alterlab-genai-camera-director
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions direct the agent to ingest external project files such as scripts, shot lists, and storyboards. This creates a surface where malicious instructions embedded in these files could potentially influence the agent's behavior.
- Ingestion points: The 'Workflow' and 'Agentic Protocol' sections specify reading 'existing project files for context — scripts, shot lists, storyboards, prior camera direction sheets'.
- Boundary markers: There are no explicit instructions or delimiters provided to warn the agent to ignore potential instructions embedded within these external project files.
- Capability inventory: The skill has significant capabilities, including writing multiple structured files to the project directory (e.g.,
{project}-camera-direction.md,{project}-energy-map.md) and searching the web for platform updates. - Sanitization: The skill lacks instructions for validating or sanitizing the content of the project files before they are incorporated into the agent's context.
- [COMMAND_EXECUTION]: The skill describes an autonomous workflow that includes running a 'content-scoring tool' and writing multiple files to the local file system. While these are intended functions of the camera director role, they represent a capability tier that requires monitoring, especially when combined with the ingestion of untrusted narrative data.
Audit Metadata