A personal link/notes archive exposed as an MCP server — the calling LLM decides categorization, you bring your own MongoDB
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"vata-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# vata-mcp Vata exposed as an MCP server: a personal link/notes archive. Save a link or note; the LLM already driving your chat session decides its title, category, description, and tags and calls `vata_save` with them — you just save and search, categories are never something you manage by hand. **Design note:** the categorization "AI" here is not a separate service — it's whichever model is already in your MCP client (Claude, etc.). The `/vata-save` prompt instructs it to call `vata_list_categories`, decide fit-or-new itself, and pass explicit values into `vata_save`. No `VATA_LLM_MODEL` or API key is needed for this to work well. That setting only matters for callers that *can't* reason (a bare script calling the tool directly) — see [Tools](#tools) below. Design docs: [VATA_MCP_PLAN.md](VATA_MCP_PLAN.md), [VATA_SCHEMA.md](VATA_SCHEMA.md), [VATA_DATAFLOW.md](VATA_DATAFLOW.md). Single-user per install — everyone who installs this runs their own server against their own database; there's no shared/multi-tenant service. ## Install ```bash pipx install vata-mcp vata-mcp setup ``` The setup wizard walks you through picking a MongoDB (local or Atlas, validated live), an optional fallback LLM, an optional access token, and an optional `/vata-clean` password — then prints a ready-to-paste MCP client config block. Full walkthrough, including MongoDB installation, in [SETUP.md](SETUP.md). Per-client config instructions (Claude Desktop/Code confirmed working; ChatGPT/Windsurf/Gemini attempted, not confirmed) in [CLIENT_SETUP.md](CLIENT_SETUP.md). Settings are saved to a config file (env vars still override it if set — see [Environment variables](#environment-variables)), so `vata-mcp setup` only needs to run once. ## Deploying remotely (optional, free tier, ~15 min) Only needed if you want Vata reachable from somewhere other than the machine it's installed on (e.g. so a phone or a second computer can use it too). Personal local use does not need this section at al…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.