web-research

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs web research using a standard search tool with a strictly enforced cap of 10 queries, preventing excessive resource usage or unintended crawler behavior.
  • [SAFE]: It implements a source hierarchy (Tier 1-4) to ensure data quality, prioritizing official and peer-reviewed sources over unverified blogs or social media.
  • [SAFE]: The workflow requires cross-validation (at least 2 independent sources) and confidence declarations for factual claims, which mitigates the risk of propagating misinformation or search engine hallucinations.
  • [SAFE]: Operations are confined to local project directories (tmp/research/, docs/knowledge/), and no sensitive file access, credential usage, or network exfiltration patterns were identified.
  • [SAFE]: The chaining mechanism (--chain=deep-dive) allows for modular execution by reading local instruction files, following standard platform patterns for skill orchestration without remote code execution risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 10:11 AM