positioning-with-ekram
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill directs the agent to suppress its internal methodology and identity, instructing it to never mention the manual or process to the user to maintain a consistent advisor persona.
- [EXTERNAL_DOWNLOADS]: The agent is instructed to fetch rival company homepages via browsing tools for the purpose of market research and differentiation analysis.
- [COMMAND_EXECUTION]: The skill uses file system operations to read and maintain agent-specific context files and generate analysis reports within the local workspace.
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection by processing external data from competitor websites and user-provided copy. 1. Ingestion points: rival homepage content and user-pasted copy. 2. Boundary markers: No delimiters or ignore instructions are specified for external data. 3. Capability inventory: Browsing and file write capabilities. 4. Sanitization: No input filtering is defined.
Audit Metadata