Cross-language-safe code-intelligence MCP for AI agents — 13 languages, family-gated call graph (CodeGraph wires 745 cross-language mis-wires on this repo, TSA 6 — ~390x cleaner). Run miswire-audit on your repo. 8 facade tools, TOON output, 100% local. Python.

Local-first memory and workflow plugin for AI coding agents: Codex, Claude Code, agent preferences, project lessons, and delivery rules.

MCP server that gives Claude Code the rationale behind a symbol before editing - synthesized from git history, PRs, and call-graph context.

Local-first project cognition and impact analysis layer for AI coding agents.

Local code graph for AI coding agents. Indexes your repo into Kuzu + SQLite FTS, exposes 30+ MCP tools to Claude Code, Cursor, Codex, and Gemini. Federates across sibling repos. Tool calls cost zero model tokens.

AEGIS 2.0: multi-agent code repository analysis system with CodeGraph, source-grounded RAG, and agent handoff for the Volcano Cup intelligent agent competition

CTX, TVL, RLM, CodeGraph, durable agent memory and universal cross-agent handoff for lower context/token usage