recoup-content-make-graphics
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill documentation describes a 'Stop' hook that prevents the agent from claiming an asset is finished until it has been processed by a verification API. This is a quality control constraint intended to ensure visual integrity.- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local artist workspace files and platform research APIs to customize generation prompts. * Ingestion points: Artist context files and research signals from the recoupable.dev API. * Boundary markers: No specific delimiters or safety instructions are defined for the inclusion of this external content. * Capability inventory: The skill uses file system access and authenticated network requests. * Sanitization: No explicit sanitization or filtering of the retrieved content is mentioned.- [EXTERNAL_DOWNLOADS]: The skill performs authenticated network requests to official recoupable.dev endpoints to generate media assets. These operations are standard for the tool's intended purpose.- [COMMAND_EXECUTION]: The provided documentation includes examples of shell commands using curl, jq, and sed to manage API communication and local metadata.
Audit Metadata