split-transcript
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues were detected. The skill's behavior is consistent with its stated purpose of organizing transcript data into structured markdown files.
- [DATA_EXFILTRATION]: The skill performs local file read and write operations required for transcript segmentation.
- Operations are limited to reading the source transcript and writing segmented content to a
split/directory or established project folders. - No network exfiltration or access to sensitive system configuration files (e.g., SSH keys, AWS credentials, or environment variables) was identified.
- [PROMPT_INJECTION]: The skill processes untrusted transcript content, presenting an indirect prompt injection surface.
- Ingestion points: Untrusted transcript data is read from local files (e.g.,
transcripts/recording.structured.md). - Boundary markers: The instructions explicitly define the task as "reorganisation, not summarisation" and direct the agent to "preserve the user's wording," which helps mitigate the risk of the agent accidentally executing embedded commands within the transcript.
- Capability inventory: The agent has the capability to write multiple markdown files to the local file system.
- Sanitization: No explicit sanitization or escaping of transcript content is performed, but the task is constrained to text preservation and organization.
Audit Metadata