Validate security findings through multi-round self-challenge and arbiter verification.
npx skills add https://github.com/avelikiy/great_cto --skill skeptical-triage
Please help me install this Agent Skill. Skill: skeptical-triage Source: https://github.com/avelikiy/great_cto/tree/main/skills/skeptical-triage Command: npx skills add https://github.com/avelikiy/great_cto --skill skeptical-triage
This Skill helps mitigate false positives in code reviews and security audits by applying a three-round self-challenge and arbiter process, reducing CTO time waste and improving overall confidence in findings.
Use the skeptical-triage skill to validate a finding in the codebase.
High-intent search queries and answers about installing and using this skill.
False positives in security audits are reduced by applying a three-round skeptical triage process that challenges findings on reachability and defenses before an arbiter issues a final impartial verdict.
The arbiter pattern for triaging security findings provides a final, impartial verdict after a multi-round self-challenge, resolving conflicting signals to validate or invalidate potential vulnerabilities before they block a release gate.
Validate a flagged vulnerability before blocking a deployment gate by conducting an in-depth skeptical triage that checks reachability, missed angles, and existing defenses, yielding a confidence-backed decision to proceed or block.
Skeptical triage applies to QA regression checks alongside security audits by using a multi-step self-challenge and arbiter verification to confirm whether flagged issues are genuine defects or false alarms.
Multi-round self-challenge code reviews require standard analysis tools like Read, Grep, Bash, and Glob to inspect the codebase and verify the reachability and context of flagged findings.
The arbiter pattern for security finding validation requires multiple analysis rounds, meaning it is designed for in-depth verification of critical blocking issues rather than rapid high-volume scanning of every minor code review flag.
fanguyunIdentify security vulnerabilities and generate structured audit reports.
ElbertePlinioScan code changes and repositories for security vulnerabilities using STRIDE threat modeling.
eliottbusinessAutomate pre-commit code review with static analysis and security scanning.
conorbronsdonCoordinate multi-agent pull request reviews with parallel adversarial and operational analysis.
