meeting-synthesizer
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external meeting notes and transcripts which may contain adversarial instructions aimed at influencing the agent's summary or extraction behavior.
- Ingestion points: Meeting content (notes, transcripts, voice-to-text) provided in various prompts defined in
SKILL.md. - Boundary markers: The instructions use simple placeholders like
[paste meeting notes/transcript]but lack strict delimiters or explicit instructions to treat the content purely as data, increasing the risk of the agent obeying instructions embedded within the transcript. - Capability inventory: The skill focuses on text processing and structuring into Markdown or JSON formats. No subprocess calls, file-writes, or network operations are present in the provided instructions.
- Sanitization: The skill does not implement any validation, escaping, or filtering of the meeting content before it is processed by the model.
Audit Metadata