Skills
skills/yonatangross/orchestkit/llm-integration/Gen Agent Trust Hub

llm-integration

Warn

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: scripts/create-lora-config.md utilizes dynamic context injection to execute shell commands such as grep and find on the host system at the time the skill is loaded to automatically populate configuration fields.
  • [DATA_EXFILTRATION]: The dynamic context commands in scripts/create-lora-config.md access and read from .env files to extract information. Accessing environment configuration files via automated shell commands can lead to unintended data exposure.
  • [EXTERNAL_DOWNLOADS]: rules/local-ollama-setup.md fetches and executes an installation script from the official Ollama domain. This uses a remote shell script for system setup.
  • [REMOTE_CODE_EXECUTION]: The skill facilitates the execution of remote code by documenting the curl | sh installation method for the Ollama platform.
Audit Metadata
Risk Level
MEDIUM
Analyzed
May 18, 2026, 12:23 AM
Security Audit — agent-trust-hub — llm-integration