vibeshot-candid-photography
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates as a pure text-to-text generation utility without any network communication, remote downloads, or external dependencies.
- [COMMAND_EXECUTION]: The README provides standard manual installation commands for the local file system. These commands do not involve automated execution, remote fetching, or privilege escalation.
- [INDIRECT_PROMPT_INJECTION]: The skill provides an interface for user input to influence the final prompt generation, which represents a typical injection surface. However, the lack of powerful capabilities minimizes the risk. 1. Ingestion points: User-specified parameters for photography subjects, scenes, and camera settings defined in SKILL.md. 2. Boundary markers: The skill instructions enforce a strict output format (numbered headers), which acts as a structural constraint on the AI's response. 3. Capability inventory: The skill environment is limited to natural language generation; no tools for system access, file writing, or network operations are requested or used. 4. Sanitization: No formal input validation is implemented, but the output is confined to descriptive photography prompts.
Audit Metadata