model-pruning
Warn
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads external research code from third-party GitHub repositories:
https://github.com/locuslab/wandaandhttps://github.com/IST-DASLab/sparsegpt. - [COMMAND_EXECUTION]: The installation instructions require executing
pip install -r requirements.txtandpip install -e .within the downloaded directories. This results in the execution of installation scripts provided by these external sources. - [PROMPT_INJECTION]: The skill presents a potential surface for indirect prompt injection during model calibration.
- Ingestion points:
calib_dataparameter inwanda_prune(found inSKILL.mdandreferences/wanda.md). - Boundary markers: Not present; the text is tokenized and fed directly to the model.
- Capability inventory: The skill can write to the file system using
model.save_pretrained. - Sanitization: None; calibration samples are processed without filtering.
Audit Metadata