DEDUPEDauthorization. Red teaming sends adversarial inputs; run it only against systems you own or have written permission to test.
Who it is forA team about to launch a customer-facing chatbot or agent that wants to find weak spots first.
The problemPrompt injection and data leaks are easy to miss in normal testing.
The outcomeA vulnerability report for your own app that shows where it fails and what to fix.
What you need first
- Node.js for npm, or Homebrew, or pip; npx works without installing
- API keys for the LLM providers you test, set as environment variables
- Only run red teaming against apps you own or are authorised to test
Steps
- Install Promptfoo with
npm install -g promptfoo - Follow the Red Teaming guide linked in the README to configure a scan of your app
- Run the scan only against apps you own or are authorised to test
- Review the security vulnerability report and fix the failing cases
- Re-run the scan after each fix
Prompt for your agent
Using promptfoo/promptfoo, follow the Red Teaming guide to scan my own app at <endpoint>, which I own. Summarise the vulnerability report by severity and propose a fix for the top three issues.
Caveat. The README links to the Red Teaming guide for configuration rather than showing it.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for promptfoo/promptfoo and "Red team your own LLM app for vulnerabilities"
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer