tax-incentive-eligibility-check-free

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The SKILL.md file contains a section titled '需要完整档时(可选)' (When you need the full version) which provides a block of text intended for the user to copy and paste to their AI assistant. This text contains explicit instructions for the agent to install a package (npx -y @alipay/agent-payment@latest install) and initiate a financial transaction, which steers the agent to perform administrative and financial actions beyond the scope of the original skill.
  • [EXTERNAL_DOWNLOADS]: The skill's documentation recommends the installation of an external Node.js package (@alipay/agent-payment) via npx. This package is associated with Alipay, a recognized financial services provider. The documentation provides instructions for users interested in the 'buyout' version of the tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external tax data tables, which presents a surface for potential indirect prompt injection if the agent were to interpret data as instructions.
  • Ingestion points: The scripts/run.mjs script accepts external data via a file path provided to the --input argument.
  • Boundary markers: The skill does not employ explicit delimiters or 'ignore' instructions when processing the input file; the content is parsed as text by the local engine.
  • Capability inventory: The skill uses Bash to execute the Node.js runtime and Read permissions to access the local data files.
  • Sanitization: The engine in scripts/engine/tax-incentive-eligibility-check.js performs arithmetic validation and regex-based parsing on the input. It does not execute the data as code or system commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 11:28 AM
Security Audit — agent-trust-hub — tax-incentive-eligibility-check-free