theme-craft
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill guides the agent to execute shell commands using the
storyCLI tool for project maintenance. - Key commands include
story reindex,story wordcount, andstory checkto update project metadata and validation states. - [INDIRECT_PROMPT_INJECTION]: The skill processes content from user-authored files (such as
story.mdand character profiles) to audit themes and draft premises, creating a potential path for malicious instructions embedded in story text to affect agent behavior. - Ingestion points: The skill reads genres and themes from
story.mdand character-specific arc fields from individual character files. - Boundary markers: The instructions lack explicit directives for using delimiters or protective warnings when interpolating user-provided story content into the agent's context.
- Capability inventory: The agent is authorized to perform file read/write operations and execute shell commands via the
storyutility. - Sanitization: No validation or sanitization steps are defined for the input data extracted from story project files.
- [DYNAMIC_EXECUTION]: The skill facilitates the execution of Node.js scripts located in relative paths or user-specified repository checkouts.
- Specifically, it uses
node ../story-maintenance/scripts/story.jsas a fallback mechanism when the global CLI is unavailable, targeting scripts within the same vendor's ecosystem.
Audit Metadata