Drillr - The Financial MCP for AI Agents MCP Server

Content

Drillr is a financial research MCP server for AI agents. It exposes a single Streamable HTTP endpoint for standardized financial data, SEC filing search, company discovery, live market signals, ticker resolution, table discovery, schema inspection, and fiscal-period utilities.

The MCP endpoint is:

https://gateway.drillr.ai/mcp/data

Drillr uses bearer API-key authentication. The public repository includes setup docs, tool references, REST API docs, and Claude Code plugin metadata.

Source Review

• https://github.com/Little-Grebe-Inc/drillr-mcp-server
• https://github.com/Little-Grebe-Inc/drillr-mcp-server/blob/main/docs/tools.md
• https://github.com/Little-Grebe-Inc/drillr-mcp-server/blob/main/docs/rest-api.md
• https://drillr.ai/developer/docs
• https://modelcontextprotocol.io/registry/about
• https://code.claude.com/docs/en/mcp

These sources were reviewed on 2026-06-05. Prefer live Drillr docs over model memory for endpoint behavior, tool coverage, API-key requirements, credit usage, data coverage, and REST/MCP response details.

Features

• Streamable HTTP MCP endpoint with bearer API-key auth.
• MCP tools backed by corresponding REST API endpoints.
• Structured financial data over 90-plus tables, including statements, ratios, earnings, insider and ownership data, prices, and alternative datasets.
• Paragraph-level semantic search over indexed SEC filings.
• Filing list lookup by ticker, form type, and date range.
• Company discovery from natural-language descriptions.
• Cross-asset signal feed for recent market events.
• Ticker resolution from company names, brands, or ticker substrings.
• Alternative-data table discovery and schema inspection.
• Fiscal-period helper for companies with non-calendar fiscal years.

Tools

• run_sql: read-only PostgreSQL SELECT queries over structured financial, market, and alternative-data tables.
• sec_report_search: paragraph-level semantic search across SEC filings such as 10-K, 10-Q, 20-F, 6-K, S-1, and DEF 14A.
• sec_report_list: list indexed filings for a ticker by form type and date range.
• company_search: discover companies by business model, supply chain, competitors, industry, or thematic fit.
• signal_list: retrieve recent news and market-event signals by ticker, sector, or time window.
• ticker_resolve: map company names, brands, or ticker fragments to canonical tickers before calling ticker-keyed tools.
• list_tables: discover available alternative-data SQL tables by category.
• get_table_schema: inspect columns and types for a selected table.
• fiscal_utility: convert fiscal years and quarters to calendar months, or resolve calendar months back to fiscal periods.

Installation

1. Sign up at https://drillr.ai.
2. Create an external-scope API key at https://drillr.ai/developer/keys.
3. Store the key as DRILLR_API_KEY or paste it only into the target MCP client's secure configuration.
4. Add the MCP endpoint with an Authorization: Bearer ... header.

Claude Code

export DRILLR_API_KEY="drl_..."

claude mcp add drillr https://gateway.drillr.ai/mcp/data \
  --transport http \
  --header "Authorization: Bearer $DRILLR_API_KEY"

Generic MCP Client

{
  "mcpServers": {
    "drillr": {
      "type": "http",
      "url": "https://gateway.drillr.ai/mcp/data",
      "headers": {
        "Authorization": "Bearer ${DRILLR_API_KEY}"
      }
    }
  }
}

Smithery

Drillr also documents a Smithery install path:

npx -y @smithery/cli install drillr/drillr --client claude

Smithery prompts for the drl_* API key and writes client configuration during installation. Review the generated config before using it in a shared machine or team environment.

Claude Code Plugin

The Drillr repository also acts as a Claude Code plugin marketplace:

/plugin marketplace add Little-Grebe-Inc/drillr-mcp-server
/plugin install drillr

After plugin installation, set DRILLR_API_KEY in the environment used by the host or add the key through the generated configuration flow.

Use Cases

• Compare company margins, ratios, growth, valuation, or price history across tickers.
• Search SEC filings for specific risk factors, accounting policies, segment commentary, related-party transactions, or guidance language.
• Resolve a company name to a canonical ticker before querying ticker-keyed data.
• Discover public companies by business description, supply-chain role, peer group, or investment theme.
• Inspect alternative datasets such as data centers, semiconductors, compute pricing, AI model benchmarks, macro, trade, contracts, patents, papers, or developer-skill demand.
• Pull recent market signals for specific tickers or sectors.
• Convert fiscal-period language into calendar ranges before querying company-reported data.

Coverage And Limits

Drillr's public docs describe coverage for US and Japan equities, financials back to the 1980s, SEC filings, earnings transcripts and summaries, markets, analyst coverage, news and signals, and AI value-chain alternative data.

Documented out-of-scope areas include private or unlisted companies, on-chain crypto metrics, options chains, real-time order books, intraday tick data, retail brokerage actions, and Drillr-produced price forecasts.

Safe Usage

• Run ticker_resolve before ticker-keyed tools when the user gives a company name, brand, or ambiguous ticker.
• Use list_tables and get_table_schema before writing SQL against unfamiliar alternative-data tables.
• Keep SQL narrowly filtered by ticker, date, table, and row limit.
• Prefer sec_report_search for narrative SEC filing content and run_sql for structured financial values.
• Compare cited paragraphs, table definitions, and source dates before including results in investment memos, diligence reports, or client-facing material.
• Monitor credit balance and avoid automated loops that repeatedly run broad SQL, SEC search, or signal queries.

Troubleshooting

The MCP client cannot connect

Confirm the URL is https://gateway.drillr.ai/mcp/data, the transport is HTTP, and the config includes Authorization: Bearer <YOUR_DRILLR_API_KEY>.

Authentication fails

Check that the key is an external-scope Drillr key and has not been copied with angle brackets, extra spaces, or shell-escaping errors.

SQL queries fail

Use list_tables and get_table_schema first. Drillr documents SQL constraints for run_sql, including read-only SELECT usage and table-specific filtering requirements.

Results do not match the company the user intended

Use ticker_resolve before searching filings or structured financial data. Some brand names, ADRs, indexes, ETFs, and foreign listings need explicit disambiguation.

Credit usage is higher than expected

Narrow the query, lower result limits, and avoid repeated broad searches. Drillr's REST responses document credit metadata, while MCP responses follow standard JSON-RPC and do not include per-call credit details inline.