oc-skill-routerSmart LLM routing brain for OpenClaw. Auto-dispatches tasks to Claude, GPT, Gemini, DeepSeek, Kimi via Evolink API. Cascade strategy cuts costs 60-85%. One A...
Install via ClawdBot CLI:
clawdbot install evolinkai/oc-skill-routerGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Sends data to undocumented external endpoint (potential exfiltration)
POST → https://direct.evolink.ai/v1/messages`Calls external URL not in known-safe list
https://evolink.aiAI Analysis
The skill's external API calls are documented and consistent with its stated purpose of routing LLM requests through Evolink's service. While it sends user prompts to a third-party proxy, this is a declared core function, not hidden exfiltration. The primary risk is dependency on the security practices of the external API provider.
Audited Apr 17, 2026 · audit v1.0
Generated Mar 22, 2026
Automates handling of customer inquiries by routing simple Q&A to lightweight models like Claude Haiku for quick responses, escalating to balanced models like Claude Sonnet for complex troubleshooting, and using flagship models like Claude Opus for high-stakes complaint resolution. Reduces response times and operational costs by 60-85% through smart routing.
Streamlines content production by assigning tasks to appropriate models: Tier 1 for quick social media posts and grammar checks, Tier 2 for article writing and SEO optimization, and Tier 3 for in-depth research reports and strategic planning. Ensures cost-effective scaling across multiple clients with transparent model selection.
Enhances coding workflows by routing bug fixes and simple refactoring to Tier 1 models, feature development and testing to Tier 2 models like Claude Sonnet, and architecture design or security audits to Tier 3 models like GPT-5.2. Supports cross-provider routing for specialized tasks such as math proofs with DeepSeek Reasoner.
Personalizes learning experiences by using lightweight models for quick concept explanations and quizzes, balanced models for interactive tutoring and content generation, and flagship models for complex problem-solving and curriculum design. Leverages cascade fallback to adapt to varying student needs efficiently.
Optimizes financial modeling and analysis by routing routine data processing to Tier 1 models, investment research and report generation to Tier 2 models, and high-risk strategy development or legal review to Tier 3 models like Claude Opus. Ensures compliance and accuracy with cost-effective model selection.
Offers tiered subscription plans (e.g., basic, pro, enterprise) providing access to the Evolink Router with varying usage limits and model tiers. Revenue is generated through monthly or annual fees, with potential upsells for additional API calls or premium support.
Charges users based on actual API consumption, with different rates per model tier (e.g., lower cost for Tier 1 models, higher for Tier 3). Attracts startups and small businesses by minimizing upfront costs and aligning expenses with usage.
Provides customized solutions for large organizations, including white-labeling, dedicated support, and integration with existing systems. Revenue comes from one-time licensing fees, ongoing maintenance contracts, and consulting services for setup and optimization.
💬 Integration Tip
Ensure the EVOLINK_API_KEY is securely stored in environment variables and test routing with sample tasks to optimize cost savings and performance.
Scored Jun 19, 2026
Use CodexBar CLI local cost usage to summarize per-model usage for Codex or Claude, including the current (most recent) model or a full model breakdown. Trigger when asked for model-level usage/cost data from codexbar, or when you need a scriptable per-model summary from codexbar cost JSON.
Check Antigravity account quotas for Claude and Gemini models. Shows remaining quota and reset times with ban detection.
使用豆包(火山引擎)语音合成大模型 API 将文本转换为语音音频文件。支持声音复刻音色(S_ 开头的音色ID)和官方预置音色。当用户要求"语音合成"、"文字转语音"、"TTS"、"朗读文本"、"生成语音"、"用我的声音读"、"豆包语音"、"声音复刻合成"等相关请求时,务必使用此 skill。即使用户只是说"帮我把...
Intelligent model routing for sub-agent task delegation. Choose the optimal model based on task complexity, cost, and capability requirements. Reduces costs...
自动生成科技新闻摘要。从多个来源(RSS、Twitter、GitHub、Web Search)抓取科技新闻,整合后生成摘要。
让 AI 代理根据对话内容自动选择最合适的模型。四层识别(系统过滤→关键词→指示词→语义相似度),四池架构(高速/智能/人文/代理),五分支路由,全自动 Fallback 回路。支持 trigger_groups_all 非连续词组命中。