skills/cjakma/k-skill/korea-weather/Gen Agent Trust Hub

korea-weather

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill demonstrates a secure design by delegating the management of the KMA_OPEN_API_KEY to a proxy server, thereby protecting sensitive credentials from being handled or exposed within the user's local environment.
  • [COMMAND_EXECUTION]: The workflow involves using curl to query an external weather proxy service. This operation is driven by the user-defined KSKILL_PROXY_BASE_URL environment variable.
  • [PROMPT_INJECTION]: The skill ingests external data from the weather proxy, creating a potential indirect prompt injection surface. 1. Ingestion points: Weather forecast data returned from the proxy service (referenced in SKILL.md). 2. Boundary markers: Absent; there are no explicit delimiters to separate external data from agent instructions. 3. Capability inventory: Network access via curl. 4. Sanitization: Absent; the skill relies on the agent to conservatively summarize specific weather fields (TMP, SKY, etc.) from the response.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 05:48 AM
Security Audit — agent-trust-hub — korea-weather