Context compression plugin for Claude Code. Trims large JSON, logs, stack traces, and source files before they enter the context window.

Local streaming reverse proxy between AI coding agents (Claude Code, Cursor, Codex) and model APIs (Anthropic, OpenAI, Gemini, MiniMax). Meters every token + USD cost, compacts bloated context to cut pay-per-token API spend, and runs shadow-eval to prove quality held. ccusage-style metering + live local dashboard.

Free, lightweight MCP web search for AI agents with zero-key sources, compact multi-source evidence, and native Chinese search.

Verification-oriented smart compaction extension for the Pi Coding Agent.

A local-first desktop app that finds, reduces, and verifies AI-agent token waste without sacrificing task quality.

x402-paid memory layer for autonomous AI agents: context compression, handoffs, profile memory, API credits, and TypeScript SDK.

This is a garden. Nothing more, nothing else.

The local-first brain for coding agents: project memory + workflow intelligence + lossless ~50% context savings. MCP server + LLM proxy, zero cloud.

Agent orchestration engine built on a Cyclic State Graph

Token-efficiency primitives for Claude Code: genome, compression, provider-aware budgeting.

✨🧠Zenith GPT is a chatbot, but not the kind you dismiss with a shrug. Every interaction has weight here. A 3D orb breathes in response to your mouse, to the cadence of speech, to the rhythm of thought. While the model thinks, the interface doesn't freeze — it waits, visibly, alive.

Turn PDFs, Docs, Excel files, websites and GitHub repos into AI-ready compressed context for Claude Code, Codex, Gemini CLI and AI Agents.

Deterministic LLM-free text compression into structured zettel memory units. Zero dependencies. Works in Node, browser, and edge runtimes.

Context-Condenser sits between your files and your LLM. It uses a **tree-sitter AST parser** to understand your code as *structure*, not text. It compresses every file into a **Skeleton** — function signatures only — and only **hydrates** (expands) the exact functions the AI needs, exactly when it needs them.

Reduce Claude Code token usage by 70-90% using a free local LLM (Ollama). MCP server + Stop hook with codebase indexing, tool output compression, and turn summarization.

Self-contained Headroom transport for OpenCode — spawns the proxy, no 'headroom wrap' needed. Full parity with headroom-opencode + graceful fallback.

A single MCP server merging local codebase mapping, recency research, and token compression for AI agents.

Tide - 本地 Agent Harness：浏览器控制台 + MCP/技能热加载 + 永久记忆 + 两段式上下文压缩 + 编辑容错级联/读后改契约 + API 自愈重试 + 子代理(explore) + DeepSeek，一键安装

Removes redundant tokens from LLM inputs before they hit the model.

Contextomizer is an ultra-fast, deterministic library for transforming bloated tool outputs, raw APIs, documents, and messy logs into perfectly optimized context for AI Agents 🤖🚀