humanizer
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted user-provided text to remove AI-like patterns but lacks boundary markers or instructions to ignore embedded commands. This creates an attack surface for indirect prompt injection, which is significant given the skill's file system modification permissions.
- Ingestion points: User-provided text via the /humanizer command or direct requests in file SKILL.md.
- Boundary markers: Absent; the skill does not use delimiters or XML tags to isolate processed text.
- Capability inventory: The skill allows the use of Read, Write, Edit, Grep, and Glob tools in file SKILL.md.
- Sanitization: None; there is no validation or filtering of external content before processing.
- [EXTERNAL_DOWNLOADS]: The README.md file provides installation instructions that involve cloning a repository from GitHub (github.com/blader/humanizer.git). As GitHub is a well-known service, this download reference is documented neutrally.
Audit Metadata