relayplaneSave ~$47/mo on a typical $100/mo OpenClaw spend. Local proxy routes simple tasks to cheaper models automatically.
Install via ClawdBot CLI:
clawdbot install RelayPlane/relayplaneSpend $100/mo on OpenClaw? Keep $47 of it. 835+ developers already installed.
RelayPlane is a local proxy that saves you real money by routing simple LLM tasks to cheaper models automatically. Complex reasoning stays on Opus — file reads, status checks, and simple edits go to Haiku. One install, no code changes, automatic fallback if anything fails.
Pro pays for itself at $60/mo API spend. If you're spending $100+/mo, expect to save $40-70/mo.
Install the proxy globally:
npm install -g @relayplane/proxy# 1. Start the proxy
relayplane-proxy
# 2. Point OpenClaw at it (add to your shell config)
export ANTHROPIC_BASE_URL=http://localhost:3001
export OPENAI_BASE_URL=http://localhost:3001
# 3. Run OpenClaw normally - requests now route through RelayPlaneOnce installed, use the CLI directly:
| Command | Description |
|---------|-------------|
| relayplane-proxy | Start the proxy server |
| relayplane-proxy stats | View usage and cost breakdown |
| relayplane-proxy telemetry off | Disable telemetry |
| relayplane-proxy telemetry status | Check telemetry setting |
| relayplane-proxy --help | Show all options |
The proxy runs on localhost:3001 by default. Configure via CLI flags:
relayplane-proxy --port 8080 # Custom port
relayplane-proxy --host 0.0.0.0 # Bind to all interfaces
relayplane-proxy --offline # No telemetry, no network except LLM APIs
relayplane-proxy --audit # Show telemetry payloads before sendingSet your API keys before starting:
export ANTHROPIC_API_KEY=sk-ant-...
export OPENAI_API_KEY=sk-...
# Optional: Google, xAI
export GEMINI_API_KEY=...
export XAI_API_KEY=...relayplane-proxy telemetry offrelayplane-proxy --offlineGenerated Mar 1, 2026
A customer service platform uses RelayPlane to route simple ticket categorization and FAQ responses to cheaper models like Haiku, while reserving Opus for complex complaint resolution and sentiment analysis. This reduces monthly API costs by 40-50% while maintaining high-quality support for intricate issues.
Marketing agencies leverage RelayPlane to automate drafting social media posts and basic blog outlines with cost-effective models, using premium models only for creative campaign ideation and brand voice refinement. This optimizes spending across high-volume content workflows without sacrificing quality for key tasks.
Edtech companies integrate RelayPlane to handle routine quiz generation and grammar checks via cheaper models, while employing advanced models for personalized learning recommendations and essay grading. This enables scalable, affordable AI features in educational applications.
Healthcare providers use RelayPlane to process simple medical note summarization and appointment reminders with low-cost models, reserving expensive models for complex diagnostic support and patient interaction analysis. This cuts API expenses while ensuring accuracy in critical clinical decisions.
RelayPlane offers a freemium model where basic proxy features are free, with paid tiers for advanced analytics, multi-team support, and priority routing. Revenue is generated from monthly subscriptions targeting businesses with high API usage, typically priced at $10-50 per user per month.
The tool provides enterprise licenses for large organizations needing custom integrations, dedicated support, and on-premise deployment options. This model generates high-value contracts through one-time fees or annual subscriptions, focusing on sectors like finance and healthcare with strict compliance needs.
RelayPlane partners with cloud providers and AI service platforms to offer bundled solutions, earning commissions on referred API usage or co-marketing deals. This diversifies income by leveraging existing ecosystems and driving adoption through strategic alliances.
💬 Integration Tip
Set environment variables for API keys before starting the proxy to ensure seamless routing, and use the offline mode for enhanced privacy in sensitive deployments.
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Search and analyze your own session logs (older/parent conversations) using jq.
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linking related objects, enforcing constraints, planning multi-step actions as graph transformations, or when skills need to share state. Trigger on "remember", "what do I know about", "link X to Y", "show dependencies", entity CRUD, or cross-skill data access.
Ultimate AI agent memory system for Cursor, Claude, ChatGPT & Copilot. WAL protocol + vector search + git-notes + cloud backup. Never lose context again. Vibe-coding ready.
Headless browser automation CLI optimized for AI agents with accessibility tree snapshots and ref-based element selection