theme-craft
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use a local CLI tool (
story) for project maintenance tasks such as reindexing, link checking, and validation. - [DYNAMIC_EXECUTION]: The instructions include a fallback mechanism to execute a script using
nodeorbunfrom a relative path (../story-maintenance/scripts/story.js), which constitutes dynamic execution of local scripts. - [INDIRECT_PROMPT_INJECTION]: 1. Ingestion points: The skill reads
story.mdand character files which contain user-authored narrative content and metadata. 2. Boundary markers: No specific delimiters are defined to separate user prose from the agent's instructions. 3. Capability inventory: The agent has the ability to write to project files and execute local shell commands. 4. Sanitization: The skill does not describe any validation or sanitization of the story content before processing it for thematic audits.
Audit Metadata