FastMCP FastMCP
Sign In
A

Adversary MCP Server

A security-focused server that integrates with Cursor IDE to provide real-time vulnerability detecti...

11 views
0 installs
Updated Feb 2, 2026
Not audited
Security Developer Tools
Tools I Recommend
ClaudeKit ClaudeKit Sponsor
A security-focused server that integrates with Cursor IDE to provide real-time vulnerability detection, exploit generation, and security insights during software development.

  1. Open the FastMCP connection interface

    • Click the "Install Now" button in your FastMCP UI to open the connection / environment variables form.

  2. Prepare to add the LLM provider keys (these are the primary ENVs you will add)

    • The Adversary MCP server uses these environment variable names in its CLI and MCP integration:
      • OPENAI_API_KEY
      • ANTHROPIC_API_KEY
    • (Optional/secondary ENVs you can add if needed)
      • ADVERSARY_LOG_LEVEL (e.g., INFO)
      • ADVERSARY_WORKSPACE_ROOT (path to override workspace root)

  3. Create / copy an OpenAI API key

    1. Sign in or sign up at the OpenAI Platform (platform.openai.com).
    2. Open the API keys page (Dashboard → API keys) and click "Create new secret key".
    3. Copy the generated secret immediately (it is shown only once). (platform.openai.com)

  4. Paste the OpenAI key into FastMCP

    1. In the FastMCP connection form, add an ENV with name OPENAI_API_KEY.
    2. Paste the secret key as the value.
    3. Save/confirm the connection entry in the FastMCP interface.

  5. Create / copy an Anthropic (Claude) API key

    1. Sign in or sign up to the Anthropic Console.
    2. Generate an API key from the Console (the Anthropic docs show creating and exporting the key as ANTHROPIC_API_KEY).
    3. Copy the generated key immediately (store it securely; the Console shows keys only at creation). (platform.claude.com)

  6. Paste the Anthropic key into FastMCP

    1. In the FastMCP connection form, add an ENV with name ANTHROPIC_API_KEY.
    2. Paste the Anthropic key as the value.
    3. Save/confirm the connection entry in the FastMCP interface.

  7. (Optional) Add the other Adversary ENVs in FastMCP

    • If you want logging or workspace overrides, add:
      • ADVERSARY_LOG_LEVEL = e.g., INFO
      • ADVERSARY_WORKSPACE_ROOT = full path to your workspace root
    • Add each as a separate ENV entry in the FastMCP form and save.

  8. Verify the MCP server picks up the keys

    • After saving ENVs in FastMCP, run the Adversary CLI or MCP status check from your environment (or via the MCP tool) to verify the LLM is configured:
      • Example CLI checks: adv configure --llm-provider openai --llm-api-key $OPENAI_API_KEY or adv status
    • Run a quick scan (e.g., adv scan-file path/to/file.py --use-llm --use-validation) to ensure the LLM integration is active.

  9. Security & best practices

    • Treat API keys as secrets: do NOT commit them to source control.
    • Store long-term secrets in a secrets manager (AWS Secrets Manager, HashiCorp Vault, 1Password/Bitwarden, etc.) and only paste into FastMCP when needed.
    • If a key is ever exposed, revoke it in the provider console and generate a new one immediately. (platform.openai.com)

  10. If you need help

    • Consult the Adversary README/CLI (adv configure setup and adv status) for configuring providers and confirming settings, then re-check the FastMCP ENVs if the server reports missing credentials.

Quick Start

View on GitHub

More for Security

View All →

More for Developer Tools

View All →

Similar MCP Servers

Report Issue

Thank you! Your issue report has been submitted successfully.