new-post
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection as it collects user-supplied strings (Title, Description, Tags) and interpolates them into local project files.
- Ingestion points: User input gathered for blog post metadata in
SKILL.md. - Boundary markers: Absent; the skill does not explicitly instruct the agent to sanitize or delimit the user-provided text to prevent instruction injection.
- Capability inventory: The agent has file-writing capabilities to
src/posts/andpublic/rss.xmlas described in Step 1 and Step 2. - Sanitization: Absent; input is placed directly into MDX frontmatter and XML tags.
- Note: This risk is minimal as the behavior is the core function of the skill and involves a human-in-the-loop confirmation before writing files.
Audit Metadata