Security audit

llama.cpp Provider

Security checks for vulnerabilities and agentic risk

Overview

This package is a coherent llama.cpp provider that installs or connects to local model servers with disclosed, user-directed setup behavior.

Before installing, confirm you want OpenClaw to manage a local llama.cpp runtime and large model downloads, or connect to a llama-server you operate. Review the endpoint and API-key choice during setup, and expect persistent model/runtime files under OpenClaw state until removed.

SkillSpector

SkillSpector was not run because this plugin release contains no bundled skills.

Static analysis

Detected: suspicious.dangerous_exec, suspicious.exposed_secret_literal

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: dist/index.js:409
Evidence: execFile(command, args, {

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: dist/index.js:1874
Evidence: const apiKey = [REDACTED]) ? await resolveLlamaServerRuntimeApiKey({ config: ctx.config }) : void 0;