← Back to Skills Marketplace
sucriss

发现、对比和配置多平台免费/低价 AI 模型

by Criss_Su · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
375
Downloads
0
Stars
2
Active Installs
1
Versions
Install in OpenClaw
/install free-model-finder
Description
发现、对比并自动配置多平台免费或低价 AI 模型，支持自动轮换以应对限流和优化成本。
README (SKILL.md)

Free Model Finder - 多平台免费模型发现与配置

核心功能

1️⃣ 模型发现

扫描多个平台的免费/低价模型：

  • OpenRouter：免费模型聚合（:free 后缀）
  • HuggingFace：免费 Inference API 额度
  • Groq：免费高速推理（Llama、Mixtral 等）
  • Google AI Studio：Gemini 免费额度
  • Ollama：本地运行（完全免费）
  • 其他平台：可扩展

2️⃣ 模型对比

按以下维度排序：

  • 价格（免费/每百万 token 成本）
  • 速度（平均响应时间）
  • 质量（上下文长度、能力评级）
  • 稳定性（rate limit 策略）

3️⃣ 自动配置

  • 直接修改 OpenClaw 配置文件
  • 设置主模型 + 备用模型列表
  • 自动重启 Gateway 生效

4️⃣ 监控轮换

  • 检测 rate limit 错误
  • 自动切换到备用模型
  • 记录切换历史

快速开始

# 查看所有平台免费模型清单
free-model-finder list

# 只看 OpenRouter 免费模型
free-model-finder list --platform openrouter

# 自动配置最优免费模型（推荐）
free-model-finder auto

# 切换到指定模型
free-model-finder switch groq/llama-3.1-70b-versatile

# 查看当前配置
free-model-finder status

# 监控 rate limit 并自动轮换
free-model-finder watch --daemon

命令参考

命令 说明
list [--platform] [--limit] 列出可用模型
compare [--top N] 对比前 N 个模型的性价比
auto [--platforms] 自动配置最优模型
switch \x3Cmodel> 切换到指定模型
status 查看当前配置
watch [--daemon] 监控并自动轮换
refresh 刷新模型缓存

平台配置

OpenRouter

需要 API Key：https://openrouter.ai/keys

export OPENROUTER_API_KEY="sk-or-v1-..."

Groq

需要 API Key：https://console.groq.com/keys

export GROQ_API_KEY="gsk_..."

Google AI Studio

需要 API Key：https://aistudio.google.com/app/apikey

export GOOGLE_API_KEY="..."

HuggingFace

需要 API Key：https://huggingface.co/settings/tokens

export HF_TOKEN="..."

Ollama

本地运行，无需 API Key：

# 安装后直接可用
ollama pull llama3.1

配置文件位置

OpenClaw 配置：~/.openclaw/openclaw.json

  • agents.defaults.model.primary：主模型
  • agents.defaults.model.fallbacks：备用模型列表

扩展新平台

在 references/platforms.md 中添加新平台支持。

故障排查

问题 解决方案
命令找不到 pip install -e . 在技能目录运行
API Key 错误 检查对应平台的环境变量
配置不生效 运行 openclaw gateway restart
模型列表为空 运行 free-model-finder refresh 刷新缓存

相关资源

Usage Guidance
This skill claims to discover and auto-rotate across live provider APIs, but the implementation is mostly static: it shows hard-coded model lists and only edits ~/.openclaw/openclaw.json. Before installing or running: (1) back up ~/.openclaw/openclaw.json, (2) inspect scripts/main.py yourself to confirm it only changes the fields you expect, (3) be aware the 'watch' / rate-limit monitoring features are not implemented and the skill will not actually call provider APIs even if you set API keys, (4) if you want true dynamic discovery/rotation, demand implementation of API calls and safe retry/monitoring logic, and (5) install with care (pip install -e .) and run commands in a test environment first.
Capability Analysis
Type: OpenClaw Skill Name: free-model-finder Version: 1.0.0 The skill is a legitimate utility designed to help users find and configure free or low-cost AI models across various platforms. The core logic in `scripts/main.py` is transparent, using standard Python libraries to read and update the local OpenClaw configuration file (`~/.openclaw/openclaw.json`) and verify the presence of API keys in environment variables without exfiltrating them. No evidence of malicious execution, data exfiltration, or prompt injection was found.
Capability Assessment
ℹ Purpose & Capability
Name/description promise: discover, compare and auto-configure multi-platform free/low-cost models. Actual code: provides a CLI that shows a static list of models (PLATFORM_MODELS), writes model selections into ~/.openclaw/openclaw.json, and checks for presence of a few API env vars. It does not actually query provider APIs or dynamically discover models. Modifying OpenClaw config is coherent with the stated purpose.
⚠ Instruction Scope
SKILL.md describes active scanning, rate-limit monitoring, automatic rotation (watch/daemon), and 'auto' behavior across platforms. The runtime code does not implement network queries, monitoring, or automatic watch/rotation — it only reads/writes the OpenClaw config and prints suggested commands. The skill instructs users to set API keys but the code only reports whether those env vars exist and does not use them to call provider APIs. It also writes to ~/.openclaw/openclaw.json (expected) — users should be aware that the skill will modify that config file.
✓ Install Mechanism
No external download URLs or extract/install from unknown hosts. Package is a local Python package with an entry point (setup.py) and standard pip install -e usage. skill.json includes an 'install' helper that runs pip in a specific workspace path — unusual but not dangerous by itself.
ℹ Credentials
The skill declares no required env vars at registry metadata level, but SKILL.md and skill.json reference optional API keys (OPENROUTER_API_KEY, GROQ_API_KEY, GOOGLE_API_KEY, HF_TOKEN). The code only checks for their presence (no exfiltration or use). Requesting/reading those env vars is proportionate to a model-management tool, but they are optional and not required for the CLI to run.
ℹ Persistence & Privilege
The skill writes to the user's OpenClaw config (~/.openclaw/openclaw.json), adding/changing agents.defaults.model and agents.defaults.models. This is within the scope of a model-switching tool but is a privileged action (it alters agent behavior). The skill is not always-enabled and does not autonomously restart services; it only prints a restart command for the user.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install free-model-finder
  3. After installation, invoke the skill by name or use /free-model-finder
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Free Model Finder - 初始版本上线 - 支持扫描和列出多平台（OpenRouter、HuggingFace、Groq、Google AI Studio、Ollama）免费/低价 AI 模型 - 可按价格、速度、质量、稳定性排序对比模型 - 提供一键自动配置主模型与备用模型，并支持自动重启 Gateway 生效 - 实现模型可用性监控与 rate limit 检测，自动切换备用模型 - 支持通过命令行查询、切换、刷新模型，查看当前配置和切换历史 - 提供详细平台接入说明及常见问题排查指南
Metadata
Slug free-model-finder
Version 1.0.0
License MIT-0
All-time Installs 2
Active Installs 2
Total Versions 1
Frequently Asked Questions

What is 发现、对比和配置多平台免费/低价 AI 模型?

发现、对比并自动配置多平台免费或低价 AI 模型，支持自动轮换以应对限流和优化成本。 It is an AI Agent Skill for Claude Code / OpenClaw, with 375 downloads so far.

How do I install 发现、对比和配置多平台免费/低价 AI 模型?

Run "/install free-model-finder" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is 发现、对比和配置多平台免费/低价 AI 模型 free?

Yes, 发现、对比和配置多平台免费/低价 AI 模型 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 发现、对比和配置多平台免费/低价 AI 模型 support?

发现、对比和配置多平台免费/低价 AI 模型 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 发现、对比和配置多平台免费/低价 AI 模型?

It is built and maintained by Criss_Su (@sucriss); the current version is v1.0.0.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463815 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259802 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200680 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191345 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190333 · by oswalpalash

💬 Comments