story-clarifier
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, which creates a potential surface for indirect prompt injection.
- Ingestion points: The skill ingests external data through its 'Draft', 'Synthesize', and 'Rewrite' modes, specifically reading "rough tickets, conversations, user stories, and feature definitions" as defined in the
Workflowsection ofSKILL.md. - Boundary markers: The skill employs structured Markdown contracts (Artifact State Contract, Story Card Contract) to organize its output, but it does not define specific escaping or delimiters for the untrusted input data within the prompt flow.
- Capability inventory: The skill possesses no dangerous capabilities. There are no subprocess calls, file-system write operations, or network requests defined in the instructions.
- Sanitization: The skill relies on strict schema adherence for its outputs, which provides some structural separation between the analysis and the data.
Audit Metadata