At a glance
- What the listing says
- A free, open-source browser extension that red-teams AI systems. An autonomous agent tests a chat UI in a browser tab or an API endpoint, scores findings deterministically, maps them to the OWASP Top 10 for LLM Applications, and exports HTML or JSON reports.
- Pricing
- Listed as Open source. For current plans, see the Mirage by KageX site.
- Category
- Automation & Agent Builders
Pros and cons
Pros
- Free and open source under the MIT license.
- Tests a chat UI in your active tab or a configured API endpoint, with the same attacks on either.
- Scoring is deterministic: a win needs a real decoded side-effect, not an LLM grading itself.
- Findings map to the OWASP Top 10 for LLM Applications, and reports export as HTML or JSON with full transcripts.
- Works with Claude, OpenAI, Grok and local Ollama models, as the attacker/judge or as the target.
Cons
- You load it as an unpacked Chrome extension from a GitHub release or a local build.
- You provide your own model API keys, and usage is billed by the provider.
- Only run it against systems you own or are explicitly authorized to test.
Sources, checked Nov 11, 2026: kagex.ai/products/mirage · github.com/kagexai/mirage · kagex.ai/blog/mirage-red-team-agent. Plans and prices change, so confirm on the vendor's site.
Other Automation & Agent Builders
zapier.com
No-code automation across thousands of apps, with AI agents.
n8n.io
Fair-code workflow automation with AI agent nodes, self-hostable.
relevanceai.com
Platform to build and run teams of AI agents.
Rising in Automation
Build Mirage by KageX? You can add this Rising badge to your site or README.
Is this your tool? Suggest an edit