Classify security scanner findings as false positives or real vulnerabilities.
npx skills add https://github.com/NOMARJ/sigil --skill fp-check-nomarj
Please help me install this Agent Skill. Skill: fp-check Source: https://github.com/NOMARJ/sigil/tree/main/packs/security/skills/owasp-security/fp-check Command: npx skills add https://github.com/NOMARJ/sigil --skill fp-check-nomarj
This Skill enforces a mandatory false-positive gate for security scanner findings so teams can separate noise from actionable vulnerabilities and avoid alert fatigue.
Use fp-check to triage the latest security scanner findings and produce a FP-CHECK TRIAGE summary.
High-intent search queries and answers about installing and using this skill.
To filter false positives from security scanner findings, apply hard exclusion rules for low-signal categories like DOS and test files, then use confidence scoring with defined thresholds to gate noise from actionable vulnerabilities.
The best way to triage automated scan outputs across repositories is to enforce a step-by-step protocol mandating reproduction attempts, TP/FP/AR classification, documented justifications, and allowlist entries with scheduled review dates.
Confidence scores in vulnerability triage work by assigning a 1-10 rating to each finding, using an explicit threshold to determine whether scanner results require manual review or can be automatically logged as false positives.
Yes, you can automatically exclude prompt injection findings from CI scan results by applying hard exclusion rules that classify common low-signal categories, including prompt injection, docs, and safe-language memory issues, as auto-excluded false positives.
You should use an allowlist when triaging scanner vulnerabilities to prevent reflagging of documented noise, ensuring that previously reviewed false positives include review dates and justifications to maintain signal quality over time.
aist-ai-sastClassify AIST pipeline findings as True Positive or False Positive with evidence.
ElbertePlinioValidate code scan findings and generate structured JSON with exploitability details.
icartshAnalyze .NET solutions for code quality, security, and dependency issues.
UnicornGenerate Prometheus alert rules and PagerDuty routing configurations.
