Stop
01Hook Claude Code
Runs when the agent finishes a response, running an inline shell check. From lkangd/cc-settings-preset.
Hook Claude Code
Runs when the agent finishes a response, running an inline shell check. From lkangd/cc-settings-preset.
Settings file Claude Code
Agent settings declaring 1 hook event (Stop).
Skill Claude CodeCodex
Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.
Plugin Claude Code
Plugin marketplace listing 12 plugins: codebase-architecture-doc-builder, commit-commands, karpathy-guidelines, rocky-pal, package-release.
Settings file Claude Code
agents skils collection.
Plugin Claude Code
macOS cleanup and disk-space recovery workflows using Mole's mo CLI, with dry-run-first safety guidance for cleanup, app uninstall, project artifact purge, installer cleanup, disk analysis, optimization, and health checks.
Skill Claude CodeCodex
This skill should be used for macOS cleanup, disk-space recovery, and Mac health checks with the Mole mo CLI. Use it when users ask to clean a Mac, free storage, fix “disk full” or “System Data” bloat, troubleshoot a slow Mac, uninstall Mac apps completely, remove app leftovers, clean caches/logs/temp files, delete…
Plugin Claude Code
Email workflows via CLI backends, starting with the himalaya CLI for reading, searching, summarizing, composing, replying, forwarding, and managing mail across IMAP/Maildir accounts, with a strict confirm-before-send workflow.
Skill Claude CodeCodex
This skill is for email tasks using the himalaya CLI: checking inbox/unread mail, searching messages, summarizing or reading threads, opening or expanding selected emails, message groups, or topics, handling attachments, composing/replying/forwarding, archiving/moving/deleting/flagging, and reviewing a day's mail for…
Plugin Claude Code
Review-verify-fix workflow in one command: parallel read-only reviewers (cross-model via a configurable runner), main-agent verification, direct fixes, a git-tracked backlog for deferred findings, and an adversarial review-fix loop.
Agent
Read-only reviewer/verifier for the code-review plugin's in-session mode. Executes one prepared angle-prompt file (or verifies candidate findings with the provided verdict ladder) and returns structured output. Never edits files and never delegates.
Command
Adversarial review loop - review, fix, re-review until clean or max rounds.
Command
Review a target (commits, staged, working tree, files), verify findings, fix confirmed issues.
Command
Configure the code-review plugin for this project (runner, concurrency, rounds, backlog).
Plugin Claude Code
Author or refresh a project's ARCHITECTURE.md in the rust-analyzer style: Bird's Eye View, Code Map with Architecture Invariant call-outs, and Cross-Cutting Concerns.
Skill Claude CodeCodex
Author or rewrite the project's ARCHITECTURE.md in the rust-analyzer style — Bird's Eye View, Code Map with per-module Architecture Invariant call-outs, and Cross-Cutting Concerns. Use when the user asks to write, generate, refresh, or review architecture documentation, an "engineering overview", a "tech doc for new…
Plugin Claude Code
Streamline your git workflow with simple commands for committing, pushing, and creating pull requests.
Command
Cleans up all git branches marked as [gone] (branches that have been deleted on the remote but still exist locally), including removing associated worktrees.
Command
Commit, push, and open a PR.
Command
Based on the above changes, create a single git commit. If recent commits are empty, use the standard commitizen commit format.
Plugin Claude Code
Compact the current conversation into a handoff document for another agent to pick up.
Skill Claude CodeCodex
Compact the current conversation into a handoff document for another agent to pick up.
Plugin Claude Code
Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.
Skill Claude CodeCodex
Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.