Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp.blust.ch search for evidence about Robert Blust's Austronesian linguistics work"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
The reference instance of CompanyGraph, robertblust/mental-model, served over MCP at
https://mcp.blust.ch/mcp. This repository pins one commit of the model and one release of
companygraph/mcp-server, builds an image that carries the model's snapshot, and runs it on
Cloud Run in Zurich behind Firebase Hosting. Everything below the Google Cloud project is
Terraform, applied by GitHub Actions.
Whoever asks about Robert Blust's work — a person, a search engine, an agent — reaches the same answer, because every surface derives from one model. This is the surface an agent queries.
Add https://mcp.blust.ch/mcp as a custom connector in Claude, or as a remote MCP server in
ChatGPT's developer mode or the Gemini CLI. No authentication. Seven tools: list_types,
describe_schema, list_entities, get_entity, find_evidence, search and fetch; every
answer names the model commit it was read from.
Related MCP server: Provenance-Aware Retrieval MCP Server
source.json names the model commit and package.json the server release. Moving either is a
pull request; the merge builds the image, applies the infrastructure with it and checks that
/healthz reports the new commit.
npm ci
npm run snapshot # writes snapshot.json from the pinned commit
npm test # the seven tools against that snapshot
docker build -t mcp-blust-ch:local .infra/bootstrap/ is applied once by the owner and holds what CI needs before it can
authenticate: the state bucket, the identity pool, the two service accounts and the image
registry. infra/ is applied by CI on every merge. The workflows in .github/workflows/ say
what runs when.
Publishing to the MCP Registry runs in the registry environment, which requires the owner's
review of every run. The signing key lives there as an environment secret, MCP_PRIVATE_KEY,
never as a repository secret, because a repository secret would be readable by any workflow on
any branch and the review gate would protect nothing.
CC BY 4.0 for the text here; the model's own license is its own.
This server cannot be deployed
Company brain for AI agents — temporal knowledge graph search, exploration, and durable memory.
Multiple MCP tools, persistent graph memory, token-saving data pointers, and more.
CompanyLens is a remote MCP server giving AI agents instant access to official company registry data across 19 jurisdictions in Europe, the Americas, and Asia-Pacific. Eighteen read-only tools let you search companies and people, look up officers and beneficial owners, map corporate networks through shared directors, screen names against the UK disqualified directors register, find every company at a registered address, and pull filing history — all from a single connector. Visit our website: https://companylens.io
Read-only hosted MCP over CanonicAI's cited Answers corpus on canonicai.com.
