transcript-to-brief
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, untrusted content (meeting transcripts), which creates a potential surface for indirect prompt injection if a transcript contains malicious instructions intended to manipulate the agent's output.
- Ingestion points: The skill accepts a transcript as its primary input argument, as specified in
SKILL.mdviaargumentHint. - Boundary markers:
references/document-contract.mdprovides a strict schema for the output (Title, Abstract, Key Concepts, etc.) and transformation rules that constrain the agent's behavior to professional research-grade synthesis. - Capability inventory: The skill's capabilities are limited to generating a research brief. There are no tools for network access, arbitrary file writing, or shell command execution detected in the provided files.
- Sanitization: The
document-contract.mdrules require the removal of conversational noise, filler, and redundancy, which acts as a basic filter, although no explicit "ignore instructions" delimiter is used.
Audit Metadata