competitive-research
Pass
Audited by Gen Agent Trust Hub on Mar 27, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted data from external competitor URLs, representing an indirect prompt injection surface. 1. Ingestion points: Competitor names or URLs provided by user input. 2. Boundary markers: None specified in the instructions. 3. Capability inventory: Local file writing and Linear issue creation. 4. Sanitization: No explicit sanitization or filtering of external content is mentioned.
- [COMMAND_EXECUTION]: The skill facilitates file system writes to store research data and integrates with the Linear service for workflow management. These operations are aligned with the skill's primary purpose and use well-known service integrations.
Audit Metadata