air-quality-advisor-skill

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security risks or malicious behaviors were found. The skill is implemented with high code quality and follows best practices for API interactions and local resource management.
  • [EXTERNAL_DOWNLOADS]: The skill communicates with official Singapore Government infrastructure (data.gov.sg, api-open.data.gov.sg, and onemap.gov.sg) to provide the service. These are well-known, trusted endpoints used strictly for the skill's primary purpose.
  • [DATA_EXFILTRATION]: No evidence of unauthorized data harvesting or transmission of sensitive local files. The skill implements a local cache in the user's home directory (~/.hermes/cache) to reduce API load, which is appropriate for a performance-conscious utility.
  • [COMMAND_EXECUTION]: All data processing is performed within Python logic. The code does not utilize subprocesses, system shells, or dynamic script execution to handle user inputs or API responses.
  • [PROMPT_INJECTION]: The instructions provided in SKILL.md are functional and focus on operational parameters without attempts to influence the agent's safety protocols or override its core guidelines.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 04:52 AM
Security Audit — agent-trust-hub — air-quality-advisor-skill