About Tawa Guard
Tawa Guard is the security gate between your AI agents and everything they read. We scan tool output, retrieved documents, web pages and email for prompt injection, and return a decision your code acts on: allow, review or block.
What we do
An agent that reads content it does not control can be told what to do by that content. We built Tawa Guard to flag that before the agent acts on it. You call POST /v1/scan or use the Python or TypeScript SDK, and your code decides what to do with the result.
The Developer plan is free every month with no card. See pricing or read the docs.
How we work
We publish how we measure. Every benchmark number we quote comes with its dataset, its operating point and its date, and we say where the comparison is not in our favour. Scans are processed in the UK and kept only for the retention window you set. Read the research.
The company
- Legal name
- Tawa Guard Ltd
- Registered in
- England and Wales
- Company number
- 17425810
- Registered office
- 66 Paul Street, London, England, EC2A 4NA
- Support
- Help centre or support@tawaguard.com
- General enquiries
- contact@tawaguard.com