publish-substack-article
Pass
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the execution of Python code blocks for Markdown-to-HTML conversion and uses a local script ('copy_to_clipboard.py') to handle system clipboard operations.
- [EXTERNAL_DOWNLOADS]: The skill specifies a dependency on the 'markdown' package, which must be installed from the official Python Package Index (PyPI).
- [PROMPT_INJECTION]: As the skill processes untrusted user-supplied Markdown files and utilizes browser automation tools (like Playwright or Chrome DevTools MCP), it is susceptible to indirect prompt injection. Malicious instructions embedded within the processed Markdown content could attempt to hijack the agent's actions during the browser automation workflow.
- Ingestion points: Reads content from user-provided file paths (e.g., '/path/to/article.md').
- Boundary markers: The skill suggests stripping YAML frontmatter but lacks explicit 'ignore instructions' delimiters for the main content body.
- Capability inventory: Includes browser navigation, element clicking, text filling, file uploads, and local command execution.
- Sanitization: Relies on standard Markdown-to-HTML conversion, which does not necessarily strip malicious natural language instructions.
Audit Metadata