The missing DevTools for Claude Code — inspect session logs, tool calls, token usage, subagents, and context window in a visual UI. Free, open source.

Gift your VS Code agent a real debugger: breakpoints, stepping, inspection.

Local execution trees for TypeScript AI agents. agent-inspect helps you understand what happened inside an AI agent run — locally. It turns manual steps, tool calls, LLM calls, structured logs, failures, durations, and run metadata into readable execution trees you can inspect from the terminal. It is built for TypeScript/Node.js developers..

Drive x64dbg with your AI. MCP server: 23 mega-tools / 153 endpoints for breakpoints, memory, disasm, tracing, anti-debug & PE dumping. Claude/Cursor/Windsurf/Cline. All local.

A stop-loss for Claude Code: detects loops, compaction danger, failed tools, and token waste before the next request.

Local-first situational awareness layer for AI agents. Single binary, written in Rust.

🤖 An AI-native MCP server for .NET memory dump investigations. Leverage LLMs (Claude/Gemini) to analyze heaps, threads, and exceptions using natural language. Built with ClrMD, token-efficient paging, and Docker support for cross-architecture (.NET/AMD64/ARM64) post-mortem analysis.

Local-first agent debugger with replay, failure memory, smart highlights, and drift detection.

Codex CLI session supervision: see failed or incomplete turns, token use, model changes, context pressure, and postmortems.

AI-Powered Execution Cinema Debugger for Elixir/BEAM

ai agent

🩺 RAG Doctor — Open-source diagnostic tool for Retrieval-Augmented Generation (RAG) systems. Analyzes codebases to detect architectural issues in LLM pipelines such as missing retrieval, bad chunking, embedding mismatches, and vector database misuse.

Claude Code skill + CLI that lets an AI agent drive Rider's MCP debugger to debug a running Unity Editor at runtime, no session restart needed.

MCP server for rr reverse debugging — enables AI assistants to record, replay, and inspect program executions via GDB/MI

Provide an MCP server to enable reverse debugging with rr, improving control and insight during program execution analysis.

QA Continuity AI — a private, on-premise platform that turns scattered project knowledge (rules, flows, bugs, tests, CI) into test plans, Playwright automation, and searchable QA memory. 100% local LLMs via Ollama + HuggingFace, zero API cost.

AI-assisted MCU debug automation framework / AI 辅助 MCU 自动调试框架

Local visual debugger for AI agent runs: replay failures, inspect prompts/tools/results, and export redacted evidence.

Quick entry for the WFGY AI Troubleshooting Atlas. Fast links, quick start, and a 60-second try. Part of the WFGY ecosystem.

🏎️ DRS: Debugging at the speed of F1. An AI-powered Rovo Agent that instantly links Jira incidents to buggy Bitbucket commits.