seoul-weather-risk

Pass

Audited by Gen Agent Trust Hub on Aug 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches primary instructions and utility files dynamically using the author's official CLI package (@nomadamas/k-skill) via npx as documented in SKILL.md.
  • [REMOTE_CODE_EXECUTION]: Executes shell commands and fetches remote instructions through the author's CLI tool. These operations are within the scope of the vendor's own infrastructure.
  • [CREDENTIALS_UNSAFE]: The Python client supports reading a MARKETPLACE_API_KEY from a local .env file or environment variables when local-direct mode is enabled. The script implements safe practices by restricting access to specific variable names and ensuring keys are not printed to logs.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes JSON data from an external weather API. Ingestion points: scripts/seoul_weather_risk.py. Boundary markers: Absent in response data. Capability inventory: Network requests and local file reading. Sanitization: Uses standard JSON parsing before displaying data to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 16, 2026, 12:50 PM
Security Audit — agent-trust-hub — seoul-weather-risk