storybook-pipeline
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external markdown files provided by the user in the
stories/directory, which presents an attack surface for indirect prompt injection if the files contain malicious instructions. - Ingestion points: Reads user-supplied markdown files (
stories/*.md) to process them through the pipeline. - Boundary markers: The orchestration instructions do not define specific delimiters for the ingested content, though sub-skills may implement them.
- Capability inventory: Performs local file system operations (directory creation, writing JSON, HTML, and binary files), executes asset scripts, and dispatches sub-skills.
- Sanitization: No explicit sanitization is performed by the orchestrator; however, the skill mandates "HARD GATES" where the user must manually review and approve the output of every stage (scene splitting, image bible generation, and narration scripting) before proceeding.
- [DYNAMIC_EXECUTION]: The skill executes a local asset script (
python .claude/skills/story-illustrator/assets/fal_image.py --check) to verify the availability of API keys for external services. This is a functional environment check and does not involve untrusted remote code.
Audit Metadata