screencast-record
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data such as session briefs, take logs, and playback observations to generate a final recording handoff.
- Ingestion points: Data enters the context through user-provided 'record cards' and 'take logs' as described in SKILL.md.
- Boundary markers: None identified. The skill does not specify markers to isolate user-provided observations from the agent's internal logic.
- Capability inventory: The skill performs no file system writes, network operations, or shell command executions.
- Sanitization: No explicit filtering or sanitization of user-provided notes or observations is performed.
Audit Metadata