io.github.ra1labsworkx-wq/snapback

up checked 2026-11-11T08:39:50.060Z

Unclaimed · Unverified

Agent searches it appeared in, last 30 days: 0. Run this server? Claim the listing: verified label, uptime alerts and analytics.

Publisher-provided title

Snapback

Publisher-provided description

Diagnose why an AI agent failed and get the verified fix instantly. Free, no token.

Summary by lookupd

Diagnoses failed agent runs, finds root-cause errors, detects loops and budget overruns, and returns verified fix suggestions for agent…

Install

Claude Code, remote (streamable-http)

claude mcp add --transport http snapback https://api.snapback.sh/mcp

Codex, remote (streamable-http)

codex mcp add snapback --url https://api.snapback.sh/mcp

Cursor, remote (streamable-http) → .cursor/mcp.json

{
  "mcpServers": {
    "snapback": {
      "url": "https://api.snapback.sh/mcp"
    }
  }
}

Other clients: docs. Or tell your agent: Set up lookupd for me: read https://lookupd.com/agent.md and follow it.

Tools (23 tools)

Publisher-provided text

diagnose_trace Diagnose why an AI agent run failed. Returns a structured verdict (failure_class, failed_at_step, root_cause, fix_suggestion, confidence). LATENCY: known patter
diagnose_batch Diagnose SEVERAL traces in one call (up to 20). Each trace is metered like a separate diagnose_trace. Returns a verdicts array (per-trace, order preserved); a b
get_verdict Fetch a previously produced verdict by its id or by trace_id (your org only).
preflight BEFORE running: get known failure patterns for a given agent setup so you can avoid them. Returns a ranked list of {failure_class, root_cause, fix_suggestion} f
submit_feedback Tell Snapback whether a verdict was correct (correct=true/false), with an optional free-text note (did the fix work? what was wrong?). ONE rating per verdict —
get_request_status Check what happened to a pattern you requested (from request_pattern's request_id). Returns pending / approved / rejected / in_library so you can see if your su
my_usage See YOUR current usage + remaining allowance so you can self-govern spend: snapbacks (diagnoses) used/cap/remaining, guard checks used/cap/remaining, estimated
my_impact See how your feedback + pattern requests have shaped the shared library — how many verdicts you've rated, patterns you've requested, and how many were approved
what_others_did THE CROWD: for a failure_class (or pass the family/error and we'll map it), see what OTHER agents did about the same failure and whether it worked — anonymized,
report_outcome Report the outcome of an auto-applied fix (the self-heal interceptor calls this after it gate-applied a fix and retried). Pass failure_class, family, fix, confi
recommend_failover Should you RETRY the same target, SWITCH provider, FALL BACK to another chain, or STOP? Pass the error + your configured topology (current_chain, available_chai
cascade_root Given an ORDERED list of errors from a run (oldest first), find the TRUE root — the error that cascaded or shouldn't have been retried — not just the final symp
suggest_budget_recovery Approaching a token/context/cost budget mid-run? Pass your counters and get the LEAST-DISRUPTIVE recovery ranked: truncate context | switch to a cheaper model |
request_pattern Leave us a message: ask us to add a failure pattern to the library, or report a problem we couldn't diagnose well. Use this when diagnose_trace didn't have a go
detect_loop MID-RUN loop check (fast, no LLM, free). Send your recent steps DURING a run; get back whether you're stuck repeating a tool call and a concrete next move. Call
session_start Open a LIVE mid-run session so Snapback can watch your run step-by-step and warn you in real time (loop / token / cost / context) - the always-on guardian mode.
session_step Report ONE step of a live run and get back any warnings immediately (loop detected / budget breach). Pass the step (action + inputs) and any counters you have (
session_end Close a live session and get a short run summary (total steps, duration). Frees the session. Free, no token.
agent_memory See what YOU (this agent) tend to fail on — your recurring failure patterns across past diagnoses. Returns your top failure classes with counts and what share o
budget_guard Live MID-RUN budget check (fast, no LLM). Send whatever counters you have and get an advisory on context %, token burn, cost burn, step budget, and off-task dri
convert_trace Turn your raw logs into a Snapback trace so you don't hand-craft JSON. Pass 'source' = a list of log/step entries, or an object with a spans/steps/messages/even
search_docs Search Snapback's documentation for how to use it — how to format a trace, what each tool does, the failure taxonomy, auth, pricing, and errors. Free and needs
diagnose_infra_error Diagnose a cryptic AGENT-INFRASTRUCTURE error — payments (x402/EIP-3009), Solana on-chain (ATA, blockhash, compute), MCP protocol, RPC/providers, EVM & Solana w

Details

Version: 1.8.4
Transports: streamable-http
Uptime (7 days): 100%
Median latency: 1519 ms
First indexed: 2026-11-01T09:50:47.061Z
Last seen in registry: 2026-11-01T09:50:47.111Z
Repository: https://github.com/ra1labsworkx-wq/snapback-selfheal

Badge for your README

Shows this live status and links here; updates hourly.

[![lookupd status](https://lookupd.com/badge/io.github.ra1labsworkx-wq/snapback.svg)](https://lookupd.com/servers/io.github.ra1labsworkx-wq/snapback)

62033 servers
Last crawl: 2026-11-11T15:38:39.561Z
API base: https://api.lookupd.com
MCP URL: https://api.lookupd.com/mcp