BigHugger
GH Repository · msoedov

agentic_security

Agentic LLM Vulnerability Scanner / AI red teaming kit 🧪

stars
2,001
30-day movement
starts with the next reading
Related entries
60
Connections
1
Pythonpythonllm-guardrailsllm-securityllm-evaluationpython/poetryllm-evaluation-frameworkllm-jailbreaksllm-fuzzerllm-scannerllm-fuzzer-aggregatorllm-fuzzingdockerai-red-teamllm-vulnerabilitiesprompt-testingagent-frameworkagent-security

agentic_security is a Python-based LLM vulnerability scanner and AI red teaming kit. Its topics indicate it covers jailbreaks, fuzzing, guardrail testing, and prompt evaluation for LLM and agent security.

Reach for it when you want to fuzz-test or red-team LLM applications for prompt injection and jailbreak vulnerabilities.

Use it to

  • Fuzz LLM endpoints for jailbreak vulnerabilities
  • Test prompt guardrails against adversarial inputs
  • Run red-team evaluations on agent pipelines
  • Aggregate fuzzing results across test suites

For Security engineers and ML teams testing LLM systems

Role
agent-framework
Language
Python
Licence
Apache-2.0
Forks
288
Open issues
40
Last push
2026-09-11
Latest release
0.1.1 · 2024-04-26
topicsllm-securityred-teamingfuzzingjailbreaksprompt-testingagent-security