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.
Claude Code plugin that replaces the compaction summary with Jev decisions: every tool call and result is scored in one fast request, stale ones are dropped or truncated, everything kept stays verbatim.
A compact Agent Skill for highly simplified, rounded, subtly neo-skeuomorphic IP mascot logos.
Cross-platform local MCP capabilities for ChatGPT with Chrome integration, Goal, Compact & Resume, and durable multi-agent workflows.
Find the ghost tokens. Fix them. Survive compaction. Avoid context quality decay.
- claude-code
- token-optimization
- context-window
- token-usage
Ghostty Blackhole puts a real, ray-traced black hole inside your terminal. It grows as Claude Code's context window fills up, live. A fresh session is a quiet hole in the corner. A full one swallows half your screen. You'll always see /compact coming.
Jev-powered model routing, memory, compaction, skill selection, computer and browser use for Hermes agents (also Claude Code and Codex)
A local plugin for image-heavy chats in Codex, Claude Code and other AI agents: earlier images become short placeholders instead of going out with every message, the model fetches the original back by its ID (also after compaction), and the prompt cache keeps working.
Remory: Learning Residual Memory for Context Compaction
- context-compaction
- inference
- llm-agents
- residual-memory
TranslatedCompress while working (边做边压) Live Compact. Compress the finished turn in place and keep the prefix cache.
An agent harness built for teaching: ~1,400 lines of Python with goal mode, permissions, skills, MCP and Claude's server-side compaction
AI-managed Auto Compact for DeepSeek Harness: the agent decides when to compact context and what to retain based on task progress, without waiting for a fixed token threshold.