code-graph-mcp
AST knowledge graph MCP server for Claude Code — semantic search, call graph traversal, HTTP route tracing, impact analysis. Auto-indexes 10 languages via Tree-sitter.
- 97
- Rust
- MIT
- 1 d ago
sdsrss
Built from sdsrss's public GitHub profile. Not on Take My Repo yet, so there are no posts or updates here.
AST knowledge graph MCP server for Claude Code — semantic search, call graph traversal, HTTP route tracing, impact analysis. Auto-indexes 10 languages via Tree-sitter.
Persistent long-term memory for Claude Code via MCP — captures coding decisions, bugfixes, and context across sessions. Hybrid FTS5 + TF-IDF search with episode batching. Single SQLite DB, no external services. Alternative to claude-mem with 600x lower cost.
AI orchestration tool for Claude Code with MCP state management, review workflows, and plugin auto-update
Claude Code plugin: AI-CODING-SPEC v6.11 enforcement via shell hooks + spec distribution into ~/.claude/. Blocks banned vocab in commit messages, unsafe rm -rf $VAR, push-on-red-CI, unread MEMORY.md before ship. Ships a standalone npx CLI for git pre-commit + GitHub Actions.
Loop Engineering plugin for Claude Code: /autoloop drives tasks to completion via builder/checker cycles; /polish iteratively raises code quality via adversarial review. Stop-hook enforced, six stop rules, install via /plugin marketplace add sdsrss/loop_eng
All-in-one engineering workflow & knowledge engine for Claude Code. One-command install, runs standalone (Node >=18, no bun). L0-L3 task classification, 13 runtime invariants, code review, browser QA, security audit, and a deduplicated knowledge base that compounds across tasks.
Autonomous QA loop for Claude Code & Codex — an AI agent uses your finished app like a real user in a sandbox, finds bugs, safely fixes them with regression tests, and loops until it converges. Hands-off, self-healing, honest.
Compile messy docs into an agent-maintained llm_wiki (Karpathy-style): whole-page Q&A — no chunks, no vector DB required. CLI + Claude Code/Codex skills + MCP server.
Make any legacy project Claude-Code-ready — without forcing AI tooling on your teammates. 10-phase /onboard Skill: discovers stacks, generates compact CLAUDE.md, configures guard hooks, batches install plan. Default --local-only mode = zero team pollution.
Multi-model committee (Mixture-of-Agents) for Claude Code — your agent chairs up to 4 heterogeneous LLMs to blind-review, decide & brainstorm in parallel, with evidence-driven convergence for more reliable conclusions.
A coding-discipline spec for OpenAI's Codex CLI, enforced by native Codex hooks + a rule-hit telemetry loop. Turns an AGENTS.md the model drifts from into a system that holds. Independent of oh-my-codex.
Git-backed version control, rollback, and provenance for your Claude Code & Codex Agent Skills — a safety-gated skill package manager (CLI + plugin).