pmem-style-generation
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the
powermemandpython-dotenvPython packages during its environment setup phase. These are functional dependencies provided by the vendor to enable memory and configuration management. - [COMMAND_EXECUTION]: The skill orchestrates several local Python scripts to handle its internal logic, including
check_config.pyfor API key validation,match_profile.pyfor stylistic intent matching, andsave_generation.pyfor file management. These operations are confined to the skill's directory and temporary folders. - [SAFE]: The skill follows security best practices for handling untrusted external data. It implements a two-step process: first, a 'neutral information extraction' prompt (dehydration) strips away stylistic or instructional noise from user-provided content using clear boundary markers (
---). This extracted, objective summary is then used as the basis for the final generation, effectively mitigating the risk of indirect prompt injection.
Audit Metadata