speculative-decoding

Warn

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to download source code from external, third-party GitHub repositories that are not included in the pre-defined trusted organizations list:
  • https://github.com/FasterDecoding/Medusa (referenced in SKILL.md and references/medusa.md)
  • https://github.com/hao-ai-lab/LookaheadDecoding (referenced in SKILL.md and references/lookahead.md)
  • [REMOTE_CODE_EXECUTION]: The documentation provides installation commands that clone these external repositories and immediately install them in editable mode (pip install -e .). This process executes build scripts (e.g., setup.py) from the downloaded source, which can lead to arbitrary code execution if the external repositories were compromised.
  • [COMMAND_EXECUTION]: The skill includes multiple shell command blocks for environment setup and software installation, involving git clone, cd, and pip install operations targeting external codebases.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 17, 2026, 07:53 PM
Security Audit — agent-trust-hub — speculative-decoding