tool-x-article-publisher
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- PROMPT_INJECTION (LOW): The skill processes untrusted Markdown content which could contain malicious instructions designed to influence the agent or the target platform.
- Ingestion points: article_md_path (Markdown file).
- Boundary markers: None explicitly implemented to isolate the content from instructions.
- Capability inventory: Local file reading, system clipboard modification, and browser automation (via Playwright).
- Sanitization: Verification is not possible as the parsing script (parse_markdown.py) was not provided in the source files.
- EXTERNAL_DOWNLOADS (LOW): The skill requires the installation of Pillow and pyobjc-framework-Cocoa. These are reputable, standard libraries on PyPI, justifying a LOW severity level.
- COMMAND_EXECUTION (SAFE): The skill executes internal helper scripts to manage content parsing and clipboard interaction, which is consistent with its stated purpose.
Audit Metadata