# Euvra

> Discover, compare, route, and execute machine-accessible capabilities for AI agents.

- name: dev.euvra/euvra
- version: 1.0.0
- connection class: R0 (autonomous)
- trust: 80/100 flags: no-repository
- quality: 73/100 (needs work)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/dev.euvra%2Feuvra
- tools: https://api.protogrid.dev/v1/servers/dev.euvra%2Feuvra/tools

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

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "euvra": {
      "type": "http",
      "url": "https://euvra.dev/mcp"
    }
  }
}
```


## Trust

- hygiene: 80
- liveness: 100
- freshness: 100
- provenance: 45
- drivers: +dns-namespace +reachable +uptime-100% +updated-20d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 73/100 (needs work)

### 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 carries no ttlMs, so clients re-fetch it

### 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: execute_capability (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); compare_capabilities (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); discover_capabilities (readOnlyHint, destructiveHint, idempotentHint, openWorldHint)
- pass `tools.token_cost`: about 644 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: 50 (weight 15)

- pass `stability.changes`: 1 tool changes in 30 days
- fail `stability.rug_pull`: 1 tools kept their name but changed most of their description in 30 days; review before trusting them

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

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

- tools: 3, about 644 tokens to load them all
- tool set hash: 5f81dfa671363b39b2d9087849ea50bc
- tools last changed: 2026-10-09T10:53:21.182Z
- badge: [![protogrid quality](https://protogrid.dev/badge/dev.euvra/euvra.svg)](https://protogrid.dev/servers/dev.euvra/euvra)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

- 2026-10-09 `execute_capability` changed (description, inputSchema) **meaning changed**
  - before: Route a natural-language capability request through Euvra and execute the selected live-discovered MCP tool. Euvra re-plans the route server-side and never trusts caller-supplied provider IDs, endpoints, or tool names.
  - after: Execute a real action using a live MCP provider selected by Euvra. Use discover_capabilities to find available capabilities and compare_capabilities to inspect route readiness before execution. Call this tool only when the user wants the action performed, not merely researched or compared. Supply a specific natural-language query describing the desired action and an arguments object containing the selected provider tool inputs when known. Euvra independently re-plans the route and does not accept caller-selected provider IDs, endpoints, or tool names. Execution may return NO_ROUTE, AUTH_REQUIRED, or BLOCKED if no eligible live tool is available. Paid execution requires an x402 payment of 0.01 USDC on the configured network; a client without payment support may receive payment requirements instead of an executed result. Do not claim the action succeeded until the response confirms success.

Full history: https://api.protogrid.dev/v1/servers/dev.euvra%2Feuvra/changes

## Remotes

- https://euvra.dev/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

_none_

## Tools (3)

- `compare_capabilities`: Ask Euvra to compare discovered providers for a request and select the strongest route candidate. This does not execute the provider.
- `discover_capabilities`: Discover and rank machine-accessible capabilities that may satisfy a natural-language request. Returns evidence, operational status, and ranking signals without executing a provider.
- `execute_capability`: Execute a real action using a live MCP provider selected by Euvra. Use discover_capabilities to find available capabilities and compare_capabilities to inspect route readiness before execution. Call this tool only when the user wants the action performed, not merely researched or compared. Supply a specific natural-language query describing the desired action and an arguments object containing the selected provider tool inputs when known. Euvra independently re-plans the route and does not accept caller-selected provider IDs, endpoints, or tool names. Execution may return NO_ROUTE, AUTH_REQUIRED, or BLOCKED if no eligible live tool is available. Paid execution requires an x402 payment of 0.01 USDC on the configured network; a client without payment support may receive payment requirements instead of an executed result. Do not claim the action succeeded until the response confirms success.