resolve-tighten-recording
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate instructions for interacting with a DaVinci Resolve MCP environment to automate video editing tasks, specifically removing dead air from recordings.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies an attack surface where transcription data (from audio) is ingested and used for editing decisions. However, the risk is mitigated by a mandatory human approval gate (Step 4) and the fact that tool operations are confined to local timeline modifications in a video editor.
- [COMMAND_EXECUTION]: The skill utilizes platform-specific MCP tools (e.g.,
media_analysis,edit_engine) for its primary purpose. These tools are part of the intended execution environment and are used according to standard automation patterns. - [DATA_EXPOSURE]: The skill reads project metadata such as clip IDs and timeline structures to perform its function. This data remains within the local environment and no network exfiltration patterns were identified.
Audit Metadata