Ship Shit Dev Skills 161 AI agent skills for development workflows. Works with Claude Code, OpenAI Codex, and Cursor. Catalog: 161 skills · 30 commands · 13 bundles · 174 plugins . Skills are model agnostic playbooks : the harness supplies the model, so no skill names a concrete model — orchestrators speak in capability tiers, and each repo's routing block maps tiers to models. Enforced by scripts/validate skill sync.sh ; standards live in .agents/memory/system/skill standards.md . Directory Structure Path Role Generated fact .agents/ Repository memory, standards, and maintenance skills Tracked .claude/ Claude loader adapters for shared maintenance content Tracked .claude plugin/ Generated Claude marketplace catalog 174 generated plugins .codex/ Codex loader adapters for shared maintenance content Tracked .github/ Issue templates and GitHub Actions workflows Tracked .husky/ Git hook configuration Tracked assets/ Static repository assets Tracked bundles/ Generated marketplace bundle snapshots 13 generated bundles commands/ Claude Code/plugin command adapters 30 command adapters prompts/ Shared prompt resources Tracked resources/ Authoring references and supporting documentation Tracked scripts/ Validation, generation, migration, and audit tooling Tracked skills/ Canonical public Agent Skills sources 161 canonical skills What's Included The Dev Loop : a board driven autonomous workflow that ships software with gh + Claude + Codex. Engineering Workflows : code review, debugging, QA, release, GitHub triage, branch cleanup, production audit, and CI repair. Implementation Specialists : frontend, backend, infrastructure, security, testing, payments, and workspace setup skills. Agent Engineering : context engineering, memory systems, evaluation, multi agent patterns, MCP building, and prompt/tool design. Maintenance Tooling : scripts that validate skills, regenerate bundles, and keep the marketplace catalog aligned with source. The Dev Loop — ship software with gh + Claude