inkos-long-market-research

Pass

Audited by Gen Agent Trust Hub on Aug 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional and guides the agent to use standard platform tools for market research. No suspicious patterns, obfuscation, or malicious instructions were detected.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, creating an inherent surface for indirect prompt injection. However, the instructions provide mitigation by directing the agent to separate observed evidence from creative recommendations.
  • Ingestion points: The skill uses research_web and ingest_material in SKILL.md to fetch external content.
  • Boundary markers: SKILL.md explicitly instructs to "Separate observed evidence, interpretation, and creative recommendation."
  • Capability inventory: The skill utilizes standard tools for research (research_web, ingest_material, retrieve_material) and project referencing (manage_book_reference). It does not use subprocesses, network shells, or sensitive file system access.
  • Sanitization: No technical sanitization is defined, but the agent is guided to handle data distinctively.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 16, 2026, 10:11 AM
Security Audit — agent-trust-hub — inkos-long-market-research