skills/brevdev/nemoclaw-demos/planet/Gen Agent Trust Hub

planet

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on a pre-defined Node.js script (planet-api.js) to interact with Planet's services. It does not utilize subprocesses or dynamic shell execution for its operations.
  • [CREDENTIALS_UNSAFE]: The implementation follows security best practices by keeping API credentials on the host machine. The skill interacts with a host-side proxy that injects the necessary authorization headers, ensuring the API key never enters the sandboxed environment.
  • [DATA_EXFILTRATION]: Network activity is restricted to a local proxy endpoint (10.200.0.1), preventing arbitrary data transmission to unknown external domains. The proxy enforces access controls, such as blocking order placement (POST) requests.
  • [PROMPT_INJECTION]: The skill documentation includes constraints (DO NOT instructions) designed to guide the agent toward safe and efficient tool usage rather than attempting to bypass safety filters or override core behavior.
  • [DATA_EXPOSURE]: The skill ingests data from external API responses via the proxy. While this represents a potential surface for indirect prompt injection, the risk is mitigated as the script parses the data into structured JSON objects and does not evaluate it as code.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 08:28 AM
Security Audit — agent-trust-hub — planet