MCP Verify
io.sentinelsignal/verify
4 toolsHTTPregistry active
Summary
If you're building agents that call third-party MCP servers, this is a policy and trust layer that sits between your agent and the rest of the ecosystem. It exposes search_servers to discover tools, decide_agent_call to enforce allow/deny rules at runtime, create_mandate for signed governance policies, and get_evidence_pack for audit trails. The backend continuously scores over 21,000 public MCP servers on freshness, auth posture, write risk, and payment signals, then serves that intel via a read-only MCP interface at verify.sentinelsignal.io/mcp. Reach for it when you need centralized visibility into which tools your agents are invoking, compliance-ready logs, or a way to block risky servers before they execute.
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Tools
Public tool metadata for what this MCP can expose to an agent.
4 tools
search_serversSearch and rank MCP servers by capability, auth preference, client target, and risk tolerance.6 params
Search and rank MCP servers by capability, auth preference, client target, and risk tolerance.
Parameters* required
limitintegerMaximum number of ranked candidates to return.
querystringOptional free-text query to bias ranking toward a specific use case or domain.
capabilitiesarrayNormalized capability taxonomy terms such as healthcare, search, read, exec, files, oauth, or prompts.
client_targetstringOptional target client or integration surface.one of
openai_connectors · claude_desktop · smithery · generic_http · healthcarerisk_tolerancestringHow much tool-surface risk is acceptable.one of
any · low · medium · highauth_preferencestringPreferred auth mode. Use unauthenticated when the agent must avoid login flows.one of
any · oauth · unauthenticatedrecommend_serversRecommend MCP servers for a plain-language task and return ranked matches with install config.6 params
Recommend MCP servers for a plain-language task and return ranked matches with install config.
Parameters* required
taskstringPlain-language task description such as 'I need an MCP for healthcare denial scoring in OpenAI connectors'.
limitintegerMaximum number of recommendations to return.
capabilitiesarrayOptional explicit capability constraints that override or extend the inferred task capabilities.
client_targetstringOptional target client or integration surface.one of
openai_connectors · claude_desktop · smithery · generic_http · healthcarerisk_tolerancestringHow much tool-surface risk is acceptable.one of
any · low · medium · highauth_preferencestringPreferred auth mode.one of
any · oauth · unauthenticatedget_server_reportReturn the full machine-readable verify report for a specific MCP server identifier.1 params
Return the full machine-readable verify report for a specific MCP server identifier.
Parameters* required
identifierstringCanonical server identifier in namespace/name format.
compare_serversCompare up to four MCP servers side by side across score, verdict, auth, tool count, prompts/resources, and freshness.1 params
Compare up to four MCP servers side by side across score, verdict, auth, tool count, prompts/resources, and freshness.
Parameters* required
identifiersarrayCanonical server identifiers in namespace/name format.
Featured
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →