← 返回 Skills 市场
Model Intel
作者
aiwithabidi
· GitHub ↗
· v1.0.0
300
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install model-intel
功能描述
Live LLM model intelligence from OpenRouter. Compare pricing, search models by name, find the best model for any task — code, reasoning, creative, fast, chea...
安全使用建议
This skill appears to do what it says — it queries OpenRouter for model info — but there are two mismatches you should be aware of: (1) the registry/README require OPENROUTER_API_KEY, yet the script does not use that key when calling the API, and (2) the script will try to read ~/.openclaw/workspace/.env to find the key if the environment variable is missing. Before installing or supplying credentials, consider: inspect the script yourself (it's short) to confirm no hidden network endpoints; avoid providing unrelated secrets in the .env file; run the script in a sandbox or isolated environment if you must test it; or ask the author to either remove the misleading metadata or to implement proper authenticated requests so the declared requirement matches actual behavior.
功能分析
Type: OpenClaw Skill
Name: model-intel
Version: 1.0.0
The model-intel skill bundle is a legitimate tool for querying LLM pricing and capabilities from the OpenRouter API. The Python script (scripts/model_intel.py) implements standard CLI functionality to list, search, and compare models using the official OpenRouter endpoint (openrouter.ai). There is no evidence of data exfiltration, malicious execution, or prompt injection; the script's access to the local .env file is limited to retrieving its own required API key.
能力评估
Purpose & Capability
Name/description match the code's functionality (fetch models, price, compare). However the registry and SKILL.md declare OPENROUTER_API_KEY as the primary credential even though the included script performs requests to https://openrouter.ai/api/v1/models without supplying any Authorization header or using the key. Requiring a key in metadata that the code does not use is an inconsistency.
Instruction Scope
SKILL.md instructs running the included Python script which calls OpenRouter. The script also attempts to read ~/.openclaw/workspace/.env to extract OPENROUTER_API_KEY if the env var is unset. The README/metadata did not call out that the script will read that local config file; reading a user's dotfile is scope creep and could expose unrelated secrets if the logic were extended.
Install Mechanism
No install spec or remote downloads — the skill is shipped as a small Python script and usage docs. No archives or third-party installers are fetched at runtime.
Credentials
The only declared required environment variable is OPENROUTER_API_KEY, which is reasonable for an OpenRouter integration — except the code does not actually use that key for its API calls. Additionally, the script will parse ~/.openclaw/workspace/.env to obtain that key if absent; that means the skill will read a local credential file even though the declared env var is supposedly sufficient.
Persistence & Privilege
The skill does not request persistent/always-on privileges, does not modify other skill configs, and does not write files or install services. Autonomous invocation is allowed but is the platform default.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install model-intel - 安装完成后，直接呼叫该 Skill 的名称或使用
/model-intel触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
- Initial release of model-intel: real-time LLM model intelligence from OpenRouter.
- Quickly compare pricing, context length, capabilities, and availability for 200+ models.
- Search, list, and compare models by name and use case (code, reasoning, creative, fast, cheap, vision).
- Find the best model for specific needs with simple command-line scripts.
- Requires Python 3.10+ and an OpenRouter API key.
元数据
常见问题
Model Intel 是什么？
Live LLM model intelligence from OpenRouter. Compare pricing, search models by name, find the best model for any task — code, reasoning, creative, fast, chea... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 300 次。
如何安装 Model Intel？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install model-intel」即可一键安装，无需额外配置。
Model Intel 是免费的吗？
是的，Model Intel 完全免费（开源免费），可自由下载、安装和使用。
Model Intel 支持哪些平台？
Model Intel 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Model Intel？
由 aiwithabidi（@aiwithabidi）开发并维护，当前版本 v1.0.0。
推荐 Skills