fine-dust-location

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill makes network requests to an unverified external domain k-skill-proxy.nomadamas.org to fetch environmental data.
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to execute shell commands using curl and a local Python script scripts/fine_dust.py to perform API calls and process data.
  • [DATA_EXFILTRATION]: User-supplied location hints (e.g., "서울 강남구") are sent as URL-encoded parameters to the external proxy server k-skill-proxy.nomadamas.org.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it ingests and processes data from an untrusted external API.
  • Ingestion points: Data returned from https://k-skill-proxy.nomadamas.org/v1/fine-dust/report (SKILL.md).
  • Boundary markers: Absent; there are no instructions to delimit or ignore potential commands within the external data response.
  • Capability inventory: Network requests via curl and local script execution scripts/fine_dust.py (SKILL.md).
  • Sanitization: Absent; no validation or escaping mechanisms for the external content are described.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 01:00 AM
Security Audit — agent-trust-hub — fine-dust-location