research-grants
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill primarily consists of reference documentation and templates. No malicious code, obfuscation, or persistence mechanisms were found.
- [INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting user-provided research drafts and data, which is a potential surface for indirect prompt injection. However, this risk is mitigated by explicit integrity rules in the operating procedures that require the agent to validate all claims against verified sources and prohibit the addition of unapproved results or references.
- [REMOTE_CODE_EXECUTION]: The skill documentation describes how users can optionally generate diagrams using a separate local script (
generate_schematic.py) and mentions human-led installation of LaTeX templates from public repositories. These are informational references for the user and do not constitute automated or hidden execution of remote code by the agent.
Audit Metadata