research-synthesizer

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted research data from multiple sources.
  • Ingestion points: Data is ingested via placeholders in SKILL.md.
  • Boundary markers: XML-style tags provide structure, but no instructions warn the agent to ignore instructions within the sources.
  • Capability inventory: No tools, subprocesses, or network operations are present.
  • Sanitization: No validation or filtering is applied to source content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 08:56 AM
Security Audit — agent-trust-hub — research-synthesizer