livekit-skills
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill provides installation commands for the LiveKit server and CLI on Linux platforms using
curl | bashpipelines targeting the officialget.livekit.iodomain. This is the documented distribution method for the platform. - [COMMAND_EXECUTION]: The skill facilitates environment setup and project management through various system commands, including
brew,winget,uv sync,pnpm install, and thelkCLI for initializing agents and running development servers. - [INDIRECT_PROMPT_INJECTION]: The skill architecture involves the agent retrieving live documentation and SDK source code via the
lk docsCLI tool, which serves as a data ingestion point. - Ingestion points: Documentation content is pulled into the agent's context through
lk docs search,lk docs get-page, andlk docs code-searchcommands. - Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore instructions embedded within the retrieved documentation.
- Capability inventory: The agent is empowered to generate code and execute shell commands (
lk,uv,pnpm) as part of the development workflow. - Sanitization: No specific filtering or validation mechanisms for the documentation data are defined in the skill instructions.
Audit Metadata