@xu-c0/cybersec-mcp

MCP server with 323 cybersecurity prompts and 7 chained workflows. Lets MCP-compatible clients (Claude Desktop, Claude Code, Cursor) call cybersecurity tools instead of asking users to copy-paste prompts.

This is the npm package. The main repo lives at github.com/xu-c0/cybersec-mcp.

Install

npx -y @xu-c0/cybersec-mcp

Claude Desktop / Claude Code / Cursor — paste into the relevant mcp.json:

{
  "mcpServers": {
    "cybersec": {
      "command": "npx",
      "args": ["-y", "@xu-c0/cybersec-mcp"]
    }
  }
}

Client-specific paths in examples/.

Tools

| Tool | What it does | |---|---| | cybersec_list_categories | Lists the 8 prompt categories with counts. Call first to orient. | | cybersec_list_prompts | Filters prompts by category, skill level, or free-text search. Returns up to 50 summaries. | | cybersec_get_prompt | Fetches a single prompt by id, substituting variables. Reports unfilled variables. | | cybersec_list_scenarios | Lists the 7 scenario workflows and their variables. | | cybersec_get_scenario | Returns a full workflow with every phase's prompt customized to the supplied variables. |

Categories

red-team · blue-team · soc-ops · cloud-security · osint · grc · vuln-analysis · ai-security

Full breakdown and MITRE ATT&CK mapping in the main README and ATTACK_MATRIX.md.

Build from source

git clone https://github.com/xu-c0/cybersec-mcp.git
cd cybersec-mcp/mcp
npm install
npm run build
node dist/index.js

npm run sync-data regenerates src/data/*.json from web-app/js/*.js — run after editing prompts.

License

MIT. For authorized security testing, defensive operations, security research, and education only.