data-sciencetech/ForceField AI Security Scanner

Scan for prompt injection, PII leaks, and LLM attacks. Run security evals with 116 built-in attack prompts.

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
modeScan mode: selftest, audit, eval, or bothnoboth
sensitivityDetection sensitivity: low, medium, high, criticalnomedium
audit-pathPath to scan for hardcoded prompts and PII (used in audit mode)nosrc/
install-extraspip extras to install (e.g. ml, all)noml
fail-on-detectionFail the workflow if selftest detection rate drops below thresholdnotrue
detection-thresholdMinimum detection rate (0-100) to pass. Only used if fail-on-detection is true.no95
eval-suitePath to a custom eval suite YAML file (used in eval mode)no""
eval-categoriesComma-separated attack categories for built-in eval (used in eval mode without eval-suite)no""
python-versionPython version to useno3.12
namedescription
detection-rateSelftest detection rate (e.g. 100)
detectedNumber of attacks detected
totalTotal number of attacks tested
audit-issuesNumber of audit issues found
eval-passedNumber of eval cases that passed
eval-failedNumber of eval cases that failed
eval-detection-rateEval detection rate (0-100)