dockkit

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill documents how to handle external data sources such as accessory events, battery states, and tracking observations from vision models. These represent ingestion points for potentially untrusted data from hardware or computer vision outputs.
  • Ingestion points: accessoryStateChanges, accessoryEvents, and trackingStates in SKILL.md and references/dockkit-patterns.md.
  • Boundary markers: Not applicable as this is framework documentation for developers/agents.
  • Capability inventory: Motor control (setAngularVelocity, setOrientation) and camera control commands.
  • Sanitization: Not explicitly implemented in the provided snippets, as they are standard API usage examples.
  • [EXTERNAL_DOWNLOADS]: The skill contains multiple references to external documentation and video tutorials hosted on sosumi.ai. These links are informational and do not involve the download or execution of remote code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 02:32 PM