pmg-choreography
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided text strings to generate presentation content, creating a vulnerability surface where malicious instructions could be embedded in data.
- Ingestion points: The skill accepts user-supplied parameters such as
topic_title,topic_subtitle,author_name, andstudent_nameas seen in the Python API usage examples inSKILL.md. - Boundary markers: The skill instructions do not define clear delimiters or include safety warnings for the agent to disregard instructions potentially hidden within these user-provided strings.
- Capability inventory: The skill has the capability to write to the local file system through the
.build_deck()method used bySpotifyAppShellBuilder,EarthSpaceMotionBuilder, andUniversityBloomBuilder. - Sanitization: No sanitization, validation, or escaping of the user-provided input strings is described in the prompt or implementation examples provided in the skill documentation.
Audit Metadata