#claude-code-hooks

Claude Code / Codex / pi plugin that hands agent steps needing no text output to Jev (TypeSafe's judgment model) — measured p50 ~230 ms and ~$0.02 per 1,000 judgments, with typed escalation back to the LLM

The control plane for your coding agents, however you run them. One checkout of rules, skills, subagents, hooks and stances, projected into Claude Code and Codex with your own primitives: hook-enforced guardrails, usage telemetry, cost postures and model tiering, under any rules library or orchestration layer.

Preserves a long-lived repository's engineering context — an architecture index, an impact map, session records, and the decisions behind them — as markdown committed beside the code, so an agent reads instead of rediscovering. Repository-local, no network, Python stdlib, MIT.

Delegate coding work from Claude Code to a local pi agent — MCP tools with mid-run steer/abort, plus per-project discipline hooks.

A curated list of Claude Code mods, installable as a plugin marketplace: themes, panes, status lines, portraits.

Save a secret: a small window opens that Claude cannot see, you paste there, it lands in 1Password. Read it back: any item in your vault loads into Claude with one fingerprint check, names only, never the values. Keys you paste in the chat are deleted before Claude reads them. Mac, Windows, Linux. MIT.

Claude Code hook that blocks fixes that already failed - fingerprints edits, links them to test outcomes, survives context compaction. 100% of disguised repeat fixes caught, zero false blocks (npm run bench).

Tested guardrail hooks and repo-hygiene skills for Claude Code. Every skill ships with an eval transcript.

One markdown file. Two hooks. Memory that survives Claude Code's context compaction. Zero deps.

Picks the model tier for a delegated Claude Code task before it spawns. Opt-in per repo, reversible in one command, fails safe. v0.1 is a lookup table on purpose.

101 open-source mods for Claude Code: guardrails, status lines, panes and slash commands, plus an in-terminal mod store

A reusable Claude Code agent team: orchestrator + subagents, cost ledger, hooks, and hire/retire/retro skills.

Auto-archive Claude Code sessions to your Obsidian vault as a Karpathy-style LLM wiki. Passive, hook-driven, knowledge converges instead of accumulating.

Claude Code plugin marketplace: the bitranox skill collection (82 skills) plus a self-improve Stop hook. Skills invoke as /bitranox:<skill>.

Your agent stopped. It could have kept going. A Claude Code plugin that reads the conversation when a turn ends, spots work that was announced but never done, and puts the agent back on.

Claude Code plugin that keeps AI attribution and AI writing tells out of commits, PRs, issues and comments.

A clearer Claude Code terminal: boxed and organized you/claude conversations, colour-coded clickable tool rows, a live status band with a timer and task info, a session panel, copy buttons, cards, alerts, readable palettes, and overall transforms Claude Code.

Spec-driven development for React: turn requirements into specifications, architecture plans, implementation, verification, and code review - guided by explicit architectural rules and consistent React conventions

Claude Code UserPromptSubmit hook that routes prompts to the right memory files instead of carpet-bombing context. Stops the 'Claude got dumb' failure mode caused by promiscuous keyword matching.

Free sample of the Claude Code Power Kit — hooks, skills & CLAUDE.md templates written by Claude itself (the €500 / 7-day experiment)

Claude Code PreToolUse hook that denies the pgrep/pkill shapes that kill the agent's own session or spin forever.

Deterministic guard hooks for Claude Code and Codex CLI: ten hooks, 1,785 paired must-fire/must-not-fire tests, zero dependencies, one-command install. Refuses 'done' without proof, piped-away git verdicts, off-scope edits, secret writes and config tampering.

Matt's Claude Code setup as an installable onboarding: one plugin, one command, a short interview, and a blank Claude Code becomes the setup I run across three companies. Every hook self-tested.

Set up Claude Code to help run a small business safely: an interview that writes your CLAUDE.md rules and memory folder, a hook that asks before Claude sends, publishes, pays or deletes, and a Friday weekly review. Claude Code plugin, MIT.

engramx's bi-temporal mistakes + structural code-graph as auto-discoverable Claude Code Skills. Catches recurring mistakes before agents repeat them. Apache 2.0, local SQLite, zero cloud.

2026 AI Agent Loop v2.0: Autonomous Coding with Guardrails & Self-Healing

Terminal tab icons and sounds for Claude Code, so you know which session wants you. Two independent plugins.

Claude Code hooks that measure live context, warn as it grows, and force a detailed handoff note before a long chat gets expensive.

Shared PreToolUse cases for coding-agent guard hooks, a runner, and results per hook and Claude Code version

Claude Code hooks, skills and a playbook for spec-first agentic development with OpenSpec in Godot. Guardrails that stop the agent from weakening tests. Extracted from a mobile game built with no hand-written code.

jevmem saves the decisions, rules and failed approaches from your Claude Code chats to JEVMEM.md in your repo, and brings the relevant ones back next session.