write-blog-post
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of natural language instructions and YAML configuration for an AI agent's interface. It contains no executable scripts, binaries, or automated tasks that could interact with the host system or external networks.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied blog notes. While this constitutes an untrusted data ingestion surface, the skill lacks dangerous capabilities such as file system writes, network requests, or command execution. Additionally, the instructions explicitly command the agent to only use supplied content and avoid inventing facts, which serves as a prompt-level defense against potential injection attempts in the user's data.
Audit Metadata