cutagent
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to interact with a locally installed command-line interface (
cutagent --json status) to verify capabilities and connection state. This is an expected behavior for an integration skill targeting DaVinci Resolve. - [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface by instructing the agent to process external data such as project timelines, media files, and clip metadata.
- Ingestion points: Data from DaVinci Resolve projects, media files, and local assets.
- Boundary markers: None explicitly mentioned in the instructions; however, the skill advises verifying actual results rather than relying on tool output alone.
- Capability inventory: The agent can perform file edits, rendering, captioning, and state inspection using the
cutagentSDK and CLI. - Sanitization: No specific sanitization or escaping of external project data is described in this routing file.
Audit Metadata