Write / Edit / MultiEdit tools for AI agent harnesses — ledger-anchored, atomic, with structured fuzzy diagnostics

Comment checker hook extension for oh-my-pi (omp). Runs @code-yeongyu/comment-checker after write, edit, multiedit, apply_patch, hashline, patch, and replace, then self-heals via omp's session_compact + TUI notifications.

Mnemoscope command-line entrypoints, including the Claude Code PostToolUse hook that auto-journals Write/Edit/MultiEdit operations.

Pluggable write-time validation framework for AI coding agents (@webpieces/ai-hook-rules). Claude Code PreToolUse + openclaw before_tool_call adapters share one rule engine.

A Vue 3 batch edit modal component for bulk editing records

Remnic memory plugin for Claude Code — hooks, skills, MCP integration

FlowRail PreToolUse hook dispatcher for Claude Code (POC). Pre-write secret scan + LLM-backed code verifier; pre-bash supply-chain check via the FlowRail MCP server.

Claude Code session + file-change capture adapter for Meterbility

Capture Claude Code agent activity as Veritio evidence via hooks, with a local file sink or ingest POST.

One-shot FlowRail installer for Claude Code. Wires PreToolUse hooks, MCP server config, and skill markdown into a tester's repo, then probes the FlowRail server with their bearer.

WakaTime plugin for OpenCode - Track AI coding activity and time spent

Claude-Code-style agent primitives for the Vercel AI SDK: FS/bash tools, an <env> system-context block, a subagent-as-tool factory, and .claude/ skills/agents loading.

> ⚠️ **This project is under active development.** APIs and features may change without notice.

Xyne CLI - A powerful AI assistant in your terminal with file operations, bash mode, drag-and-drop support, and multi-provider AI integration

OpenCode plugin that converts raw tool-call markup from local Ollama models into standard OpenAI tool_calls, plus line-ending normalization for file edits.

AI Traffic Control — inspect and govern AI prompts in Claude Code

Record a decision once and Claude Code stops re-deciding (and contradicting) it later. A local decision ledger for Claude Code sessions — free, open source, runs entirely on your machine.

A comprehension checkpoint for AI-assisted development. Verifies you understand the code Claude Code just wrote before you keep building.

Pre-execution gate that asks AI agents to produce an Understanding Report before acting

거북이코드 — 계획↔구현↔검증 게이트. Claude Code PreToolUse hook으로 구현 전 계획 케이스 누락·시나리오 미지정을 차단하고, 응답 종료(Stop) 시 scope 사후판정(축A 파급반경·축B 최소구현 사다리)으로 품질을 권고하며, gbc verify로 구현 후 케이스를 증거(JUnit·코드 독해)와 대조한다(verified>reviewed>unverifiable).