han-river-water-level

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx to fetch and execute the vendor-owned package @nomadamas/k-skill for instruction retrieval and utility updates. This is an expected management pattern for the nomadamas ecosystem.
  • [COMMAND_EXECUTION]: The skill instructions involve executing curl commands to query the vendor-owned proxy server at k-skill-proxy.nomadamas.org.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user-supplied station names and codes to construct data queries. [1] Ingestion points: stationName and stationCode inputs defined in instruction.md. [2] Boundary markers: Example commands use the curl --data-urlencode flag to separate parameters from the command structure. [3] Capability inventory: The skill is limited to shell execution of network requests via curl. [4] Sanitization: User input is URL-encoded by the curl tool before being sent to the endpoint, preventing simple command injection or parameter manipulation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 07:50 AM