scene-user-demand-research

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because its core workflow involves collecting and analyzing untrusted feedback from third-party sources.
  • Ingestion points: The agent ingests data from external platforms through "public pages, permitted APIs, research datasets, or user-authorized exports" as specified in SKILL.md.
  • Boundary markers: The instructions do not specify the use of delimiters or protective headers to isolate untrusted external text when it is provided to the agent for synthesis or analysis.
  • Capability inventory: The skill facilitates the collection, normalization, and synthesis of data into "opportunity cards" and product decisions.
  • Sanitization: The research protocol does not include explicit instructions for the agent to sanitize or filter potential malicious instructions embedded within the feedback text.
  • [EXTERNAL_DOWNLOADS]: The research protocol involves connecting to and fetching content from external web platforms and APIs.
  • Evidence: SKILL.md and references/research-protocol.md define a workflow for mining reviews and forums, though they explicitly instruct the agent to respect access controls, login walls, and robots.txt restrictions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 07:57 AM
Security Audit — agent-trust-hub — scene-user-demand-research