content-researcher

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to its interaction with untrusted web content.
  • Ingestion points: Web page content is retrieved and processed by subagents during the source extraction phase (references/guide-reader.md) and the question research phase (references/question-researcher.md).
  • Boundary markers: The subagent prompts lack explicit instructions or delimiters to isolate the agent from potentially malicious commands embedded in the retrieved source material.
  • Capability inventory: The skill utilizes web browsing and search capabilities to collect data and has the ability to write extracted knowledge to local markdown files (knowledge-base.md and under-discussed.md).
  • Sanitization: No sanitization, validation, or filtering mechanisms are implemented to process the external content before it is processed by the subagents.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 03:22 AM