kr-whois-lookup

Pass

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to use npx to download the @nomadamas/k-skill package from the npm registry to fetch instructions and manage skill files.
  • [REMOTE_CODE_EXECUTION]: The skill uses npx to execute remote code from the @nomadamas/k-skill package. This package is used for dynamic instruction loading, listing files, and updating the skill context.
  • [COMMAND_EXECUTION]: The skill uses the curl command-line tool to perform GET requests against the k-skill-proxy.nomadamas.org API for WHOIS lookups.
  • [INDIRECT_PROMPT_INJECTION]: The skill handles data from the Public Data Portal (data.go.kr) WHOIS API, which represents an untrusted data source.
  • Ingestion points: API responses from k-skill-proxy containing domain registration, IP allocation, and AS number details.
  • Boundary markers: No specific delimiters or protective instructions are used to separate the ingested WHOIS data from the agent's prompt context.
  • Capability inventory: The skill possesses network access via curl and the ability to execute a specific Node.js CLI tool.
  • Sanitization: Input values (domains, IPs, AS numbers) are normalized and validated, but the output data returned from the API is summarized without explicit sanitization for embedded instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 1, 2026, 02:41 PM