z-smart-xparse

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). Two URLs point to direct installer scripts (.sh and .ps1) intended to be fetched-and-executed (curl|bash, irm|iex), which is a high-risk distribution pattern even if hosted on a vendor subdomain.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The required workflow runs xparse-cli parse <FILE> (and for large PDFs: splits into chunks then parses each chunk), and the tool’s extracted document content is returned as Markdown/JSON and then read by the agent (Step 6 “Read the markdown result”), so outsider-authored free text from the user-provided document is ingested into the LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's setup/install instructions explicitly fetch and execute remote install scripts (e.g., curl/irm → shell/iex) from https://dllf.intsig.net/download/2026/Solution/xparse-cli/install.sh and https://dllf.intsig.net/download/2026/Solution/xparse-cli/install.ps1, which is a runtime dependency that executes remote code.