NGUYEN VIET NAM

neural-memory

AI 智慧體v4.63.0
作者:NGUYEN VIET NAM更新於 2026-10-23共 57 個版本 · 最新版本釋出於 2026-10-23@clawhub_nhadaututtheky/neural-memory
上下文管理記憶增強#Learning

據 AlphaMoat 統計,截至 2026-10,neural-memory 累計下載 29,887 次、獲 13 個星標,按累計下載量在本次參與排名的 10,914 個 AI 技能中排名第 918 位,在「AI 智慧體」分類(1,632 個)中排名第 141 位。

列表預設按累計下載量排序,也可按星標排序;詳情頁排名按累計下載量計算。

累計下載
3萬
星標
13
全站下載量排名
#918
全站前 9% · 「AI 智慧體」前 9%
平台累計下載星標評論
ClawHub ↗1.9萬10-
SkillHub ↗3萬13—

各平台的計數口徑不同,統計時間不一致。

這個技能是做什麼的

記憶組成神經圖,常被同時訪問的記憶之間會加強連線,並帶有衰減、矛盾檢測與時間推理。外掛佔用獨佔的記憶插槽:每次 agent 執行前注入相關記憶,結束後自動抽取事實、決策與待辦,共註冊 6 個工具。首次使用時在 ~/.neuralmemory/ 自動建立資料與配置;抽取與自動儲存支援中文、越南語與英文,日文與韓文會明確返回不支援。 自動擷取可能讓私人對話內容跨工作階段保留;先檢查設定,必要時按不同情境拆分記憶區。來源中的 pip 與 npm 安裝路徑不一致,安裝前應確認目標套件。啟用外部 embedding 服務會改變預設的本機資料流。

資料釋出於 2026-10-27

可以用來做什麼

  • 讓 agent 在多次執行之間保留事實與待辦
  • 把重複出現的話題在神經圖裡相互強化

什麼時候用

  • 需要跨會話記憶且不想依賴 LLM 抽取服務時
  • 希望記憶按關聯而非關鍵詞命中時

示例提示詞

以下是根據技能文件生成的示例,不代表實際執行結果。

把這次討論裡的決策和待辦記進記憶
回想一下之前關於這個專案的結論

月度資料趨勢

累計下載3萬
73341.9萬3.2萬2026-03 · 累計下載 90052026-05 · 累計下載 1.3萬2026-06 · 累計下載 1.6萬2026-07 · 累計下載 1.8萬2026-08 · 累計下載 1.9萬2026-09 · 累計下載 2萬2026-10 · 累計下載 3萬26-0326-0926-10
星標13
911132026-03 · 星標 92026-05 · 星標 112026-06 · 星標 132026-07 · 星標 132026-08 · 星標 132026-09 · 星標 132026-10 · 星標 1326-0326-0926-10

2026-09 起統計方法調整:同名技能按作者分別統計,並覆蓋全部技能。圖中虛線為調整前的月份,與之後的資料不直接比較,也不計算跨越調整的增長。

月份累計下載環比星標
2026-103萬—13
2026-092萬—13
2026-081.9萬+3.1%13
2026-071.8萬+10.9%13
2026-061.6萬+23.0%13
2026-051.3萬—11
2026-039005—9

SKILL.md 原文

以下是作者釋出的原始檔案,按純文字顯示,本站未做修改或稽核;其中的連結不可點選,命令請核實後再執行。抓取於 2026-10-23。

9.6 KB · ClawHub v4.63.0
---
name: neural-memory
description: |
  Associative memory with spreading activation for persistent, intelligent recall.
  Use PROACTIVELY when:
  (1) You need to remember facts, decisions, errors, or context across sessions
  (2) User asks "do you remember..." or references past conversations
  (3) Starting a new task — inject relevant context from memory
  (4) After making decisions or encountering errors — store for future reference
  (5) User asks "why did X happen?" — trace causal chains through memory
展開全部(共 259 行)
  Zero LLM dependency. Neural graph with Hebbian learning, memory decay, contradiction detection, and temporal reasoning.
homepage: https://github.com/nhadaututtheky/neural-memory
metadata: {"openclaw":{"emoji":"brain","primaryEnv":"NEURALMEMORY_BRAIN","requires":{"bins":["python3"],"env":["NEURALMEMORY_BRAIN"]},"os":["darwin","linux","win32"],"install":[{"id":"pip","kind":"node","package":"neural-memory","bins":["nmem"],"label":"pip install neural-memory"}]}}
---

# NeuralMemory — Associative Memory for AI Agents

A biologically-inspired memory system that uses spreading activation instead of keyword/vector search. Memories form a neural graph where neurons connect via 20 typed synapses. Frequently co-accessed memories strengthen their connections (Hebbian learning). Stale memories decay naturally. Contradictions are auto-detected.

**Why not just vector search?** Vector search finds documents similar to your query. NeuralMemory finds *conceptually related* memories through graph traversal — even when there's no keyword or embedding overlap. "What decision did we make about auth?" activates time + entity + concept neurons simultaneously and finds the intersection.

## Setup

### 1. Install NeuralMemory

```bash
pip install neural-memory
```

The brain and config at `~/.neuralmemory/` are auto-created on first use.

### 2. Install the OpenClaw Plugin (Recommended)

The plugin occupies the exclusive **memory slot** — auto-injects context before each agent run and auto-captures memories after.

```bash
# Install from npm
npm install -g neuralmemory
```

Add to `~/.openclaw/openclaw.json`:

```json
{
  "plugins": {
    "load": {
      "paths": ["<path-to-installed-plugin>"]
    },
    "entries": {
      "neuralmemory": {
        "enabled": true,
        "config": {
          "pythonPath": "python",
          "brain": "default",
          "autoContext": true,
          "autoCapture": true
        }
      }
    },
    "slots": {
      "memory": "neuralmemory"
    }
  }
}
```

**Plugin features:**
- 6 tools registered automatically (nmem_remember, nmem_recall, nmem_context, nmem_todo, nmem_stats, nmem_health)
- `before_agent_start` hook: injects tool instructions + relevant memories as context (persists across `/new`)
- `agent_end` hook: auto-extracts facts, decisions, and TODOs from the conversation
- Configurable: `contextDepth` (0-3), `maxContextTokens` (100-10000)

**After installing, build the plugin:**

```bash
cd <path-to-installed-plugin>
npm run build
```

This compiles TypeScript to JavaScript in `dist/`. The plugin entry point is `dist/index.js`.

#### Windows Installation

On Windows, use forward slashes or escaped backslashes in `openclaw.json` paths:

```json
{
  "plugins": {
    "load": {
      "paths": ["C:/Users/<you>/AppData/Roaming/npm/node_modules/neuralmemory"]
    }
  }
}
```

To find the installed path:

```powershell
npm list -g neuralmemory --parseable
```

If `openclaw plugins list` doesn't show the plugin:
1. Verify the path in `openclaw.json` points to the package root (where `package.json` is)
2. Ensure `npm run build` was run (the `dist/` folder must exist with compiled `.js` files)
3. Use `python` instead of `python3` in the plugin config (Windows default)

### Alternative: MCP Configuration (Manual)

If you prefer MCP over the plugin, add to `~/.openclaw/mcp.json`:

```json
{
  "mcpServers": {
    "neural-memory": {
      "command": "python",
      "args": ["-m", "neural_memory.mcp"],
      "env": {
        "NEURALMEMORY_BRAIN": "default"
      }
    }
  }
}
```

On Windows, use `"python"` (not `"python3"`). This gives you all 63 MCP tools but without the auto-context/auto-capture hooks.

### 3. Verify

```bash
nmem stats
```

You should see brain statistics (neurons, synapses, fibers).

### Troubleshooting

| Symptom | Cause | Fix |
|---------|-------|-----|
| `openclaw plugins list` doesn't show plugin | Plugin path wrong or not built | Run `npm run build`, verify path in `openclaw.json` |
| Agent runs `nmem remember` in terminal | Agent confused CLI vs tool | Plugin now auto-injects tool instructions via `systemPrompt` |
| Agent forgets tools after `/new` | No tool instructions in new session | Plugin now injects `systemPrompt` on every `before_agent_start` |
| `python3 not found` (Windows) | Windows uses `python` not `python3` | Set `pythonPath: "python"` in plugin config |
| Timeout errors | Slow machine or large brain | Increase `timeout` in plugin config (max 120000ms) |

## Tools Reference

### Core Memory Tools

| Tool | Purpose | When to Use |
|------|---------|-------------|
| `nmem_remember` | Store a memory | After decisions, errors, facts, insights, user preferences |
| `nmem_recall` | Query memories | Before tasks, when user references past context, "do you remember..." |
| `nmem_context` | Get recent memories | At session start, inject fresh context |
| `nmem_todo` | Quick TODO with 30-day expiry | Task tracking |

### Intelligence Tools

| Tool | Purpose | When to Use |
|------|---------|-------------|
| `nmem_auto` | Auto-extract memories from text | After important conversations — captures decisions, errors, TODOs automatically |
| `nmem_recall` (depth=3) | Deep associative recall | Complex questions requiring cross-domain connections |
| `nmem_habits` | Workflow pattern suggestions | When user repeats similar action sequences |

### Management Tools

| Tool | Purpose | When to Use |
|------|---------|-------------|
| `nmem_health` | Brain health diagnostics | Periodic checkup, before sharing brain |
| `nmem_stats` | Brain statistics | Quick overview of memory counts |
| `nmem_version` | Brain snapshots and rollback | Before risky operations, version checkpoints |
| `nmem_transplant` | Transfer memories between brains | Cross-project knowledge sharing |

## Workflow

### At Session Start
1. Call `nmem_context` to inject recent memories into your awareness
2. If user mentions a specific topic, call `nmem_recall` with that topic

### During Conversation
3. When a decision is made: `nmem_remember` with type="decision"
4. When an error occurs: `nmem_remember` with type="error"
5. When user states a preference: `nmem_remember` with type="preference"
6. When asked about past events: `nmem_recall` with appropriate depth

### At Session End
7. Call `nmem_auto` with action="process" on important conversation segments
8. This auto-extracts facts, decisions, errors, and TODOs

## Examples

### Remember a decision
```
nmem_remember(
  content="Use PostgreSQL for production, SQLite for development",
  type="decision",
  tags=["database", "infrastructure"],
  priority=8
)
```

### Recall with spreading activation
```
nmem_recall(
  query="database configuration for production",
  depth=1,
  max_tokens=500
)
```
Returns memories found via graph traversal, not keyword matching. Related memories (e.g., "deploy uses Docker with pg_dump backups") surface even without shared keywords.

### Trace causal chains
```
nmem_recall(
  query="why did the deployment fail last week?",
  depth=2
)
```
Follows CAUSED_BY and LEADS_TO synapses to trace cause-and-effect chains.

### Auto-capture from conversation
```
nmem_auto(
  action="process",
  text="We decided to switch from REST to GraphQL because the frontend needs flexible queries. The migration will take 2 sprints. TODO: update API docs."
)
```
Automatically extracts: 1 decision, 1 fact, 1 TODO.

## Key Features

- **Zero LLM dependency** — Pure algorithmic: regex, graph traversal, Hebbian learning
- **Spreading activation** — Associative recall through neural graph, not keyword/vector search
- **20 synapse types** — Temporal (BEFORE/AFTER), causal (CAUSED_BY/LEADS_TO), semantic (IS_A/HAS_PROPERTY), emotional (FELT/EVOKES), conflict (CONTRADICTS)
- **Memory lifecycle** — Short-term → Working → Episodic → Semantic with Ebbinghaus decay
- **Contradiction detection** — Auto-detects conflicting memories, deprioritizes outdated ones
- **Hebbian learning** — "Neurons that fire together wire together" — memory improves with use
- **Temporal reasoning** — Causal chain traversal, event sequences, temporal range queries
- **Brain versioning** — Snapshot, rollback, diff brain state
- **Brain transplant** — Transfer filtered knowledge between brains
- **Chinese + Vietnamese + English** — Extraction and auto-save triggers across all three (sentiment lexicons remain English + Vietnamese); Japanese and Korean return an explicit unsupported-language hint instead of a silent empty result

## Depth Levels

| Depth | Name | Speed | Use Case |
|-------|------|-------|----------|
| 0 | Instant | <10ms | Quick facts, recent context |
| 1 | Context | ~50ms | Standard recall (default) |
| 2 | Habit | ~200ms | Pattern matching, workflow suggestions |
| 3 | Deep | ~500ms | Cross-domain associations, causal chains |

## Notes

- Memories are stored locally in SQLite at `~/.neuralmemory/brains/<brain>.db`
- No data is sent to external services (unless optional embedding provider is configured)
- Brain isolation: each brain is independent, no cross-contamination
- `nmem_remember` returns fiber_id for reference tracking
- Priority scale: 0 (trivial) to 10 (critical), default 5
- Memory types: fact, decision, preference, todo, insight, context, instruction, error, workflow, reference

「AI 智慧體」分類下的熱門技能

資料月份:2026-10 · 下載/星標數聚合自公開技能註冊庫(ClawHub、SkillHub),按月更新。口徑說明見資料方法。