Find security tools with your AI assistant
CyberAgents Exchange publishes all 115 of its agents, skills, MCP servers and playbooks in a form AI assistants can read. Copy one prompt into Claude, ChatGPT, Copilot or Cursor, answer a few questions about your work, and get recommendations that fit it.
Start here
You're helping me find tools on CyberAgents Exchange, a directory of open-source security AI agents, skills, MCP servers and playbooks. First, read https://exchange.tenable.com/llms.txt so you know how the directory's data is laid out. Don't recommend anything yet. Then interview me, one question at a time, until you know: - my role and what I'm trying to get done - the security tools and platforms I already use - the AI assistant or coding tool I'd run it in - whether I need vetted listings only Ask only what you need, usually three or four questions. Then fetch the listings index, shortlist what fits, and open each candidate's detail record before recommending it. For each recommendation, tell me what it does, why it fits my situation, its trust tier, and the link to its Exchange page. If nothing fits, say so.
Paste it into any assistant that can open a web page, press Enter, and answer its questions. If yours can't open links, turn on its web access or browsing first.
Keep going
Once it has recommended a few listings, carry on the conversation. For example:
- "Which of these work with Claude Code?"
- "Is there an MCP server that does the same job?"
- "Only show me vetted listings."
- "Walk me through installing the first one."
Before you install
The Exchange is a directory. Each listing's code lives in its author's GitHub repository and runs with whatever access you give it, so read it before you install it, as you would any open-source tool. Every listing carries a trust tier, and the prompt asks your assistant to tell you which:
- Contributed — passed automated validation and a baseline maintainer review.
- Vetted — passed the Exchange Inspector security review process.
For developers
Contribute a listing
Built an agent, skill, MCP server or playbook? The Submission Builder skill checks your repository and opens the pull request for you.
Prefer to do it by hand? The contributing guide covers the frontmatter for each listing type and the review process.
Build on the data
An index of every listing, with each row linking to its full record:
https://exchange.tenable.com/api/listings.json- Agent
https://exchange.tenable.com/api/agents/<slug>.json - Skill
https://exchange.tenable.com/api/skills/<slug>.json - MCP server
https://exchange.tenable.com/api/mcp-servers/<slug>.json - Playbook
https://exchange.tenable.com/api/playbooks/<slug>.json
CORS is open. Every field and vocabulary is documented in /llms-full.txt; /llms.txt is the short version.