技能详情（站内镜像，无评论）
许可证：MIT-0
MIT-0 ·免费使用、修改和重新分发。无需归因。
版本：v1.0.0
统计：⭐ 0 · 534 · 1 current installs · 1 all-time installs
⭐ 0
安装量（当前） 1
🛡 VirusTotal ：可疑 · OpenClaw ：可疑
Package：aiwithabidi/model-intel-pro
安全扫描（ClawHub）
- VirusTotal ：可疑
- OpenClaw ：可疑
OpenClaw 评估
The skill largely does what it claims (calls OpenRouter to list and compare models), but there are mismatches around credential usage and an undeclared file read that warrant caution.
目的
Name/description match the code: the script fetches model data from https://openrouter.ai and provides list/search/compare/best/price commands. However the skill declares OPENROUTER_API_KEY as required even though the HTTP requests in the included script do not actually use that key (no header/param). That mismatch is unexpected — either the API key is unnecessary or the script is buggy/unfinished.
说明范围
SKILL.md instructs running the bundled Python script, which is fine. The script, however, will attempt to read ~/.openclaw/workspace/.env to extract OPENROUTER_API_KEY if the env var is not set. The manifest said no required config paths; reading a workspace dotfile in the user's home is an access beyond the declared scope and should have been listed. The script does not read other arbitrary files, but the undeclared home path access is a scop…
安装机制
No install spec is present (instruction-only). That minimizes install-time risk. The bundle includes a single Python script which will be executed; there's no downloader, third-party install, or archive extraction.
证书
The only required env var is OPENROUTER_API_KEY, which is reasonable for an OpenRouter integration. But the script reads the API key from ~/.openclaw/workspace/.env as a fallback and then never uses the key in its requests. This either indicates sloppy/unfinished code or unnecessary access to a file that may contain other secrets. Requesting a credential and not using it is disproportionate and confusing.
持久
The skill does not request always: true and does not attempt to persist itself or modify other skills or system settings. It runs on-demand and performs network calls only to openrouter.ai.
安装（复制给龙虾 AI）
将下方整段复制到龙虾中文库对话中，由龙虾按 SKILL.md 完成安装。
请把本段交给龙虾中文库（龙虾 AI）执行：为本机安装 OpenClaw 技能「Model Intel Pro」。简介：Live LLM model pricing and capabilities from OpenRouter. List top models, searc…。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装：https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/aiwithabidi/model-intel-pro/SKILL.md
（来源：yingzhi8.cn 技能库）SKILL.md
---
name: model-intel
version: 1.0.0
description: >
Live LLM model pricing and capabilities from OpenRouter. List top models, search by name,
compare side-by-side, find best model for a use case, check pricing. Always up-to-date
from the OpenRouter API. Triggers: model pricing, compare models, best model for,
cheapest model, model cost, LLM comparison, what models are available.
license: MIT
compatibility:
openclaw: ">=0.10"
metadata:
openclaw:
requires:
bins: ["python3"]
env: ["OPENROUTER_API_KEY"]
---
# Model Intel 🧠💰
Live LLM model intelligence — pricing, capabilities, and comparisons from OpenRouter.
## When to Use
- Finding the best model for a specific task (coding, reasoning, creative, fast, cheap)
- Comparing model pricing and capabilities
- Checking current model availability and context lengths
- Answering "what's the cheapest model that can do X?"
## Usage
```bash
# List top models by provider
python3 {baseDir}/scripts/model_intel.py list
# Search by name
python3 {baseDir}/scripts/model_intel.py search "claude"
# Side-by-side comparison
python3 {baseDir}/scripts/model_intel.py compare "claude-opus" "gpt-4o"
# Best model for a use case
python3 {baseDir}/scripts/model_intel.py best fast
python3 {baseDir}/scripts/model_intel.py best code
python3 {baseDir}/scripts/model_intel.py best reasoning
python3 {baseDir}/scripts/model_intel.py best cheap
python3 {baseDir}/scripts/model_intel.py best vision
# Pricing details
python3 {baseDir}/scripts/model_intel.py price "gemini-flash"
```
## Use Cases
| Command | When |
|---------|------|
| `best fast` | Need lowest latency |
| `best cheap` | Budget-constrained |
| `best code` | Programming tasks |
| `best reasoning` | Complex logic/math |
| `best vision` | Image understanding |
| `best long-context` | Large document processing |
## Credits
Built by [M. Abidi](https://www.linkedin.com/in/mohammad-ali-abidi) | [agxntsix.ai](https://www.agxntsix.ai)
[YouTube](https://youtube.com/@aiwithabidi) | [GitHub](https://github.com/aiwithabidi)
Part of the **AgxntSix Skill Suite** for OpenClaw agents.
📅 **Need help setting up OpenClaw for your business?** [Book a free consultation](https://cal.com/agxntsix/abidi-openclaw)