agent-memory-bridge
extensionmaintainedCross-session memory for pi-coding-agent (pi-mono). A thin client for the local claude-mem worker: records tool activity, injects relevant past context, and registers a memory_recall search tool. Share memory across pi, Claude Code, Cursor, Codex and Open
by — · v1.0.4 · published 1w ago
$ pi install npm:agent-memory-bridgeSignals
Download trend
418 downloads · last 12 weeks (weekly)
README
agent-memory-bridge (pi adapter)
持久记忆,随拿随用。 一个为 pi-coding-agent 打造的自我维护记忆扩展。自动记录你的每轮对话、工具调用与助手回复,跨会话、跨引擎调用同一份可搜索记忆。
This is the pi adapter of the
agent-memory-bridgeproject — a bridge that gives every AI coding agent you use (OpenCode, CodeBuddy, pi, Hermes) one shared memory.
✨ Highlights
| 亮点 | 说明 |
|---|---|
| 🔄 多 Agent 适配 | 与 OpenCode / CodeBuddy / Hermes 共享同一记忆库,pi 里记的,其它 agent 也能搜到 |
| 🚀 一键安装 | pi install npm:agent-memory-bridge 一条命令,npm 市场直接装 |
| 📥 自动捕获 | 自动记录用户提问、工具调用、助手回复、会话结束,零手动操作 |
| 🔌 后端可拔插 | 底层可切换 claude-mem / mem0 / both,一个环境变量搞定 |
| 🌐 中英双语 | 中英双语文档 |
Install (npm registry)
# 1. make sure the worker is up
curl -s http://127.0.0.1:37701/api/health
# 2. install from the npm registry (recommended)
pi install npm:agent-memory-bridge
# or install from source (local path):
cd /path/to/agent-memory-bridge/agents/pi
pi install "$PWD"
pi install appends the package to the packages array of
~/.pi/agent/settings.json:
{
"packages": [
"npm:agent-memory-bridge"
]
}
Restart pi. At startup it logs the resolved worker URL, e.g.
[pi-claude-mem] worker → http://127.0.0.1:37701 (端口来源: settings.json).
Then run the /memory-status slash command — it should report the worker
version and degraded: false.
Uninstall
pi remove npm:agent-memory-bridge
then restart pi.
How it works
pi-coding-agent
└── this extension (extensions/pi-claude-mem.ts)
session start local session id / project name
before agent start -> POST /api/sessions/init (with prompt)
context -> GET /api/context/inject
tool result -> POST /api/sessions/observations (fire-and-forget)
agent end -> POST /api/sessions/summarize
memory_recall tool -> GET /api/search
/memory-status -> GET /api/health
│
▼
local claude-mem worker (default 127.0.0.1:37701)
SQLite (FTS5) + Chroma, shared across engines
Every observation is tagged platformSource: "pi-agent".
It is a thin client for the local claude-mem worker HTTP API: it records
tool activity to the worker, injects relevant past context each turn, and
registers a memory_recall search tool. pi itself does no summarization — the
worker at 127.0.0.1:37701 handles summarization, embeddings and vector search.
This is a fork, not an independent reimplementation. It derives from the ArtemisAI
pi-agent-memory@0.3.4adapter, which itself derives from the OpenClaw plugin shipped with claude-mem. See NOTICE for the full provenance and the list of modifications.
Configuration
| Variable | Default | Description |
|---|---|---|
CLAUDE_MEM_HOST | 127.0.0.1 | Worker host |
CLAUDE_MEM_PORT | 37777 (fallback) | Worker port; the extension also reads CLAUDE_MEM_WORKER_PORT from ~/.claude-mem/settings.json (accepts both number and numeric-string) |
CLAUDE_MEM_DATA_DIR | ~/.claude-mem | Worker data dir (where settings.json is read from) |
PI_MEM_PROJECT | derived from cwd (pi-<dir>) | Project grouping for observations |
PI_MEM_DISABLED | — | Set to 1 to disable the extension |
Resolution order for the port: CLAUDE_MEM_PORT env →
CLAUDE_MEM_WORKER_PORT in ~/.claude-mem/settings.json → fallback 37777.
The runtime only imports the two peer modules pi injects
(@mariozechner/pi-ai, @mariozechner/pi-coding-agent) plus Node built-ins, so
the directory needs no node_modules. The @earendil-works pi distribution
exposes compatible extension APIs and loads the package as-is.
License
AGPL-3.0-or-later — see LICENSE. The original copyright and
notices must be retained. Note this is stricter than the rest of
agent-memory-bridge, which is MIT; this agents/pi/ subdirectory is a
self-contained AGPL-licensed component.
The claude-mem worker backend is a separate program (Apache-2.0) that this extension only talks to over a local HTTP API; it is neither contained in nor published by this package.
For a from-scratch machine setup (installing pi, the worker and the AI provider) see DEPLOY.md.