second-opinion-review
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows a structured, multi-step workflow focused on text analysis and verification. It includes specific guardrails against following instructions embedded in untrusted user data (Section 0 of SKILL.md).
- [SAFE]: External dependencies are restricted to standard Python libraries (argparse, datetime, json, re, sys). The provided
scripts/claims_scan.pyscript performs local text processing with no network access or file-writing capabilities. - [SAFE]: Remote operations are limited to standard platform tools (
web_search,web_fetch) for factual verification, governed by strict source-quality hierarchies and human-in-the-loop (HILT) principles. - [SAFE]: The skill implements a 'Zero-Assumption Rule' and mandatory evidence chains for all factual claims, requiring primary sources or explicit placeholders for unverified data.
- [SAFE]: Instruction overrides and safety guidelines are explicitly addressed, treating all external input as data rather than instructions.
Audit Metadata