# Vouch

> Vouch — independently measured reliability scores for MCP tools, not self-reported claims.

- name: io.github.AivaStack/vouchtools
- version: 0.1.0
- connection class: R0 (autonomous)
- trust: 94/100
- quality: 83/100 (good)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/io.github.AivaStack%2Fvouchtools
- tools: https://api.protogrid.dev/v1/servers/io.github.AivaStack%2Fvouchtools/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "vouchtools": {
      "type": "http",
      "url": "https://vouch.tools/api/mcp"
    }
  }
}
```


## Trust

- hygiene: 100
- liveness: 100
- freshness: 85
- provenance: 90
- drivers: +repository +namespace-matches-repo +website +reachable +uptime-100% ~updated-48d-ago
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 83/100 (good)

### protocol: 88 (weight 25)

- pass `protocol.modern`: supports 2026-07-28 (server/discover)
- pass `protocol.stateless`: answers without a session
- pass `protocol.transport`: streamable HTTP
- warn `protocol.list_ttl`: tools/list declares ttlMs 0, so clients re-fetch it on every use

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 71 (weight 30)

- pass `tools.descriptions`: every tool has a description
- pass `tools.description_length`: descriptions are concise
- pass `tools.schemas`: every tool has an object input schema
- fail `tools.annotations`: no tool declares readOnlyHint or destructiveHint, so clients cannot tell safe calls from risky ones
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: vouch_find (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); vouch_compare (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); vouch_check (readOnlyHint, destructiveHint, idempotentHint, openWorldHint)
- pass `tools.token_cost`: about 454 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: 100 (weight 15)

- pass `stability.changes`: no tool changes in 30 days
- pass `stability.rug_pull`: no tool changed its meaning under the same name

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 3, about 454 tokens to load them all
- tool set hash: 5c80f2c7aea581bf17b838f12641c54c
- badge: [![protogrid quality](https://protogrid.dev/badge/io.github.AivaStack/vouchtools.svg)](https://protogrid.dev/servers/io.github.AivaStack/vouchtools)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

_No tool definition changes recorded._

## Remotes

- https://vouch.tools/api/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

_none_

## Tools (3)

- `vouch_check`: Get an MCP tool's independently measured trust score before relying on it: real invocation trials, not self-reported metadata. Call this before adding a new MCP server to a project, or when deciding between two tools that do similar things. Look up by tool name, package, or server — free text.
- `vouch_compare`: Compare 2-5 MCP tools' measured behaviour side by side before picking one. Call this when weighing alternatives for the same task. Never collapses the comparison into a single ranked number — each tool keeps its own score and tier, and a caveat is surfaced when compared tools weren't measured at the same tier.
- `vouch_find`: Search Vouch's independently measured corpus of MCP tools by free-text query. Call this when choosing an MCP tool for a task and you want options ranked by measured reliability rather than popularity — optionally filtered by measurement tier (deep = full behavioural battery, shallow = schema + sample-call only).