Home / MCP registry index / io.github.ra1labsworkx-wq/snapback
Snapback
io.github.ra1labsworkx-wq/snapback in the official MCP registry.
Endpoint https://api.snapback.sh/mcp. Source: github.com/ra1labsworkx-wq/snapback-selfheal.
The registry describes it as: Diagnose why an AI agent failed and get the verified fix instantly. Free, no token.
We read its 23 tools on 2026-10-26. Nothing in them matched our tool-poisoning rules.
A clean read means these rules found nothing, not that the server is harmless. A server can change its tools at any time, which is why the index reads it again every week.
Tools it advertises
diagnose_trace diagnose_batch get_verdict preflight submit_feedback get_request_status my_usage my_impact what_others_did report_outcome recommend_failover cascade_root suggest_budget_recovery request_pattern detect_loop session_start session_step session_end agent_memory budget_guard convert_trace search_docs diagnose_infra_error
Run this server? Show the check in your README
The badge reflects the latest weekly read: 
[](https://lazaretto.dev/mcp-registry/io-github-ra1labsworkx-wq-snapback)How this was checked
Lazaretto connects to the endpoint the way any MCP client does, asks the server to introduce itself and list its tools, and reads every name, description, parameter schema and server instruction for tool poisoning: hidden directive blocks, orders to read local secrets, parameters built to carry secrets or the conversation out, orders that redirect another server's tools, and invisible characters. It never calls a tool. What we found across the registry.
To run the same check every time Claude Code adds a server, install the free lazaretto-guard plugin. Browse every server in the index.