Skills
Navigate
skills/zhizhunbao/workbuddy/repo-reuse-flow/Gen Agent Trust Hub

repo-reuse-flow

Warn

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: MEDIUMREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill performs dynamic module loading using __import__ and importlib.import_module within the extracted evaluation modules to resolve embedding providers.
  • [REMOTE_CODE_EXECUTION]: The FaithfulnesswithHHEM metric in _faithfulness.py utilizes trust_remote_code=True when downloading models from Hugging Face, which permits the execution of arbitrary code defined in the model repository.
  • [DATA_EXFILTRATION]: Usage events and evaluation statistics are transmitted to a remote telemetry endpoint (https://t.explodinggradients.com) as part of the analytics system in _analytics.py.
  • [EXTERNAL_DOWNLOADS]: Multimodal prompts in multi_modal_prompt.py are capable of fetching and processing image data from external HTTP/HTTPS URLs.
  • [COMMAND_EXECUTION]: The workflow involves executing various Python scripts to search, clone, and perform static analysis on external GitHub repositories.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 17, 2026, 02:23 AM
Security Audit — agent-trust-hub — repo-reuse-flow