Make your coding agent work like a senior teammate instead of an eager intern. /sonu:build writes every change in your style and test-first, then hands it back with the riskiest parts flagged before you review. /sonu:ship takes it from working tree to merged PR on its own — collecting every AI reviewer bot enabled on…
★not rated 3 3d agoA
tokens not measured
copy · 89%MIT
Enforces an 8-phase SDLC workflow for Claude Code: Plan, Analyze, Design, Build, Test, Deploy, Support, Docs. Keeps the human in the lead through explicit phase gates, enforces plan-before-code and surgical-edit discipline, and degrades gracefully when Git/ticketing/observability integrations are absent.
★not rated 3 4mo agoA
tokens not measured
originalMIT
Read-only check for pre-commit setup state. Detects missing pre-commit binary, uninstalled .git/hooks/pre-commit, and unreachable Docker daemon when the config requires it. Walks you through approval-gated installs via SessionStart and PostToolUse:EnterWorktree hooks.
★not rated 3 27d agoA
tokens not measured
originalMIT
MagicTeX — a LaTeX editor for AI agents: live PDF preview, PDF comments that drive edits, a Visual (WYSIWYG) mode, and git-tracked history, with slash commands to run it hands-off.
A carefully crafted Claude Code statusline written in bash. Shows model, effort level, git branch (state-aware color palette), context window with absolute-token traffic light, rate limits, and optional cost.
★not rated 2 4mo agoA
tokens not measured
originalMIT
Tells you what your branch actually changed — what was done, why, and what deserves a second look, instead of the diff — then files the fixes into the commits they belong to.
★not rated 2 13d agoA
tokens not measured
originalMIT
Per-agent diff review for Claude Code parallel subagent workflows. Each subagent worktree captured automatically, tabbed TUI to approve, revert, or merge each one independently.
★not rated 2 3mo agoA
tokens not measured
originalMIT
Claude Code plugin: Claude orchestrates, Executor (grok or codex) implements tasks in isolated git worktrees, codex reviews read-only, Claude judges/reworks/merges. Review is never bypassed — including Claude's own fallback code.
★not rated 2 25d agoA
tokens not measured
originalMIT
Generic, plugin-based PR review for GitHub, Azure DevOps, and GitLab. Reviews with parallel passes built from synced skill packs (stack-aware, zero built-in reviewers) in a single session (Copilot CLI or Claude Code), with optional Codex second opinion.
Catches the garbage your AI assistant left behind — security shortcuts, secrets buried in git history, code nobody understands. One 0-100 score, 100% local, no LLM in the loop.
★not rated 2 1mo agoA
tokens not measured
AGPL-3.0
Every skrrt skill in one bundle for the skills.sh CLI — git shipping workflow plus structured markdown authoring with auto-fixed lint-clean output. Claude Code installs per bucket instead; see marketplace.json.
★not rated 2 27d agoA
tokens not measured
originalMIT
Automatic post-commit code review using OpenAI Codex CLI. After every git commit, Codex reviews the changes and provides a PASS/FAIL verdict. On FAIL, Claude fixes issues and re-commits. On PASS, prompts to push to GitHub.
★not rated 2 1mo agoA
tokens not measured
originalMIT
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: