keyword-research
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions define a workflow for processing external, untrusted data which could potentially contain malicious instructions targeting the agent.
- Ingestion points: The 'Report Workflow' section in SKILL.md mentions parsing Excel/CSV files and visiting competitor or product pages via web search.
- Boundary markers: There are no explicit instructions for the agent to ignore or delimit instructions found within these external files or web content.
- Capability inventory: The agent has the capability to perform web searches and read local project context files (SKILL.md).
- Sanitization: No specific sanitization or validation steps are defined for the parsed data.
Audit Metadata