blog-post
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill contains logic for the agent to execute a local shell file using the
sourcecommand (source .env.local). This allows for the execution of arbitrary commands if the target file is manipulated or contains untrusted input. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it instructs the agent to retrieve and process external content from an API without specific sanitization or boundary markers.
- Ingestion points: Content is retrieved from endpoints like
GET /api/v2/blog-posts/{blog_post_id}andGET /api/v2/blog-posts/{id}/generation/{ai_edit_id}. - Boundary markers: The instructions do not define delimiters or warnings to ignore instructions embedded within the retrieved blog content.
- Capability inventory: The skill uses
curlfor network operations andsourcefor command execution. - Sanitization: There are no instructions for the agent to validate or sanitize external data before processing it as part of the blog management workflow.
Audit Metadata