# MCP Verify

> Trust, freshness, policy, and discovery layer for public MCP servers.

- name: io.sentinelsignal/verify
- version: 1.0.269
- connection class: unknown (unknown)
- trust: 46/100 flags: no-repository, unreachable
- quality: 85/100 (good)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/io.sentinelsignal%2Fverify
- tools: https://api.protogrid.dev/v1/servers/io.sentinelsignal%2Fverify/tools

**Not classified yet.** The remote endpoints have not been probed successfully, or the listing has nothing to connect to.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "verify": {
      "type": "http",
      "url": "https://verify.sentinelsignal.io/mcp"
    }
  }
}
```


## Trust

- hygiene: 80
- liveness: 0
- freshness: 65
- provenance: 55
- drivers: +dns-namespace +website -unreachable(5 failures) ~updated-149d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 85/100 (good)

### protocol: 75 (weight 25)

- warn `protocol.modern`: newest supported version is 2025-03-26; 2026-07-28 not supported, older versions are deprecated until 2027-07-28
- n/a `protocol.stateless`: only applies to 2026-07-28 servers
- pass `protocol.transport`: streamable HTTP
- n/a `protocol.list_ttl`: only applies to 2026-07-28 servers

### 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: 86 (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
- pass `tools.annotations`: 22 of 22 tools declare readOnlyHint or destructiveHint
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: get_server_report (destructiveHint, idempotentHint, openWorldHint); search_servers (destructiveHint, idempotentHint, openWorldHint); recommend_servers (destructiveHint, idempotentHint, openWorldHint); …
- pass `tools.token_cost`: about 2,562 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: 22, about 2,562 tokens to load them all
- tool set hash: 92274d246f2886f26e4dd08d2d6bb550
- badge: [![protogrid quality](https://protogrid.dev/badge/io.sentinelsignal/verify.svg)](https://protogrid.dev/servers/io.sentinelsignal/verify)
- 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://verify.sentinelsignal.io/mcp (streamable-http, auth none, reachable false, uptime30d 0.9381443)

## Packages

_none_

## Tools (22)

- `compare_servers`: Compare up to four MCP servers side by side across score, verdict, auth, tool count, prompts/resources, and freshness.
- `compile_docs`: Generate agent-readable MCP artifacts from product docs: server card, llms.txt, tool schemas, examples, policy metadata, score preview, and hosted docs page.
- `create_mandate`: Create a Verify Mandate: signed delegated authority for an agent/user/server scope, budget, validity window, and tool constraints.
- `decide_agent_call`: Verify Gateway decision point. Decide whether this agent may call this tool, with this payload, for this user, right now. Writes an audit-ledger decision event.
- `explain_agent_risk`: Agent Sprawl Radar: explain risk findings and the recommended action for one registry agent.
- `export_policy`: Export a JSON TrustOps policy for one MCP server with allow, blocked tools, required scopes, freshness, and approval gates.
- `fetch`: ChatGPT-compatible read-only fetch alias. Returns a full MCP server item with id, title, text, url, and metadata.
- `generate_remediation_plan`: Agent Sprawl Radar: generate deterministic remediation steps for risky or drifting agents.
- `get_agent`: Agent Sprawl Radar: return one agent with tools, data access, findings, drift, and recommended action.
- `get_drift_report`: Agent Sprawl Radar: return agent drift events across latest snapshots.
- `get_evidence_pack`: Return Verify Ledger evidence packs with invocations, hash-chained audit events, decisions, outcomes, mandates, and cost summaries.
- `get_gateway_options`: Return Verify Gateway semantics, /v1/route versus /v1/decide guidance, mandate inputs, ledger evidence hooks, and decision fields.
- `get_hosting_options`: Return Verify Hosted MCP runtime/sandbox hosting capabilities, controls, and endpoints.
- `get_server_report`: Return the full machine-readable verify report for a specific MCP server identifier.
- `get_subscription_options`: Return alert types, subscription channels, watch scopes, and existing subscription endpoints.
- `list_agents`: Agent Sprawl Radar: list discovered AI agents, MCP servers, bots, workflows, risk levels, and owners.
- `list_high_risk_agents`: Agent Sprawl Radar: list high and critical risk agents for review.
- `recommend_servers`: Recommend MCP servers for a plain-language task and return ranked matches with install config.
- `route_task`: Choose the best MCP server/tools for a task. Use decide_agent_call when evaluating one specific attempted tool call.
- `score_agent`: Agent Sprawl Radar: score a normalized agent payload or an existing registry agent.
- `search`: ChatGPT-compatible read-only search alias. Returns MCP server results with id, title, and url.
- `search_servers`: Search and rank MCP servers by capability, auth preference, client target, and risk tolerance.