uninhibited-scholar/Agent Redteam Scan

AI agent security red-team scanner for CI — OWASP LLM Top 10 coverage, SARIF output, PR comments. Like `npm audit` for AI agents.

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
modelModel ID to test (e.g. gpt-4o, glm-4-plus, llama3)yes
targetTarget: openai (default), claude, zai, ollama, deepseek, azure, qwen, localnoopenai
api-keyAPI key (use secrets!). Not needed for ollama/local.no""
base-urlCustom API base URLno""
fail-belowFail CI if score below this (0-100, 0 = no gate)no0
suitesComma-separated suites (default: all 13)no""
limitMax samples per suite (default: 20)no20
max-tokensMax response tokens (default: 500)no500
workersParallel API calls (default: 4)no4
max-attemptsMaximum total attempts per model call (1-10, default: 3)no3
upload-sarifUpload SARIF to GitHub Security tab (default: true)notrue
pr-commentPost results as PR comment (default: true)notrue
namedescription
scoreOverall security score (0-100)
total-samplesTotal samples tested
total-failedTotal failed samples
total-errorsTotal samples without a security verdict
run-statusExecution health: complete, incomplete, or no_data
sarif-filePath to SARIF file