Connect your AI

Let your agent find research, check evidence, and track changes.

Copy. Paste into your AI.

The prompt guides setup, then requests research leads with sources.

Preparing prompt…

View full prompt
Help me start industry research with EquityLayer. First check whether EquityLayer MCP is connected and inspect the available tools and their parameters. If it is not connected, use this HTTP MCP endpoint: . If your environment supports MCP setup, help add it under the client's rules. Otherwise, give me the shortest setup instructions and wait for connection before continuing. Do not report a successful connection without verification. A localhost or 127.0.0.1 endpoint works only on the same computer. For a remote environment, use the public setup instructions at https://equitylayer.io/docs. Once connected, use discover_theses to explore AI Compute. Return up to 3 research candidates. For each, explain why it matters, cite sources and dates, identify unknowns, and suggest what to track next. State missing coverage. Answer in English and preserve company tickers and source links. For each candidate, return a Dashboard link by combining  with its dashboard_path. If dashboard_path is absent, use /dashboard?research= followed by the URL-encoded thesis_id. The user can open the latest public evidence and save it for tracking.
Manual setup & service check

Add EquityLayer MCP

Add this setup to your selected AI client.

Discoverdiscover_theses
Verifytest_thesis
Trackget_research_update

Check that this site's MCP responds. Verify your client connection in your AI.

Call get_research_update with {"theme_id":"ai_compute"}. Save caller_action.baseline in my workspace. Call the tool again with that baseline. Report the result_state and source dates.
For Claude, Codex, and Hermes

Connect the curated research harness.

Connect once. EquityLayer helps your agent use available financial data, check what changed, and keep the view current.

00 / Connect

Connect once. Keep the research state.

  1. Add EquityLayer as the research layer.
    claude mcp add --transport http equitylayer https://equitylayer.io/mcp

    Codex, Hermes, and other MCP clients: point them at https://equitylayer.io/mcp (HTTP transport).

  2. Give your agent the setup prompt.
    Set up EquityLayer as the research layer for this workspace.
    
    Use this remote MCP server:
    - Name: equitylayer
    - URL: https://equitylayer.io/mcp
    - Transport: HTTP
    
    If you can add MCP servers, add this server now. Ask for approval if this client requires it.
    If you cannot add MCP servers, show the exact command or configuration for this client. Then wait for me.
    
    After the MCP connection works, create a local research view and show the first useful update.
    
    1. Inspect the financial MCP servers, data sources, and skills that are available in this workspace.
    2. Do not add a source or access private data without my permission.
    3. Never send private or portfolio data to EquityLayer.
    4. Start with my company, theme, or investment view. Ask one question if the scope is unclear.
    5. Use EquityLayer for research context, source rules, and saved research state.
    6. Use an available connector only when it matches the required evidence and its terms allow the use.
    7. Prefer approved official sources. Record the source, date, and reporting period.
    8. Separate reported facts, management claims, inference, and missing evidence.
    9. Compare new evidence with the previous investment view.
    10. Show what changed, why it matters, affected companies, counterevidence, and the next check.
    11. Save the research view and its next trigger in this workspace. Use the returned update tool for later checks.
    
    Do not ask for email, account, broker, wallet, or execution access.
    If the MCP is unavailable, say so and do not fabricate results.
  3. Verify the connection, then start research.

    Call get_service_status to check access and language rules. A copied URL does not confirm a successful connection.

    Your agent uses approved sources and the financial connectors already in its workspace. It returns source-linked changes, affected companies, and the next check.

Machine-readable index: /llms.txt. Everything below is the reference contract behind those three tools.

01 / Principles

Models change. Research contracts should persist.

A useful capability has a bounded job, explicit inputs, structured outputs, source provenance, a data-as-of timestamp, known limitations, and a maintainer. MCP may be the interface, but the managed research service is the product.

Every entry in the source index is published against this contract, and every research note shows how sources become dated, testable claims.

02 / Research object

Suggested output envelope

{
  "question": "What changed since the prior quarter?",
  "data_as_of": "ISO-8601 timestamp",
  "facts": [{ "claim": "...", "source": "..." }],
  "inferences": [{ "claim": "...", "confidence": 0.72 }],
  "assumptions": ["..."],
  "risks": ["..."],
  "invalidation_conditions": ["..."],
  "capability_version": "semis-earnings@0.3.1"
}
03 / Quality signals

Trust should be inspectable.

The first version should avoid one composite “accuracy” score. Show separate signals for source coverage, point-in-time integrity, schema reliability, latency, cost, maintenance recency, benchmark task results, and user reuse.

  • Never infer commercial rights from “open source” alone.
  • Separate data-provider terms from repository license.
  • Version reputation by capability release, not only creator.
  • Evaluate bounded tasks before subjective full reports.