Skills
Navigate
skills/polyai/adk/poly-adk-rtc/Gen Agent Trust Hub

poly-adk-rtc

Pass

Audited by Gen Agent Trust Hub on Oct 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute various poly rtc commands (pull, push, diff, validate, edit) to manage environment configurations. It specifically warns against pushing to live environments without explicit user consent.
  • [EXTERNAL_DOWNLOADS]: The skill documentation specifies the installation of the polyai-adk package using the uv tool. This is a standard vendor-provided utility for the PolyAI platform.
  • [INDIRECT_PROMPT_INJECTION]: The skill manages and processes external configuration data which could theoretically contain malicious instructions.
  • Ingestion points: poly rtc pull fetches remote environment configurations into local schema.json and data.json files.
  • Boundary markers: None specified for the content of the JSON files.
  • Capability inventory: Shell execution of the poly CLI tool for environment management.
  • Sanitization: Uses the poly rtc validate command to ensure data.json conforms to the expected schema.json structure.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 15, 2026, 01:06 PM