nts-tax-delinquency

Pass

Audited by Gen Agent Trust Hub on Aug 23, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions in SKILL.md suggest using npx -y @nomadamas/k-skill@0 to fetch instructions and helper files. These commands execute a utility package maintained by the skill's author to manage execution across different environments.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests and processes external HTML content from the National Tax Service website (nts.go.kr).
  • Ingestion points: The scripts/nts_tax_delinquency.py script fetches HTML data using urllib.request from the NTS search endpoint.
  • Boundary markers: The script wraps the output in a structured JSON envelope and uses a regex-based _strip_tags function to clean HTML content.
  • Capability inventory: The skill performs read-only operations and presents the results in the chat. It does not have capabilities for file writes or arbitrary command execution based on the ingested data.
  • Sanitization: Basic HTML tag stripping is performed via regex before the data is returned to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 23, 2026, 08:22 AM
Security Audit — agent-trust-hub — nts-tax-delinquency