hooks

Put this Claude Code session on a full-duplex voice call. You talk while it codes, interrupt it at any time, and hear results read back. A spoken "yes" never approves anything: permission prompts stay on your keyboard. macOS only.

An in-process `tool.call` mod that denies Bash git commands which silently discard uncommitted work. It is the mod form of `hooks/git-guard.sh` and replaces it once cut over, saving the shell forks that script cost on every Bash call (about 34 ms per call, against under a microsecond here).

compound is a Claude Code **mod** that makes each session start from what your earlier sessions built and learned. A mod is a plugin made of hooks: code that runs when you submit a prompt, when Claude calls a tool, and when Claude is about to stop.

A Claude Code mod that brings a second model into your chat when the first one gets tired, stuck or sloppy.

Claude Code plugin providing an autonomous development workflow (brainstorm, auto, iterate, audit, review, ship) with a prd.json sprint system, subagents, and sprint/typecheck/safety hooks.

A Claude Code mod that draws a live forecast of the context window above the prompt, showing fill percentage, the last turn's addition, and a colored bar of what fills the window.

A Claude Code mod that shows which files each agent has in its context, and how much of each.

A per-session mod manager for Claude Code. A toolbar above the prompt lets you switch any locally installed mod on or off for the current session.

Probe mod for the reins project: a Claude Code plugin that registers tool.call hooks to study how the mod engine treats slow, failing or dialog-opening hooks and where mods sit against settings hooks.

A pixel-art companion drawn in the band above the Claude Code prompt that acts out the tool in use, reacts with one of 15 emotions when the answer arrives, and shows each running subagent as a mini robot.

A Claude Code mod that opens a terminal in the worktree you or your agents are working in.

Discover Agent Skills under .agents/skills (project) and ~/.agents/skills (user), expose them as slash commands, list them in prompt.context, and optionally intercept the Skill tool so names from those trees load without living under .claude/skills.