The Agent Skills Directory

[SAFE]: No malicious patterns, such as hardcoded credentials, data exfiltration, or obfuscated commands, were found in the skill. The instructions are transparent and focused on linguistic analysis.
[PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection as it is designed to ingest and analyze untrusted data. 1. Ingestion points: The skill reads user-provided text and files using the Read tool for style calibration. 2. Boundary markers: Absent; the instructions do not provide the agent with methods to distinguish between data and instructions within the input. 3. Capability inventory: The skill utilizes the Read, Write, Edit, and AskUserQuestion tools. 4. Sanitization: No sanitization or validation of external input is mentioned in the processing flow.

humanizer-zh