film-location
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill documentation focuses on creative prompting techniques and workflow organization for film production, which is consistent with its stated purpose.
- [COMMAND_EXECUTION]: The skill includes example shell commands (
batch.sh,poll.sh) to automate image generation and status polling. These commands are executed from a local path (<abs path of film-pipeline>/scripts) and use standard placeholders, representing a benign local workflow without remote dependencies. - [INDIRECT_PROMPT_INJECTION]: The skill generates 'spatial locks' meant for consumption by other agents in a sequence. 1. Ingestion points: User-provided project slugs and location names in
SKILL.md. 2. Boundary markers: None explicitly defined for the generated text blocks. 3. Capability inventory: Execution of local bash scripts (batch.sh) as described inSKILL.md. 4. Sanitization: Not applicable to the instructional documentation provided.
Audit Metadata