delivery-tracking

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 the @nomadamas/k-skill Node.js package, which is a vendor-owned resource, to fetch updated instructions and manage the skill environment via npx commands.
  • [COMMAND_EXECUTION]: The instruction set includes bash scripts that execute curl for network requests and python3 for data processing. These scripts follow safe practices by using temporary files for intermediate storage and cleaning them up after execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external tracking pages at cjlogistics.com and epost.go.kr.
  • Ingestion points: Data is fetched via curl into local variables or temporary files in instruction.md.
  • Boundary markers: No specific delimiters are used, but data is treated as a string for parsing.
  • Capability inventory: The skill can execute network requests (curl) and local Python scripts.
  • Sanitization: High. The skill uses strict JSON path navigation and regular expressions to extract only specific logistics fields (status, timestamp, location), effectively sanitizing the input before presenting it to the agent/user.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 07:37 AM