The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (such as rm and npm run) that directly incorporate a user-supplied slug argument. This creates a potential for command injection if the input is not strictly validated (e.g., a slug containing shell metacharacters like ; or &).
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface where untrusted data could influence agent behavior:
Ingestion points: Untrusted data enters the agent context via the story.json configuration files and the user-provided slug directory name.
Boundary markers: No delimiters or instructions to ignore embedded commands are specified when reading or processing these inputs.
Capability inventory: The skill leverages powerful tools including Bash, Write, Edit, and Read, which could be abused if the agent follows instructions embedded in the data.
Sanitization: There are no explicit instructions for sanitizing or validating the content of story.json before it is used to perform file edits or shell operations.

story-render