优化Token消耗 AI Skills
23 skills, sorted by popularity. Let AI handle 优化Token消耗 for you.
Agents
A framework-agnostic AI Agent debugging and optimization tool
agent-expenditure
Monitor token consumption in multi-agent parallel scheduling, identify waste, and optimize efficiency.
agent-expenditure
Monitor token consumption in parallel agent scheduling, identify waste, and optimize scheduling strategies.
AGI记忆模组
Building Persistent Memory and Context Engineering Capabilities for AI Agents
AGI记忆模组
A meta-skill that provides persistent memory and context engineering capabilities for AI agents.
Agented
A stateful text editor designed specifically for LLMs
Agented
A command-line text editor designed for LLMs
Agent Mailbox Light
Provides a lightweight, cross-workspace internal mail system for temporary coordination among multiple AI agents.
Agent Mailbox Light
A lightweight, cross-session ephemeral coordination mailbox for multi-agent workspaces.
Agent Mailbox Light
A lightweight, cross-workspace local mailbox system for multi-agent temporary coordination and context sharing.
Agent Mailbox Light
A lightweight, ephemeral internal mail coordination system for multi-agent workspaces.
Agent Mailbox Light
A lightweight, cross-session ephemeral coordination mailbox for local AI workspaces.
@4fu/pi-bin-hints
Automatically detects modern Unix tools and injects AI prompts
@2008muyu/pi-plan
A two-stage planning workflow: strong model for planning, lightweight model for execution.
@adamjen/pi-vcc
Algorithmic Dialogue Compressor with Structured Summaries
agent-context-card
Long-session context projection to reduce token consumption
@adamjen/pi-compact-fast
Compress Pi sessions with a fast local model to save tokens.
@a-canary/pi-upskill
Learn from failures, automatically reduce token waste
agent-halter
Set session budgets for AI coding agents to control time and conversation turns.
9router
Intelligent AI Router and Token Saver, connecting various coding tools to 40+ AI providers.
agent-registry
A lazy loading system for Claude Code that provides on-demand Agent loading, significantly saving context tokens.
AgentAutoFlow
A customizable AI coding team instruction set designed to reduce overall token consumption through intelligent planning.
agentic-mcp-skill
Optimizes token consumption during MCP tool invocation through a three-tier progressive loading approach.