humaniseur-fr
Pass
Audited by Gen Agent Trust Hub on May 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional content and evaluation data for a text-processing task, with no malicious scripts or code detected.
- [EXTERNAL_DOWNLOADS]: The skill references documentation from Wikipedia and the author's own GitHub repository (github.com/samber/cc-skills), which is consistent with the skill's development context.
- [COMMAND_EXECUTION]: The skill utilizes tools like Read, Edit, and Write to interact with documents, which is necessary for its function as a text editor.
- [DATA_EXFILTRATION]: Analysis for Category 8 (Indirect Prompt Injection) reveals that while the skill processes untrusted text and has write access, the specific linguistic rewriting focus serves as a safeguard. Mandatory Evidence: Ingestion points: Input text processed by the skill; Boundary markers: Absent; Capability inventory: File system access and Agent tools; Sanitization: Implicitly performed through the detailed humanization transformation process.
Audit Metadata