Neura Market
ZengLiangYi/ChatCrystal logo

ZengLiangYi/ChatCrystal

Free

面向 AI 编程对话的本地优先 PKM 记忆服务器，从 Claude Code、Cursor、Codex CLI、Trae、GitHub Copilot 导入对话，生成结构化笔记、标签图谱、语义搜索、Markdown 导出，并通过 MCP 提供任务记忆召回/写回。

FreeFree tier
Type
Open Source
Visit Website GitHub (40 stars)

About ZengLiangYi/ChatCrystal

ChatCrystal is a local-first AI personal knowledge management (PKM) app designed for developers who use AI coding assistants like Claude Code, Cursor, Codex CLI, Trae, and GitHub Copilot. It imports conversations from these tools, distills them into structured notes with titles, summaries, conclusions, snippets, and tags, and enables semantic search with embedding-based relation-aware result expansion. Users can build a tag knowledge graph where knowledge points are tags connected by normalized co-occurrence, export notes to Markdown, and expose CLI and MCP tools so AI agents can recall and write back reusable experience. It runs locally with configurable LLM and embedding providers, and offers desktop (Windows), CLI, and Docker deployment options.

Key Features

Imports AI coding conversations from Claude Code, Cursor, Codex CLI, Trae, and GitHub Copilot
Distills conversations into structured notes with titles, summaries, conclusions, code snippets, and tags
Semantic search with embeddings and relation-aware result expansion
Builds a tag knowledge graph where knowledge points are tags and edges show normalized co-occurrence
Exposes CLI and MCP tools so AI agents can recall and write back reusable experience
Runs locally with configurable LLM and embedding providers
Cross-platform deployment: desktop app (Windows), CLI/Web, and Docker Cloud

Pros & Cons

Pros
  • Local-first and privacy-preserving – no cloud dependency required
  • Supports multiple major AI coding tools (Claude Code, Cursor, Codex CLI, Trae, Copilot)
  • Rich feature set: structured note-taking, semantic search, knowledge graph, MCP memory
  • Open source and free to use with no paid tiers
  • Flexible deployment options: desktop app, CLI, or Docker
Cons
  • Requires manual configuration of LLM and embedding providers before use
  • Desktop app currently only available for Windows (no macOS/Linux desktop builds mentioned)
  • Dependency on external LLM and embedding providers – may incur costs for API usage
  • Small community and relatively new project (51 stars on GitHub as of content)

Best For

Developers using AI coding assistants who want to preserve and reuse conversation knowledgeBuilding a personal knowledge base from AI interactions across multiple toolsSemantic search across past coding conversations to find solutions quicklyEnabling AI agents to recall and write back reusable experience via MCP integrationKnowledge management for teams using AI-assisted development workflows

FAQ

What AI coding tools does ChatCrystal support?
ChatCrystal can import conversations from Claude Code, Cursor, Codex CLI, Trae, and GitHub Copilot.
Is ChatCrystal free?
Yes, ChatCrystal is open source and free to use. You can download it from GitHub or install via npm.
How do I install ChatCrystal?
You can install the desktop app via the Windows installer from GitHub Releases, or use npm (npm install -g chatcrystal) for CLI/Web access, or deploy via Docker for multi-device use.
Does ChatCrystal require an internet connection?
ChatCrystal is local-first, but it requires an LLM provider and an embedding provider to be configured for summarization and semantic search, which typically need API calls.