poetry
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill creates a vulnerability surface for indirect prompt injection by instructing the agent to ingest untrusted content.
- Ingestion points: Untrusted data enters the agent's context through user-provided 'Style examples' and 'Words / dictionary' lists (in the file SKILL.md), the latter of which can be read directly from a user-specified file path.
- Boundary markers: The skill lacks any instructions to use delimiters or ignore embedded prompts within the external data sources.
- Capability inventory: The skill is primarily a text generation assistant. While it does not define direct command or network operations, it utilizes a file-reading capability (in the SKILL.md process) to process external dictionary files, which can be a vector for malicious input.
- Sanitization: No sanitization, validation, or filtering is specified for the content ingested from files or style examples before it is used to influence the poem's creation.
Audit Metadata