MCP Quickstart

By the end of this page InvesTeam's analysis is available as tools inside your AI client — so you can say "create an investment brief on X" in Claude and watch it run. First success is a tool call that returns a session_id. The output is decision-support analysis, not investment advice.

How do I get a key for MCP?¶

You use the same hfk_… programmatic API key as the HTTP API; the MCP server forwards it to the API. You need the pipeline:write scope to create a brief and pipeline:read to poll it. See Authentication & API Keys.

How do I connect InvesTeam to Claude?¶

Add the InvesTeam MCP server to your client's config with your key. The block below is a single-language JSON config — it is not affected by the code-language toggle.

{
  "mcpServers": {
    "investeam": {
      "url": "https://mcp.investeam.io",
      "headers": { "Authorization": "Bearer hfk_your_key_here" }
    }
  }
}

How do I verify it works?¶

Prompt the client to create a brief. The create_brief tool returns a BriefResponse with a session_id and a status — that returned id is your first success.

# Conceptually, the tool call the agent makes maps to this HTTP request:
curl -sS https://investeam.io/api/v1/briefs \
  -H "Authorization: Bearer hfk_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{"input": "Is NVDA a buy right now?"}'

# From an MCP-capable client the same call is a tool invocation:
#   create_brief(input="Is NVDA a buy right now?")
# which returns {"session_id": "...", "status": "needs_clarification" | "completed" | "rejected"}

Then let the agent poll get_brief every two to three seconds until the status is completed or rejected.

What's next?¶

Point the agent at the full MCP Quickstart loop and, for cadence and terminal states, the agent-patterns guide. Prefer plain HTTP? See the API Quickstart.