Standard tool implementations (grep/glob/read/...) for sirius — product-layer reusable tool library, not framework-builtin

AI root-cause investigation from your terminal — point it at a repo, pipe it an alert, get an evidence-linked report

Structure-aware JSX/TSX editing for coding agents.

Run commands on files when Claude Code edits them

Pi extension that enforces self-documenting code by detecting and blocking unnecessary comments

OpenCode plugin that blocks reads/edits on files matching .agentignore / .claudeignore (gitignore syntax)

Auto-commit after every AI coding agent turn

A lightweight harness for Claude Code

Run a NanoCodana coding agent over your real filesystem, with a sandboxed shell that can escalate to the host on approval.

Claude Code hooks for logging prompts and file changes

MCP server that shows what your coding agent can do without asking you: every settings file that contributes permission rules, which one wins, what runs unattended, and the rules Claude Code accepts and then ignores.

OXLint/ESLint plugin for empirically-backed code quality metrics (WMC, Halstead, LCOM, CBO, DIT)

Lint/format a single file directly, without touching git — a nano-staged-config-compatible replacement for the naive nano-staged-in-a-hook approach

Protect sensitive files from AI coding assistants with .aiignore

Basic tools package for Magic Chat with filesystem, process, and utility operations

最小化 Agent 系统 —— 10 工具 + 插件系统 + workflow DSL + 自动压缩 + OpenAI 兼容 + Ink TUI；NodeNext + tsc 原地编译，dev 用 Bun .ts、install 用 Node .js（学习/教学用）

OpenCode plugin that substitutes unicode characters with ASCII equivalents in AI responses and file edits

Type-safe TypeScript SDK for building Claude Code hook extensions with edit tracking, context correlation, transform utilities, and non-blocking error handling

Per-project memory wiki for Claude Code, driven by hooks and a nightly compiler.

Claude Code quality-of-life improvements: auto-approve hooks, context monitoring, and status line enhancements