open-meteo-historical-fetch
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DATA_EXFILTRATION]: The skill initiates network requests to the official Open-Meteo API endpoint (archive-api.open-meteo.com) to retrieve historical weather and soil data. This operation is consistent with the skill's primary purpose and targets a well-known technology service.
- [PROMPT_INJECTION]: The skill displays an indirect prompt injection surface by processing external weather data. 1. Ingestion points: External data is fetched from the Open-Meteo API via RetryableHttpJsonClient in scripts/open_meteo_historical_fetch.py. 2. Boundary markers: The validated data is printed as JSON; however, it lacks explicit safety delimiters to instruct the agent to ignore any potential instructions embedded in the API response content. 3. Capability inventory: The skill has the ability to write JSON artifacts to the local file system and execute logical checks via a Python script. 4. Sanitization: The validate_records and validate_section functions in the script perform extensive verification of JSON structure, variable types, and date ranges before the data is utilized.
- [COMMAND_EXECUTION]: The skill documentation provides command-line examples for running the local Python script. The script uses the argparse library to safely handle parameters and does not use os.system or eval on untrusted inputs.
Audit Metadata