ray-x-article
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes local Python scripts
prepare_article.pyandrecord_draft.py. These scripts are used for article validation, converting content to HTML, and updating local metadata within Markdown files. These actions are consistent with the skill's primary function of article preparation and tracking. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from local Markdown files via the
--articleparameter, which presents an indirect prompt injection surface. This risk is addressed through explicit validation logic inprepare_article.py(e.g., checking for heading counts and bold statements) and the use ofhtml.escapeto sanitize content before it is pasted into the browser environment. - [EXTERNAL_DOWNLOADS]: The skill interacts with the well-known service
x.comthrough browser-control tools to save article drafts. This interaction is specific to the official articles composer and does not involve downloading or executing untrusted code from the domain.
Audit Metadata