repository also configures its own agents. See what agent-coding-setup tells them →
michaelcjoseph/agent-coding-setup
Agent Claude Code
Reviews changes for system-level architectural concerns — module boundaries, resource leaks, startup/shutdown correctness, error propagation, and centralization violations.
michaelcjoseph/agent-coding-setup
Agent Claude Code
Reviews code changes for bugs, type-safety issues, security problems, and project convention violations. Read-only.
michaelcjoseph/agent-coding-setup
Agent Claude Code
Checks code for dead code, over-abstraction, duplication, and unnecessary complexity. Read-only — reports findings only.
michaelcjoseph/agent-coding-setup
Agent Claude Code
Updates CLAUDE.md and project docs after structural changes — new modules, commands, agents, env vars, scripts, or dependencies.
michaelcjoseph/agent-coding-setup
Agent Claude Code
Generates a changelog from recent git history, or a commit message from staged changes.
michaelcjoseph/agent-coding-setup
Agent Claude Code
Audits changes for hardcoded secrets, PII exposure, path traversal, injection risks, and unsafe shell/subprocess usage. Read-only.
michaelcjoseph/agent-coding-setup
Agent Claude Code
Writes and runs tests for code changes using the project's existing test framework; bootstraps test infrastructure if missing.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: