competitor-finder
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze content from untrusted external sources, including competitor websites, marketplace profiles (Upwork, Toptal), and third-party review platforms (G2, Clutch, GoodFirms). This creates a surface for indirect prompt injection attacks where malicious actors could host hidden instructions on these platforms to influence the agent's behavior.
- Ingestion points: External URLs from competitor sites and review platforms specified in Step 2, Step 3, and Step 4 of the instructions.
- Boundary markers: Absent. The instructions do not provide delimiters or specific prompts to the agent to treat external content as data rather than instructions.
- Capability inventory: Low risk. The skill only produces a text-based report and does not utilize dangerous tools like shell execution or file system modification.
- Sanitization: Absent. There are no instructions to sanitize or escape data retrieved from the web before interpolation into the agent's reasoning process.
- [DATA_EXPOSURE]: The skill prompts the user to provide sensitive internal business information, specifically CRM "lost-reason" data and details regarding deals lost in the last 12 months. While this is used for the intended purpose of the skill, users should be aware of the privacy implications of feeding proprietary sales data into an AI model's context.
Audit Metadata