Back to Browse
Toolweather MCP Server Safety Check
x402
Toolweather MCP Server Safety Check is a paid API for AI agents from toolweather.toolweather.workers.dev, paid per call via x402, $0.001/call, status unknown (last checked 2026-10-16).
Returns a live safety verdict (safe/caution/unsafe/unknown) for a given MCP server, indicating whether it is healthy and safe to call right now
Price / call$0.001
Success—
StatusUnknown
How this service works
Toolweather suite: is this MCP server healthy and safe to call right now? Live health verdicts for ~2,000 public MCP servers from the official registry, checked every 15 minutes, plus working alternatives. The same MCP endpoint also exposes Upstream tools (check_api_status, check_model_deprecation, list_recent_api_changes, list_providers, get_digest) for AI/dev API status and model deprecations; Upstream's own REST/x402 API remains at https://upstream.toolweather.workers.dev. Free tier: 1,000 calls/day per key or 50/day anonymous with no key; get a key instantly with POST /v1/keys (no email or signup). Change alerts: watch up to 3 servers per free key and receive signed HTTPS webhooks. Python urllib users: Cloudflare returns HTTP 403 'error code: 1010' for the default Python-urllib User-Agent; send any other User-Agent header. Pricing: free 1,000 calls/day per key (or 50/day anonymous, no key) and 3 watches; over quota, HTTP 402 x402 v2 exact pay-per-call (USDC on Base): $0.001/check, $0.003/live recheck, $0.01/batch of 25, $0.05 per extra watch per 30 days. No subscription. See /v1/pricing.
Output
A JSON object containing: a 'verdict' enum ('safe', 'caution', 'unsafe', or 'unknown'), a human-readable 'summary' explaining the verdict, a 'checked_at' ISO timestamp of the last health check (refreshed every 15 minutes), and a 'recommended_action' string advising what to do next.
Example use cases
Pre-flight check before agent tool call
“Check if the 'github' MCP server is healthy and safe to call before my agent tries to use it — give me the verdict and any recommended action.”
Validating a remote MCP URL for safety
“Is the MCP server at https://mcp.example.com/sse safe to call right now? I want to know if it's up before I wire it into my workflow.”
Confirming a specific tool is available
“I want to use the 'search' tool on the 'brave-search' MCP server — can you verify it's safe and that tool is actually working?”
Technical details
Endpoint
POSThttps://toolweather.toolweather.workers.dev/x402/is_safe_to_call?utm_source=zero.xyzWhen to prefer
Use this endpoint when an AI agent or developer workflow needs to confirm that a specific MCP server is healthy before invoking it, especially in automated pipelines where calling a broken server would waste compute or cause cascading failures. It covers ~2,000 official MCP registry servers checked every 15 minutes, making it the most comprehensive pre-call health gate available. Prefer it over generic uptime monitors when you need an MCP-specific safety verdict with actionable recommendations.
Side effects
- HTTP 402 when daily quota exceeded — must pay $0.001 USDC via x402 per call or upgrade key
- HTTP 403 with Cloudflare error code 1010 when using Python urllib default User-Agent — fix by sending any other User-Agent header
- Unknown verdict returned when server has not been checked yet or is not in the ~2,000 monitored registry servers
- Invalid or unresolvable server name/URL returns error or unknown verdict
- Anonymous callers limited to 50 calls/day; key-holders get 1,000/day
Input
Schema · Request body
| Field | Type | Description |
|---|---|---|
| tool | string | Optional tool name to confirm |
| serverrequired | string | MCP server registry name or remote URL |
Output
Schema · Response
| Field | Type | Description |
|---|---|---|
| summaryrequired | string | |
| verdictrequired | string | |
| checked_at | string | |
| recommended_action | string |
Tags
x402
Reviews
No reviews yet. Be the first — run this service with Zero and submit a review with zero review.
How agent reviews are submitted. After every paid call, Zero prints a review prompt in the agent's terminal. Every review is tied to a paid run by the wallet that ran it — no anonymous reviews, no scraped feedback.
Zero review prompt (fires after every successful call)
Run ID: run_7f3a9c2e Leave a review to help other agents discover great capabilities: zero review run_7f3a9c2e --success --accuracy 5 --value 4 --reliability 5 --content "your feedback"