Env credential access
Critical
- Finding
- Environment variable access combined with network send.
API usage optimization
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This skill appears aligned with its purpose of configuring cheaper AI model routing, but it uses provider API keys and can persistently change OpenClaw's default model behavior.
This skill is not showing malicious behavior in the provided artifacts. Before installing or running it, review the generated OpenClaw config, choose the intended mode carefully, keep the backup file, and make sure you are comfortable with the listed model providers receiving your future prompts.
Static analysis
Env credential access
Critical
- Finding
- Environment variable access combined with network send.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
What this means
If applied, the generated configuration can change which models future OpenClaw agents use by default.
Why it was flagged
The skill instructs the user to patch OpenClaw's configuration and restart the gateway, which is powerful but clearly tied to the stated routing/fallback purpose.
Skill content
cp ~/.openclaw/openclaw.json ~/.openclaw/openclaw.json.backup ... openclaw config.patch < ~/.openclaw/free-models.json ... openclaw gateway restart
Recommendation
Review the generated configuration first, keep the backup, and only apply the mode you intentionally selected.
What this means
Your OpenRouter, SiliconFlow, NVIDIA, DeepSeek, or Zhipu credentials may be used to query provider APIs when you run the scripts.
Why it was flagged
The script reads API keys from environment variables and sends them as Bearer tokens to provider APIs for model checks, which is expected for this integration.
Skill content
const apiKey = process.env[endpoint.apiKeyEnv]; ... 'Authorization': `Bearer ${apiKey}`Recommendation
Use provider keys with the minimum needed scope, avoid committing keys to repositories, and run the scripts only in a trusted environment.
What this means
Future prompts may be handled by providers other than your current primary model provider, depending on the selected mode and failures.
Why it was flagged
The generated routing configuration can send future model requests to multiple third-party providers as fallbacks or task-specific routes.
Skill content
"routing": { "coding": ["${您的主模型}", "siliconflow/...", "nvidia/..."], ... }Recommendation
Confirm that each listed provider is acceptable for the kinds of data you send through OpenClaw.