Piia Engram MCP Server

@Patdolitse

代码仓库主页

基本信息

分类

开发工具

许可证

AGPL-3.0

传输方式

stdio

作者

Patdolitse

提交者

Patdolitse

关于 Piia Engram MCP Server

Local-first AI memory you can see, edit, and override — portable across Claude Code, Codex, Cursor, Windsurf, and other MCP coding tools.

#memory #identity #local-first #knowledge

Piia Engram MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "engram": {
      "command": "python",
      "args": [
        "-m",
        "piia_engram.mcp_server"
      ],
      "env": {
        "ENGRAM_TOOLS": "all"
      }
    }
  }
}

Piia Engram MCP Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表，不代表服务没有工具。请查阅项目文档了解支持的能力。

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Piia Engram?

Piia Engram is a local-first AI work identity layer for developers. It stores your personal preferences, coding standards, lessons learned, decisions, and project context in local files under ~/.engram/ as JSON and Markdown. Multiple MCP-compatible coding tools (Claude Code, Codex, Cursor, Windsurf, and others) can read this same approved context, so you never have to re-explain yourself when switching tools or starting new chats. It is not an agent memory database; it is a user-owned layer above your tools.

How to use Piia Engram?

Install with pip install piia-engram and run engram setup. The wizard auto-detects supported AI tools, lists the config files it will touch, writes the MCP connection after confirmation, and backs up every file before modifying it. After restarting your configured tool, its first conversation can load your approved identity and knowledge through startup or search tools.

Key features of Piia Engram

Local files you own, no cloud account required
AI-proposed knowledge reviewed by user before high-risk writes
Portable identity across multiple MCP-compatible coding tools
Memory Lens (engram preview --html) shows what AI callers receive
Field-level optional AES-256-GCM encryption for sensitive profile fields
17 core MCP tools (loaded by default) + 40 advanced tools (opt-in)

Use cases of Piia Engram

Developers who switch between Claude Code, Codex, and Cursor and want consistent context
Users who open many AI chat windows per week and want each to start from the same identity
Developers who lost preferences after a tool update and want persistent local memory
System architects maintaining living Architecture Decision Records queryable by AI tools
Vibe coders who want consistent patterns and preferences across sessions

常见问题

How is Piia Engram different from native AI tool memory?

Native memories (Claude Code, Codex, Cursor) are scoped to one tool or workspace. Piia Engram provides one portable identity layer above them: local files you own, AI-proposed knowledge you review, and context that follows you across tools.

What are the dependencies and runtime requirements?

Python 3.10 or higher. The core store runs locally with no network calls by default. Optional features like `read_web_content` may make network requests; remote telemetry requires explicit opt-in.

Where does my data live?

All identity and knowledge files are stored under `~/.engram/` as plain JSON and Markdown by default. You can inspect, edit, and override them at any time. Encryption is optional for specific profile fields.

Does Piia Engram require approval for every AI write?

By default, high-risk items (credentials, shell commands, MCP config, permission rules) wait for your review; low/medium writes are auto-absorbed but fully auditable and reversible. Set `ENGRAM_APPROVAL=strict` to gate every write.