blog-context
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or data exfiltration behaviors were detected. The skill maintains a local file, .agents/blog-context.md, as a persistent store for blogging-related context, which is the intended functionality.
- [INDIRECT_PROMPT_INJECTION]: The skill reads external web content from user-provided URLs to draft context, which represents a potential attack surface for indirect prompt injection from malicious sites. 1. Ingestion points: homepage, about page, and recent posts from user-supplied URLs (SKILL.md). 2. Boundary markers: No specific delimiting instructions are provided for parsing the web content. 3. Capability inventory: Network read (URL fetching), file system write and read (.agents/blog-context.md). 4. Sanitization: The skill includes a requirement to provide a summary and identify gaps for the user to review after the automated drafting process, facilitating human verification.
Audit Metadata