my-project-public

repository

loading code, commits, and activity

repositories

loading repo index

#1	---
#2	title: OpenClaw
#3	---
#4
#5	Add long-term memory to [OpenClaw](https://github.com/openclaw/openclaw) agents with the `@mem0/openclaw-mem0` plugin. Your agent forgets everything between sessions — this plugin fixes that by automatically watching conversations, extracting what matters, and bringing it back when relevant.
#6
#7	## Overview
#8
#9	<Frame>
#10	<img src="/images/openclaw-architecture.png" alt="OpenClaw Mem0 Architecture" />
#11	</Frame>
#12
#13	The plugin provides:
#14	1. Auto-Recall — Before the agent responds, memories matching the current message are injected into context
#15	2. Auto-Capture — After the agent responds, the exchange is sent to Mem0 which decides what's worth keeping
#16	3. Agent Tools — Five tools for explicit memory operations during conversations
#17
#18	Both auto-recall and auto-capture run silently with no manual configuration required.
#19
#20	## Installation
#21
#22	```bash
#23	openclaw plugins install @mem0/openclaw-mem0
#24	```
#25
#26	## Setup and Configuration
#27
#28	### Understanding `userId`
#29
#30	The `userId` field is a string you choose to uniquely identify the user whose memories are being stored. It is not something you look up in the Mem0 dashboard — you define it yourself.
#31
#32	Pick any stable, unique identifier for the user. Common choices:
#33
#34	- Your application's internal user ID (e.g. `"user_123"`, `"alice@example.com"`)
#35	- A UUID (e.g. `"550e8400-e29b-41d4-a716-446655440000"`)
#36	- A simple username (e.g. `"alice"`)
#37
#38	All memories are scoped to this `userId` — different values create separate memory namespaces. If you don't set it, it defaults to `"default"`, which means all users share the same memory space.
#39
#40	<Tip>In a multi-user application, set `userId` dynamically per user (e.g. from your auth system) rather than hardcoding a single value.</Tip>
#41
#42	### Platform Mode (Mem0 Cloud)
#43
#44	<Note>Get your API key from [app.mem0.ai](https://app.mem0.ai).</Note>
#45
#46	Add to your `openclaw.json`:
#47
#48	```json5
#49	// plugins.entries
#50	"openclaw-mem0": {
#51	"enabled": true,
#52	"config": {
#53	"apiKey": "${MEM0_API_KEY}",
#54	"userId": "alice" // any unique identifier you choose for this user
#55	}
#56	}
#57	```
#58
#59	### Open-Source Mode (Self-hosted)
#60
#61	No Mem0 key needed. Requires `OPENAI_API_KEY` for default embeddings/LLM.
#62
#63	```json5
#64	"openclaw-mem0": {
#65	"enabled": true,
#66	"config": {
#67	"mode": "open-source",
#68	"userId": "alice" // any unique identifier you choose for this user
#69	}
#70	}
#71	```
#72
#73	Sensible defaults work out of the box. To customize the embedder, vector store, or LLM:
#74
#75	```json5
#76	"config": {
#77	"mode": "open-source",
#78	"userId": "your-user-id",
#79	"oss": {
#80	"embedder": { "provider": "openai", "config": { "model": "text-embedding-3-small" } },
#81	"vectorStore": { "provider": "qdrant", "config": { "host": "localhost", "port": 6333 } },
#82	"llm": { "provider": "openai", "config": { "model": "gpt-4o" } }
#83	}
#84	}
#85	```
#86
#87	All `oss` fields are optional. See [Mem0 OSS docs](/open-source/node-quickstart) for available providers.
#88
#89	## Short-term vs Long-term Memory
#90
#91	Memories are organized into two scopes:
#92
#93	- Session (short-term) — Auto-capture stores memories scoped to the current session via Mem0's `run_id` / `runId` parameter. These are contextual to the ongoing conversation.
#94
#95	- User (long-term) — The agent can explicitly store long-term memories using the `memory_store` tool (with `longTerm: true`, the default). These persist across all sessions for the user.
#96
#97	During auto-recall, the plugin searches both scopes and presents them separately — long-term memories first, then session memories — so the agent has full context.
#98
#99	## Agent Tools
#100
#101	The agent gets five tools it can call during conversations:
#102
#103	\| Tool \| Description \|
#104	\|------\|-------------\|
#105	\| `memory_search` \| Search memories by natural language \|
#106	\| `memory_list` \| List all stored memories for a user \|
#107	\| `memory_store` \| Explicitly save a fact \|
#108	\| `memory_get` \| Retrieve a memory by ID \|
#109	\| `memory_forget` \| Delete by ID or by query \|
#110
#111	The `memory_search` and `memory_list` tools accept a `scope` parameter (`"session"`, `"long-term"`, or `"all"`) to control which memories are queried. The `memory_store` tool accepts a `longTerm` boolean (default: `true`) to choose where to store.
#112
#113	## CLI Commands
#114
#115	```bash
#116	# Search all memories (long-term + session)
#117	openclaw mem0 search "what languages does the user know"
#118
#119	# Search only long-term memories
#120	openclaw mem0 search "what languages does the user know" --scope long-term
#121
#122	# Search only session/short-term memories
#123	openclaw mem0 search "what languages does the user know" --scope session
#124
#125	# View stats
#126	openclaw mem0 stats
#127	```
#128
#129	## Configuration Options
#130
#131	### General Options
#132
#133	\| Key \| Type \| Default \| Description \|
#134	\|-----\|------\|---------\|-------------\|
#135	\| `mode` \| `"platform"` \\| `"open-source"` \| `"platform"` \| Which backend to use \|
#136	\| `userId` \| `string` \| `"default"` \| Scope memories per user \|
#137	\| `autoRecall` \| `boolean` \| `true` \| Inject memories before each turn \|
#138	\| `autoCapture` \| `boolean` \| `true` \| Store facts after each turn \|
#139	\| `topK` \| `number` \| `5` \| Max memories per recall \|
#140	\| `searchThreshold` \| `number` \| `0.3` \| Min similarity (0–1) \|
#141
#142	### Platform Mode Options
#143
#144	\| Key \| Type \| Default \| Description \|
#145	\|-----\|------\|---------\|-------------\|
#146	\| `apiKey` \| `string` \| — \| Required. Mem0 API key (supports `${MEM0_API_KEY}`) \|
#147	\| `orgId` \| `string` \| — \| Organization ID \|
#148	\| `projectId` \| `string` \| — \| Project ID \|
#149	\| `enableGraph` \| `boolean` \| `false` \| Entity graph for relationships \|
#150	\| `customInstructions` \| `string` \| (built-in) \| Extraction rules — what to store, how to format \|
#151	\| `customCategories` \| `object` \| (12 defaults) \| Category name → description map for tagging \|
#152
#153	### Open-Source Mode Options
#154
#155	\| Key \| Type \| Default \| Description \|
#156	\|-----\|------\|---------\|-------------\|
#157	\| `customPrompt` \| `string` \| (built-in) \| Extraction prompt for memory processing \|
#158	\| `oss.embedder.provider` \| `string` \| `"openai"` \| Embedding provider (`"openai"`, `"ollama"`, etc.) \|
#159	\| `oss.embedder.config` \| `object` \| — \| Provider config: `apiKey`, `model`, `baseURL` \|
#160	\| `oss.vectorStore.provider` \| `string` \| `"memory"` \| Vector store (`"memory"`, `"qdrant"`, `"chroma"`, etc.) \|
#161	\| `oss.vectorStore.config` \| `object` \| — \| Provider config: `host`, `port`, `collectionName`, `dimension` \|
#162	\| `oss.llm.provider` \| `string` \| `"openai"` \| LLM provider (`"openai"`, `"anthropic"`, `"ollama"`, etc.) \|
#163	\| `oss.llm.config` \| `object` \| — \| Provider config: `apiKey`, `model`, `baseURL`, `temperature` \|
#164	\| `oss.historyDbPath` \| `string` \| — \| SQLite path for memory edit history \|
#165
#166	Everything inside `oss` is optional — defaults use OpenAI embeddings (`text-embedding-3-small`), in-memory vector store, and OpenAI LLM.
#167
#168	## Key Features
#169
#170	1. Zero Configuration — Auto-recall and auto-capture work out of the box with no prompting required
#171	2. Dual Memory Scopes — Session-scoped short-term and user-scoped long-term memories
#172	3. Flexible Backend — Use Mem0 Cloud for managed service or self-host with open-source mode
#173	4. Rich Tool Suite — Five agent tools for explicit memory operations when needed
#174
#175	## Conclusion
#176
#177	The `@mem0/openclaw-mem0` plugin gives OpenClaw agents persistent memory with minimal setup. Whether using Mem0 Cloud or self-hosting, your agents can now remember user preferences, facts, and context across sessions automatically.
#178
#179	<CardGroup cols={2}>
#180	<Card title="OpenAI Agents SDK" icon="robot" href="/integrations/openai-agents-sdk">
#181	Build agents with OpenAI's SDK and Mem0
#182	</Card>
#183	<Card title="LangGraph Integration" icon="diagram-project" href="/integrations/langgraph">
#184	Create stateful agent workflows with memory
#185	</Card>
#186	</CardGroup>
#187

z6Mkq5mY3JWtxoxUobWcfNHm7AkRubgSWEZTkBVqZXJviFZ5/my-project-public