process-transformation-story
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external prompt excerpts and summaries from the
references/cases.mdfile to guide the generation of new artifacts. - Ingestion points: The agent is required to read
references/cases.mdon every invocation to select an 'anchor case'. - Boundary markers: The workflow includes a 'Preserve / Replace / Avoid' reference contract and instructions to 'Distinguish observed evidence from your own recommendation', which serves to separate external content from agent logic.
- Capability inventory: The skill does not request or use tools for shell execution, file writing, or unauthorized network access.
- Sanitization: No explicit sanitization or filtering of the ingested prompt text is performed.
Audit Metadata