speculative-decoding

Warn

Audited by Snyk on Oct 29, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 0.70). The skill executes git clone from external GitHub repositories (https://github.com/FasterDecoding/Medusa and https://github.com/hao-ai-lab/LookaheadDecoding) and installs them via pip install -e . during setup, introducing arbitrary third-party code as an operational dependency with weak/unknown provenance.

Issues (1)

W012
MEDIUM

Unverifiable external dependency detected (runtime URL that controls agent).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 29, 2026, 03:22 PM
Issues
1