social-reframe
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The workflow involves running a local script
scripts/social_reframe.py. While the script content was not provided for review, invoking local scripts is a standard capability for AI agent skills to perform deterministic tasks. - [INDIRECT_PROMPT_INJECTION]: The skill processes external project data from
../../references/SOCIAL_CAMPAIGN.md. This represents a potential surface for indirect instructions, though it is used here for legitimate configuration of asset reframing. - Ingestion points: Reads campaign metadata and focus-region information from
../../references/SOCIAL_CAMPAIGN.md. - Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are defined.
- Capability inventory: Limited to running a Python script for image/video reframing.
- Sanitization: No specific sanitization or validation of the input file content is described.
Audit Metadata