[Feature]: Add support for Superagent Firewall Guardrail #14664
Description
The Feature
Superagent just launched a safety model designed for coding tasks. Would be cool to support in the guardrails of LiteLLM
Motivation, pitch
Code agents are especially risky since they can install packages, edit CI workflows, or touch sensitive files. A firewall guardrail would help catch things like prompt-injection, typosquatted dependencies, unsafe shell commands, or PRs that modify workflows/secrets.
Superagent Firewall is purpose built to identify vulnerabilities for ai generated code.
LiteLLM is hiring a founding backend engineer, are you interested in joining us and shipping to all our users?
No
Twitter / LinkedIn details
@pelaseyed