Connect Claude Code

Connect Claude Code to Tares so it can read your data over MCP. Three ways: the plugin (one install, with MCP read-back plus session capture), stdio (the client spawns a local proxy, simplest for an agent on the same machine as the daemon), or HTTP (connect to a running MCP server, for a remote or deployed instance).

Choose a connection

Plugin (recommended)

One install wires the MCP server and streams your sessions into the claude_code source:


/plugin marketplace add glassflow/tares
/plugin install tares@tares

The install prompts for the Tares URL (the default matches a local tares up), an optional access token (on a secured instance, a read + ingest API key), and a capture toggle. Requires the tares package on PATH (uv tool install tares). The plugin spawns tares-mcp like the stdio path does.

stdio (local)

The client runs tares-mcp, which proxies to the daemon. The tares package must be installed where Claude Code runs.


claude mcp add tares \
  --env TARESD_URL=http://localhost:8787 \
  -- tares-mcp

If the daemon requires a token, add --env TARES_AUTH_TOKEN=<token>.

HTTP (remote)

First run the MCP server (or use a deployed instance that already serves /mcp):


tares mcp --transport streamable-http --port 8788 --taresd http://localhost:8787

Then add it by URL:


claude mcp add --transport http tares http://localhost:8788/mcp

If the instance sets TARES_AUTH_TOKEN, pass it:


claude mcp add --transport http tares https://tares.example.com/mcp \
  --header "Authorization: Bearer <token>"

The console’s Agents → Connect tab generates these commands for your instance, with the endpoint and token filled in.

Verify the connection


claude mcp list
# tares: connected

Use it

In a Claude Code session, refer to Tares by name. The agent discovers and calls the tools:

Use tares: which sources am I ingesting, and what does each contain?

Use tares to describe source:metrics and tell me the entities it carries.

With tares, create a view correlating my logs and metrics by service, then show me the timeline for service=checkout.

The agent uses catalog_list / catalog_describe to learn the data, derive to compose a view, query to read it, and remember to write back observations.

Watch the calls

In the Tares console, Agents → Reads lists each read as it happens, tagged client = mcp.

Other clients

The same endpoint works with any MCP client (Codex, Claude Desktop, Cursor). See Connecting agents for the config formats.