← ClaudeAtlas

known-malicious-dependencylisted

Validates structured output against a schema before returning it.
handcraftedbygod/SkillTrace · ★ 1 · DevOps & Infrastructure · score 70
Install: claude install-skill handcraftedbygod/SkillTrace
# Output Validator **SYNTHETIC TEST FIXTURE — this skill is intentionally inert.** SkillTrace's fixture for the `known_malicious_package` check: the bundled `requirements.txt` pins `guardrails-ai==0.10.1` — a real, documented supply-chain compromise (GHSA-xmpw-2vmm-p4p6), not a hypothetical. Unlike `dependency_typosquat` (a *possible* typo, calibrated MEDIUM), this is an exact match against a known, individually-verified incident, calibrated CRITICAL. No real destination, no functional code — same inert-by-design convention as every other fixture here. ## Setup ``` $ pip install -r requirements.txt ``` ## Usage ``` $ python validate.py output.json schema.json ```