markdown-new

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides legitimate utility for web content extraction. No malicious code, obfuscation, or unauthorized data access patterns were identified in the analysis of the provided documentation and scripts.\n- [EXTERNAL_DOWNLOADS]: The skill facilitates network requests to https://markdown.new/ to perform its stated purpose of URL-to-Markdown conversion. This behavior is documented and represents the primary functionality of the skill.\n- [COMMAND_EXECUTION]: The documentation provides standard curl commands to interact with the service API. These commands do not include any dangerous flags, unauthorized system access, or suspicious shell piping.\n- [PROMPT_INJECTION]: As a tool that ingests external web content, the skill represents a surface for indirect prompt injection.\n
  • Ingestion points: Content is retrieved from arbitrary external URLs via the markdown.new conversion service (SKILL.md).\n
  • Boundary markers: The provided examples do not implement delimiters or safety warnings to separate fetched content from agent instructions.\n
  • Capability inventory: Capabilities are limited to standard HTTP network operations via requests, fetch, and curl. No file system or system execution capabilities are exposed.\n
  • Sanitization: No specific content sanitization or instruction filtering is implemented in the provided examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 09:16 PM