@smyslenny/agent-memory 3.1.0 → 4.0.0-alpha.1

package/README.en.md

@@ -1,153 +0,0 @@
-# 🧠 AgentMemory v3
-
-> Structured long-term memory layer for AI agents: write, recall, decay, and auto-ingest.
-
-[![npm](https://img.shields.io/npm/v/@smyslenny/agent-memory)](https://www.npmjs.com/package/@smyslenny/agent-memory)
-[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)
-[![Node.js](https://img.shields.io/badge/Node.js-≥18-green.svg)](https://nodejs.org/)
-[![MCP](https://img.shields.io/badge/MCP-9_tools-orange.svg)](https://modelcontextprotocol.io/)
-
-**[简体中文](README.md)** | **English**
-
----
-
-## Positioning in v3
-
-AgentMemory v3 is explicitly a **structured memory companion** to OpenClaw memory-core, not a second full-stack retrieval system.
-
-- Markdown (`memory/*.md` + `MEMORY.md`) remains the human-readable source of truth
-- agent-memory is a derived, structured lifecycle layer
-
-Core capabilities:
-
-- **Typed memory model**: `identity / emotion / knowledge / event`
-- **URI path addressing**: `core://`, `emotion://`, `knowledge://`, `event://`
-- **Write Guard** for dedup/conflict gating
-- **BM25 recall** with priority × vitality weighting
-- **Sleep-cycle maintenance** via `reflect` (decay / tidy / govern)
-- **Ingest** for markdown-to-memory extraction
-- **Surface** for readonly context surfacing (no access side effects)
-- **Warm boot / reflect narrative output**
-- **Multi-agent isolation** by `agent_id`
-
----
-
-## Quick Start
-
-### Install
-
-```bash
-npm install -g @smyslenny/agent-memory
-```
-
-### CLI examples
-
-```bash
-# Initialize DB
-agent-memory init
-
-# Store memory
-agent-memory remember "User prefers dark mode" --type knowledge --uri knowledge://preferences/theme
-
-# Search
-agent-memory recall "user preferences" --limit 5
-
-# Startup boot (narrative output)
-agent-memory boot
-
-# Run sleep cycle
-agent-memory reflect all
-```
-
----
-
-## MCP Server
-
-### Example config
-
-```json
-{
-  "mcpServers": {
-    "agent-memory": {
-      "command": "node",
-      "args": ["node_modules/@smyslenny/agent-memory/dist/mcp/server.js"],
-      "env": {
-        "AGENT_MEMORY_DB": "./agent-memory.db",
-        "AGENT_MEMORY_AGENT_ID": "noah",
-        "AGENT_MEMORY_AUTO_INGEST": "1",
-        "AGENT_MEMORY_WORKSPACE": "/home/user/.openclaw/workspace"
-      }
-    }
-  }
-}
-```
-
-### MCP tools (9)
-
-- `remember`
-- `recall`
-- `recall_path`
-- `boot`
-- `forget`
-- `reflect`
-- `status`
-- `ingest`
-- `surface`
-
-> `link` and `snapshot` were removed in v3.
-
----
-
-## Auto-Ingest (file change watcher)
-
-When MCP server starts, watcher is enabled by default (`fs.watch`) for:
-
-- `~/.openclaw/workspace/memory/*.md`
-- `~/.openclaw/workspace/MEMORY.md`
-
-On file changes, ingest runs automatically (still guarded by Write Guard and dedup).
-
-Environment variables:
-
-- `AGENT_MEMORY_AUTO_INGEST`
-  - `1` (default): enabled
-  - `0`: disabled
-- `AGENT_MEMORY_WORKSPACE`
-  - default: `$HOME/.openclaw/workspace`
-
----
-
-## Recommended OpenClaw integration
-
-Use a 3-stage cron pipeline:
-
-1. `memory-sync` (14:00 / 22:00)
-   - dynamic session JSONL discovery
-   - append incremental entries to `memory/YYYY-MM-DD.md`
-   - best-effort `agent-memory.remember`
-   - emit health metrics (scan path / file count / extracted / synced)
-
-2. `memory-tidy` (03:00)
-   - markdown consolidation/distillation
-   - call `agent-memory.reflect phase=all`
-
-3. `memory-surface` (14:05 / 22:05)
-   - generate `RECENT.md`
-
-Design principle: **Markdown is source of truth; agent-memory is a derived index layer.**
-
----
-
-## Development
-
-```bash
-npm install
-npm test
-npm run build
-```
-
----
-
-## License
-
-MIT