blog-rewrite
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [SAFE]: A comprehensive review of the skill's workflow and instructions reveals no malicious patterns, such as prompt injection overrides, credential harvesting, or unauthorized network activity. The skill operates within its intended functional scope of content optimization.\n- [COMMAND_EXECUTION]: The skill utilizes the Bash tool for standard repository maintenance tasks, including using Grep and Glob to identify keyword cannibalization and executing build verification for MDX-formatted files to ensure no compilation errors.\n- [EXTERNAL_DOWNLOADS]: Performs web searches and fetches media from well-known services (Pixabay, Unsplash) and reputable research sources. These operations are conducted to replace fabricated data with sourced statistics and provide legitimate visual elements.\n- [PROMPT_INJECTION]: The skill was analyzed for indirect prompt injection risks (Category 8):\n
- Ingestion points: Reads existing blog content (markdown, MDX, HTML) from the filesystem in Phase 1.\n
- Boundary markers: No explicit delimiters or ignore-instructions are defined for the ingested content.\n
- Capability inventory: Employs Write, Edit, and Bash tools to modify the repository and verify build integrity.\n
- Sanitization: No specific automated sanitization or escaping of ingested content is mentioned.
Audit Metadata