academic-deep-research

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: A thorough analysis of the skill's scripts, instructions, and metadata shows no evidence of malicious patterns, obfuscation, persistence mechanisms, or unauthorized privilege escalation.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes web_search and web_fetch to gather research data from the internet. This behavior is consistent with the skill's stated purpose of academic investigation and relies on standard platform capabilities.
  • [PROMPT_INJECTION]: While the skill ingests external data from web sources, it implements robust mitigations against indirect prompt injection (Ingestion: web_search and web_fetch in Phase 3; Boundary markers: Absent; Capability inventory: web_fetch and sessions_spawn; Sanitization: Mandatory cross-referencing of results across multiple sources and required human-in-the-loop approval before finalizing research plans or reports). This structured methodology reduces the risk of the agent following malicious instructions embedded in retrieved web content.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 10:49 AM
Security Audit — agent-trust-hub — academic-deep-research