Safe Labs AI Inc.
AI agent red-teaming and evaluation. Open-source tools for testing LangGraph, CrewAI, AutoGen, and custom AI agents against OWASP ASI01–ASI10.
- 53 followers
- United States of America
- https://agentsafelabs.com
Pinned Loading
Repositories
Showing 5 of 5 repositories
- safelabs-eval Public
Red-teaming and evaluation framework for AI agents, built around an OWASP-inspired ASI01–ASI10 taxonomy
- safelabs-trace Public
Trace event model, frozen severity tagger and action-aware metrics for tool-using LLM agents, with the runner, analysis and human checks of a pre-registered study of what text-only scoring misses. Source available for research reproducibility; all rights reserved.
- agentport-bench-results Public
- safelabs-research Public
Safe Labs AI research — attack taxonomy, advisories, and blog posts on AI agent security.
Top languages
Loading…
Most used topics
Loading…