skills/quan0715/open-press/openpress/Gen Agent Trust Hub

openpress

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill manages workspace operations by executing local Node.js scripts (node engine/cli.mjs) and npm scripts (npm run build, npm run dev, npm run openpress:pdf). These are used for validation, rendering, and export tasks.
  • [EXTERNAL_DOWNLOADS]: It provides instructions for bootstrapping new environments using npm create @open-press, which involves fetching the framework's creation tool from the npm registry.
  • [PROMPT_INJECTION]: The skill processes user-controlled MDX source files, creating an indirect prompt injection surface.
  • Ingestion points: Source text within the press/ directory.
  • Boundary markers: None identified for distinguishing content from instructions.
  • Capability inventory: Subprocess execution via npm/node and file modification via replace.
  • Sanitization: Relies on human-in-the-loop verification (previewing replacements) rather than automated filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 04:24 AM
Security Audit — agent-trust-hub — openpress