DebtStack AI

Provides an MCP server to access DebtStack AI corporate credit data for AI agents and workflows.
  • python

0

GitHub Stars

python

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "debtstack-ai-debtstack-python": {
      "command": "debtstack-mcp",
      "args": [],
      "env": {
        "DEBTSTACK_API_KEY": "your-api-key"
      }
    }
  }
}

You can run the DebtStack AI MCP server locally and connect Claude, Cursor, or other MCP clients to access comprehensive corporate credit data. This server exposes the DebtStack data and tools to your AI agents, enabling fast, structured queries and seamless integration into your workflows.

How to use

You install and run an MCP server instance locally, then configure your MCP client to connect via the provided stdio commands or the Python module option. Once connected, you can perform cross-company credit searches, drill into bonds, map guarantor structures, and retrieve pricing, documents, and covenant information. Use the MCP tools to ask natural-language questions or execute scripted queries, and let the server return structured results in milliseconds.

How to install

Prerequisites you need before starting: Python and pip installed on your system.

pip install debtstack-ai[mcp]

Additional configuration and usage notes

The MCP server supports multiple local configurations. You can run the server using a Python module or connect via dedicated desktop code clients. The following configurations are provided as examples to enable quick setup for Claude Desktop, Claude Code, and Cursor clients, as well as a Python module option.

{
  "mcpServers": {
    "debtstack-ai": {
      "command": "debtstack-mcp",
      "env": {
        "DEBTSTACK_API_KEY": "your-api-key"
      }
    }
  }
}

What to run in your MCP client per environment

Use the following MCP configurations exactly as shown to connect Claude Desktop, Claude Code, and Cursor, or to run the server via the Python module. Each entry includes the required command or program and environment variables.

{
  "mcpServers": {
    "debtstack-ai": {
      "command": "debtstack-mcp",
      "env": {
        "DEBTSTACK_API_KEY": "your-api-key"
      }
    }
  }
}

How the MCP server is started from different clients

Claude Desktop, Claude Code, and Cursor can all run the MCP server locally using a standard executable and the API key provided by you. If you prefer a Python-based startup, you can run the MCP server through Python’s module entry point.

{
  "mcpServers": {
    "debtstack-ai-python": {
      "command": "python",
      "args": ["-m", "debtstack.mcp_server"],
      "env": {
        "DEBTSTACK_API_KEY": "your-api-key"
      }
    }
  }
}

Security and access control

Keep your API key secure. Do not commit it to public repositories or share it in unsecured channels. Use per-user keys where possible and rotate keys periodically.

Examples of common queries you can run via MCP

After connecting, you can perform typical queries such as searching for high leverage companies, drilling into bonds to view pricing and covenants, and mapping guarantor chains. The server exposes functions that let you filter by sector, yield, maturity, seniority, and pricing, and to retrieve full document text from SEC filings when needed.

Troubleshooting

If you cannot connect, verify the API key is correct and the MCP server process is running. Check client logs for connection errors and ensure network access between the client and the local MCP server. If the server reports missing environment variables, confirm that the key is provided in the environment block for the respective configuration.

Notes on language and tooling

This server is intended for Python-centric environments and can be extended to work with various MCP clients. You can install the MCP package with the Python extras for MCP support and, if needed, connect to clients using the provided stdio commands or the Python module startup path.

What’s included and capabilities

The MCP server provides programmatic access to DebtStack AI data, including search capabilities for companies and bonds, resolution of identifiers, traversal of guarantor networks, pricing data, document search, and tracking changes in debt structures over time.

Available tools

search_companies

Screen companies by leverage, sector, coverage ratios, and risk flags.

search_bonds

Filter bonds by yield, spread, seniority, and maturity.

resolve_bond

Look up a bond by CUSIP, ISIN, or a description.

get_guarantors

Find all entities that guarantee a bond.

get_corporate_structure

Get the full parent-subsidiary hierarchy for a company.

search_pricing

Retrieve FINRA TRACE bond prices, yields, and spreads.

search_documents

Search SEC filing sections and credit documents.

get_changes

See changes in a company's debt structure over time.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational