brandkit
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a helper script,
scripts/brandkit.py, to execute local image generation backends and a design token CLI tool viasubprocess.run. These executions target specific sibling scripts located in the agent's skill directory or the user's home folder. - [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific data from
jtbd.jsonto inform its brand strategy generation. - Ingestion points: Project strategy is ingested from
jtbd.jsonas specified in the primary instructions. - Boundary markers: The skill does not employ explicit delimiters or 'ignore' instructions for the content of the ingested file.
- Capability inventory: Capabilities include writing to local JSON token files and executing local scripts for image generation.
- Sanitization: The skill uses standard JSON parsing but does not sanitize the internal string values used for strategy inference.
Audit Metadata