Shell command execution detected (child_process).
Critical
- Code
- suspicious.dangerous_exec
- Location
- dist/index.js:403
- Evidence
execFile(command, args, {
Security audit
llama.cpp Provider
Security checks for vulnerabilities and agentic risk
Overview
This plugin coherently provides llama.cpp model support and discloses its local server setup, downloads, and external-server connection behavior.
Before installing, expect this provider to use disk space for verified llama.cpp binaries and GGUF models, run a local llama-server under OpenClaw when you choose managed setup, and store an API key only if you configure an existing server that requires one.
SkillSpector was not run because this plugin release contains no bundled skills.
Static analysis
Detected: suspicious.dangerous_exec, suspicious.exposed_secret_literal
File appears to expose a hardcoded API secret or token.
Critical
- Code
- suspicious.exposed_secret_literal
- Location
- dist/index.js:1793
- Evidence
const apiKey = [REDACTED]) ? await resolveLlamaServerRuntimeApiKey({ config: ctx.config }) : void 0;