Compaction

"Work appears completed or recorded. Run /compact to save tokens."

FirstCoder is an open-source local coding agent that’s useful enough to run, small enough to read end to end, and structured enough to learn from. If you’re building your own agent, it’s a practical reference—not a black box.

Zero-config Claude Code setup with enforced task scaffolding, structured memory, persistent context after compaction, plug-in code standards, optional TDD mode, and zero behavior changes for developers.

Find the ghost tokens. Fix them. Survive compaction. Avoid context quality decay.

Use ChatGPT Pro as a native model in the Codex app — with context, compaction, streaming, images, and optional MCP tools.

The agent harness you can fork. A working agent loop with compaction, sessions, approvals, a terminal REPL, a local dashboard and ACP, where every tool, skill, hook and policy is a file you can change. Build your own internal or coding agent on any model.

The harness layer for Claude Code — a reference implementation of harness engineering with hook-enforced dual review, state-machine gates that survive context compaction, and fail-closed safety where it counts. Quality gates that AI can't skip.

General-purpose agent in one static Go binary: console TUI, ACP server for editors, OpenAI-compatible API with embedded web UI, Telegram gateway, cron scheduler, swarm relay, remote mode. Rules, skills, subagents, hooks, MCP, background tasks, long-term memory, context compaction. Any OpenAI-compatible provider, distroless-ready.

Stop re-explaining yourself to AI. MARM offers persistent memory, cross-agent context sharing, write queues, swarm-ready presets, and compaction for clean recall. Includes a live web dashboard for managing memories, logs, and sessions. MCP connects over HTTP & STDIO and works with major LLMs like Claude, Codex, Grok, Gemini, and other models.

Spec-driven development with smart compaction. Claude Code plugin combining Ralph Wiggum loop with structured specification workflow.

Pre-compaction hooks + a CAS-guarded task pointer that keep Codex agents on-goal across context auto-compaction — no mainline drift, no duplicate work, no conflicting progress in long runs.

Persistent file-based planning for AI coding agents and long-running tasks. Crash-proof markdown plans, session recovery after /clear and compaction, per-turn re-injection against context rot, deterministic completion gate. Manus-style. Install from npm, the Claude Code plugin marketplace, or npx skills. Codex, Cursor, OpenCode, 60+ agents.

Persistent local memory for AI coding agents. Preserve decisions and project context across sessions, compaction, and tools.

Drop-in memory harness for AI agents — 3-tier memory, compaction tree, hybrid search. One command to set up. Works with Claude Code and OpenClaw.

A local plugin for image-heavy chats in Codex, Claude Code and other agents: earlier images stop going out with every message, and the model fetches the one it needs by its ID, also after compaction.

Claude Code memory, Codex memory, Cursor memory, OpenCode memory, Gemini CLI memory, Windsurf memory & Cline memory across 43 agents. AI coding memory, Claude Code history, Cursor history, local coding agent memory, MCP memory server & Claude compaction.

The open-source alternative to Paper.design. A multiplayer design canvas where humans and AI agents design together, live. MCP built in.

Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.

Context window optimization for AI coding agents. Sandboxes tool output (98% reduction), persists session memory, and enforces routing across 17 platforms via MCP + hooks.

Turn Claude Code into a full game dev studio — 49 AI agents, 72 workflow skills, and a complete coordination system mirroring real studio hierarchy.

Never stop coding. Free MIT AI gateway: one endpoint, 375 providers (150+ free), 1,400+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 900+ contributors

Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.

A collection of projects designed to help developers quickly get started with building deployable applications using the Claude API

Academic Research Skills for Claude Code: research → write → review → revise → finalize

Compact the current conversation into a handoff document for another agent to pick up.

AINL helps turn AI from "a smart conversation" into "a structured worker." It is designed for teams building AI workflows that need multiple steps, state and memory, tool use, repeatable execution, validation and control, and lower dependence on long prompt loops. AINL is a compact, graph-canonical, AI-native programming system for (READ: README)