research-powerpack
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists exclusively of Markdown documentation and reference files. It does not include any scripts, binaries, or automated installation commands that could lead to unauthorized code execution.
- [SAFE]: No hardcoded credentials, sensitive file paths, or suspicious network operations were identified. Instructions for tool usage refer only to well-known research platforms (Google, Reddit) and official documentation sources.
- [SAFE]: There is no evidence of prompt injection, role-play bypasses, or instructions to ignore safety guidelines. The skill uses instructional language consistent with its primary purpose of enhancing agent research capabilities.
- [SAFE]: The skill correctly addresses the risk of processing untrusted external data (Category 8) by advising the agent to cross-verify claims across multiple tools and by using structured extraction targets in scraping tools to limit the ingestion of irrelevant or potentially malicious boilerplate content.
- [SAFE]: All external resources mentioned, such as StackOverflow, GitHub, and various documentation sites, are well-known technology services and fall within the trusted scope for technical research.
Audit Metadata