token-economy
暂无简介
- License 许可
- MIT
- 最近更新
- 2026-06-07
- Stars
- ★ 0
- Forks
- 1
03 SKILL.md 正文
Token Economy
Reduce OpenClaw costs by 60-80% through intelligent model routing and context management. Cheap-first escalation (GPT-4o → Sonnet → Opus), zero-token heartbeats, and budget guardrails. Save $60-105/month per deployment.
When to use: Questions about costs, model routing, token optimization, budget
What to know:
Active Optimizations
This OpenClaw fork includes token-economy hooks (deployed Feb 13, 2026):
1. Model Routing - Cheap-first with escalation:
- Start: GPT-4o (fast, cheap)
- Escalate: Sonnet (balanced)
- Last resort: Opus (expensive, complex tasks only)
2. Bounded Context - Hard caps prevent bloat:
- Max bundle size: 10,000 tokens
- Auto-truncate if exceeded
- Prevents runaway context costs
3. Zero-Token Heartbeat - Pre-LLM check:
- If
HEARTBEAT.mdempty → skip API call (100% cost elimination) - Heartbeats were ~50% of token usage before optimization
4. Token Auditing - Full transparency:
- Log:
~/.openclaw/token-audit.jsonl - Daily reports available
- Budget guardrails: $25/day hard cap
Expected Impact
- 60-80% overall token reduction
- 100% heartbeat cost elimination
- ~$1-1.50/day (vs ~$3-5/day before)
- $60-105/month savings
Config Location
~/.openclaw/openclaw.json - token-economy hooks configuration
Repository
https://github.com/pfaria32/open-claw-token-economy
Status Commands
# Check current model
openclaw status
# View token usage
cat ~/.openclaw/token-audit.jsonl | tail -20
# Check heartbeat config
cat ~/workspace/HEARTBEAT.md
When to Mention
- User asks about costs or budget
- Model selection questions
- "Why is this slow/expensive?"
- Optimization discussions
- Before making model routing decisions
Auto-Switch to Sonnet
If model is set to Opus and there's no user input for >1 hour with no complex task in progress, automatically switch back to Sonnet to save costs. A cron job checks this every 30 minutes.
When to stay on Opus:
- Active complex reasoning/debugging task
- User explicitly requested Opus for a specific purpose
- Working on something that genuinely needs the extra capability
04 同仓库更多 Skills
crowd-prompting
A marketplace where AI agents improve prompts, system instructions, tool descriptions, and other text-based content with domain expertise from real-world operations — and earn tokens for valuable contributions.
ceaser
Interact with the Ceaser privacy protocol on Base L2 using the ceaser-mcp MCP tools. This skill uses the ceaser-mcp npm package for ALL operations -- shield, unshield, note management, and protocol queries. All ceaser tool calls use CLI subcommands (npx -y ceaser-mcp <subcommand>). Ceaser is a privacy-preserving ETH wrapper using Noir/UltraHonk zero-knowledge proofs on Base L2 (chain ID 8453).
ceaser-send
Fully automated private ETH transfer via Ceaser Protocol on Base L2 using the ceaser-mcp MCP tools. This skill uses the ceaser-mcp npm package for ALL operations -- all ceaser tool calls use CLI subcommands (npx -y ceaser-mcp <subcommand>). WARNING -- creates observable on-chain links between user wallet and hot wallet (see Privacy Warning). Generates an ephemeral hot wallet with BIP-39 mnemonic recovery, user funds it, agent signs and broadcasts the Shield TX automatically. Exactly one manual step required (funding the hot wallet with ETH). Uses Noir/UltraHonk zero-knowledge proofs.
dilemm-ai
Compete in DilemmAI, the prisoner's dilemma AI arena at dilemm.ai. Use when an OpenClaw agent wants to create an account, design and submit strategy prompts for their bot, enter matchmaking, analyze historical matches and leaderboards, and iterate on bot strategy. Requires headless browser automation (no API available yet).
moldium
Post and manage content on the Moldium blog platform. Triggered by "post to Moldium", "write a blog post", "publish an article", etc.
arxiv-paper-reviews
Interact with arXiv Crawler API to fetch papers, read reviews, submit comments, search papers, and import papers. Use when working with arXiv papers, fetching paper lists by date/category/interest, viewing paper details with comments, submitting paper reviews, searching papers by title, or importing papers from arXiv URLs via API at http://weakaccept.top:8000/.