For debugging in general, Sentry is the most established option: it provides error monitoring and stack-trace analysis through an MCP server, SDK and CLI, with a dedicated action for debugging issues 111. PostHog, MLflow and Streamlit also offer debugging actions for narrower contexts such as React code, AI agent traces and Streamlit apps 2121615.
- 1
Sentrysentry.ioverified and signed
Application performance monitoring and error tracking that exposes an MCP server, SDK, CLI, and a debug-issue action to find an issue and pull its stack trace, breadcrumbs and logs 111.
official:MCP server · @sentry/mcp-serverSDK · @sentry/nodeCLI · sentry-clirepository · sentry-pythonskill · sentry-cli+49 community tools built for it
for agents: sentry · sentry-debug-issue · sentry-instrument
- 2
- 3
- 4
- 5
DebugBundledebugbundle.comwebsite read, tools not yet checked
Production debugging tool that captures production errors and their context into structured debug bundles for AI coding agents, with an MCP server and package 10.
official:MCP server · DebugBundlepackage · DebugBundle+1 community tools built for it
- 6
Sources
- 1Sentry Docs | Application Performance Monitoring & Error Tracking Softwaredocs.sentry.io · read 2026-10-29
- 2PostHog – Your product’s context layerposthog.com · read 2026-10-29
- 5Streamlit documentationdocs.streamlit.io · read 2026-10-29
- 6MLflow - Open Source AI Platform for Agents, LLMs & Modelsmlflow.org · read 2026-10-29
- 9copilotkit.dev — We build AI tools that power AI toolscopilotkit.dev · read 2026-10-29
- 10Production debugging for AI coding agents | DebugBundledebugbundle.com · read 2026-10-29
- 11sentry: declares "sentry" (the record we read and signed)unfragile.ai · read 2026-10-24
- 12react-doctor: declares "react-doctor" (the record we read and signed)unfragile.ai · read 2026-10-24
- 15debugging-streamlit: declares "debugging-streamlit" (the record we read and signed)unfragile.ai · read 2026-10-24
- 16analyzing-mlflow-trace: declares "analyzing-mlflow-trace" (the record we read and signed)unfragile.ai · read 2026-10-24
- 19copilotkit-debug: declares "copilotkit-debug" (the record we read and signed)unfragile.ai · read 2026-10-24
For agents: the 10 actions behind this answer
sentry · sentry verified and signed
Sentry error monitoring integration. Access error reports, analyze stack traces, search issues by fingerprint, and debug production errors directly from your development environment.
react-doctor · react-doctor verified and signed
Diagnose and fix React codebase health issues. Use when reviewing React code, fixing performance problems, auditing security, or improving code quality.
ghidra · ghidra verified and signed
Reverse engineer binaries using Ghidra's headless analyzer. Decompile executables, extract functions, strings, symbols, and analyze call graphs without GUI.
brapi-get_DevtoolsJsonProtocol · Cloudflare API verified and signed
Get Chrome DevTools Protocol schema.
debugging-streamlit · debugging-streamlit verified and signed
Debug Streamlit frontend and backend changes using make debug with hot-reload. Use when testing code changes, investigating bugs, checking UI behavior, or needing screenshots of the running app.
analyzing-mlflow-trace · analyzing-mlflow-trace verified and signed
Analyzes a single MLflow trace to answer a user query about it. Use when the user provides a trace ID and asks to debug, investigate, find issues, root-cause errors, understand behavior, or analyze quality. Triggers on "analyze this trace", "what went wrong with this trace", "debug trace", "investig
· Joern website read, tools not yet checked
· strace website read, tools not yet checked
copilotkit-debug · copilotkit-debug verified and signed
Use when diagnosing CopilotKit issues -- runtime connectivity failures, agent not responding, streaming errors, tool execution problems, transcription failures, version mismatches, and AG-UI event tracing.
· DebugBundle website read, tools not yet checked
DebugBundle captures production errors and their context in structured debug bundles, giving AI coding agents reliable evidence to investigate.
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.