Reusable setup prompts for optimizing Claude Code documentation. Achieve 90% token savings on any project in 5 minutes.

Claude Code plugin that tracks token usage, identifies wasted context, and saves 30-50% on API costs. Heatmaps, ROI reports, budget alerts, efficiency scores, git-aware suggestions — all local, zero config.

45% cost reduction measured. The only Claude Code plugin built from CC source analysis — cache expiry prevention, SubTask auto-delegation, zero-cost context restoration, real-time dashboard. Max Plan + API pay-per-use.

Local-first context compression for AI coding tools. One binary saves 85-93% of redundant tokens across every LLM call.

reShapr website

Claude Code CLI skill that delegates complex tasks to an OpenCode subagent via ACP protocol, saving 50-90% tokens.

Open-source library of token-efficient prompts — 18 prompts, 14 categories, 3 variants each (Lean/Balanced/Max Quality). Covers code, research, creative writing, career, mental health, and more.

Techniques to optimize token usage on GitHub Copilot

50%+ fewer input tokens. 20%+ shorter output. Do more work in the same context window.

Chrome Extension that lets you continue any AI conversation anywhere—without losing context. No more copy-paste—move full context across AI tools.

Reduce noisy shell, CI, diff, and MCP-adjacent output into compact answers your coding agent can actually use. Alembic is a local, skill-first tool for Codex and Claude that cuts context waste without adding a network dependency.

serena MCP (38% less tokens). Quick setup: npx serena-slim --setup

The missing Middleware for reducing LLM API costs through TOON format by converting JSON to TOON automatically with 30-60% token savings with no code changes.

Diagnose Claude Code context burn and generate practical fixes as a tiny npm/npx-ready CLI.

Compress LLM prompts 30-60% — CLI, Claude Desktop MCP, browser extension. Zero API calls. Zero cost.

Break through Cursor / Windsurf quota limits and achieve true autonomous programming loops. 10x-Agent-Loop enables infinite interactions within a single Fast Request through file-watching technology, while providing a Consultant Mode that leverages external AI models (Gemini/Claude) to break through technical bottlenecks.

Brain Synapse A cognitive memory OS for AI agents, featuring biologically-inspired dual-track retrieval, associative recall, and low-latency memory access. Designed for next-generation agent frameworks like OpenClaw.

⚡ Semantic compression for IDE↔LLM communication. Save 80%+ tokens with radical glyphs. Supports OpenAI, Claude, VS Code, Antigravity.

Auto-learning codebook compression for AI agent memory. Makes context files 2-10x smaller while staying fully readable by any LLM. Zero deps, pure JS.

A fully local toolkit for OpenClaw, Claude Code, Codex, Gemini, and other heavy-token users to reduce token waste, improve performance, and detect security risks.