Categories › AI & Agents › Agent memory & context
MCP tool observatory: do registry servers answer, and are their answers true? No key.
Remote server: your requests and data go to robinsaige.com. Claude Code: claude mcp add --transport http verifier https://robinsaige.com/mcp · Claude app: Settings → Connectors → Add custom connector.
The tools this server listed when we connected to it, in its own words.
registry_pulse — Call this FIRST for the state of the MCP tool ecosystem in one shot: how many servers exist, how many actually answer a real handshake, how should_i_use — THE GATE — call this before your agent depends on a tool you don't already trust. Give an exact server registry name and get a one-word verdfind_tools — Find and RANK the trustworthy tool for a NEED. Describe the task in plain words ('screen a company for sanctions', 'US tariff data', 'vesselresolve_server_name — Resolve a partial or misspelled server name to its EXACT registry id before calling should_i_use / check_server. Matches a fragment against check_server — The FULL evidence behind the verdict for ONE server, by exact registry name — call should_i_use first for the one-word decision; call this wreport_call — AFTER your agent calls a tool, fire-and-forget how it went — success/failure, latency, cost — so the observatory accumulates realized reliabchanges_since — What changed in the tool economy since a date: verdict flips, deaths, revivals, new servers, confirmed drift — the census diff as data. Polllist_findings — The observatory's ruled findings — each with its claim and its falsification test — plus the public corrections log (what we published, thenNot tested by KitScout yet. When our harness runs it, pros, cons and the tasks we tried appear here, with the date and evidence.