ad-brief
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's behavior is entirely consistent with its purpose as a research and planning tool for advertising. It manages project-specific data within its own dedicated directory and interacts with the user for decision-making.\n- [EXTERNAL_DOWNLOADS]: The skill is instructed to visit external websites, including the user's own product page and competitor sites, to perform market research. These network operations are standard for a research-oriented agent and do not target untrusted or suspicious repositories.\n- [PROMPT_INJECTION]: The skill identifies an indirect prompt injection surface as it ingests and processes content from external websites and local files.\n
- Ingestion points: Website content from user-provided URLs and competitor pages, plus local project files like package.json and README (SKILL.md).\n
- Boundary markers: There are no explicit instructions or delimiters used to isolate processed external data from the agent's core instructions.\n
- Capability inventory: The agent can read and write files within the project directory and perform network requests for browsing (SKILL.md).\n
- Sanitization: The skill does not describe any specific sanitization or filtering of content retrieved from external sources before it is interpreted by the agent.
Audit Metadata