my-project-public

repository

loading code, commits, and activity

repositories

loading repo index

my-project-public — gitlawb

#1	---
#2	title: MCP Integration
#3	description: "Connect any AI client to Mem0 using Model Context Protocol for universal memory access"
#4	---
#5
#6	> Model Context Protocol (MCP) provides a standardized way for AI agents to manage their own memory through Mem0, without manual API calls.
#7
#8	## Why use MCP
#9
#10	When building AI applications, memory management often requires manual integration. MCP eliminates this complexity by:
#11
#12	- Universal compatibility: Works with any MCP-compatible client (Claude Desktop, Cursor, custom agents)
#13	- Agent autonomy: AI agents decide when to save, search, or update memories
#14	- Zero infrastructure: No servers to maintain - Mem0 handles everything
#15	- Standardized protocol: One integration works across all your AI tools
#16
#17	## Available tools
#18
#19	The MCP server exposes 9 memory tools to your AI client:
#20
#21	\| Tool \| Purpose \|
#22	\|------\|---------\|
#23	\| `add_memory` \| Store conversations or facts \|
#24	\| `search_memories` \| Find relevant memories with filters \|
#25	\| `get_memories` \| List memories with pagination \|
#26	\| `update_memory` \| Modify existing memory content \|
#27	\| `delete_memory` \| Remove specific memories \|
#28	\| `delete_all_memories` \| Bulk delete memories \|
#29	\| `delete_entities` \| Remove user/agent/app entities \|
#30	\| `get_memory` \| Retrieve single memory by ID \|
#31	\| `list_entities` \| View stored entities \|
#32
#33	## Deployment options
#34
#35	Choose the deployment method that fits your workflow:
#36
#37	<AccordionGroup>
#38	<Accordion title="Python package (recommended)">
#39	Install and run locally with uvx:
#40
#41	```bash
#42	uv pip install mem0-mcp-server
#43	```
#44
#45	Configure your client:
#46	```json
#47	{
#48	"mcpServers": {
#49	"mem0": {
#50	"command": "uvx",
#51	"args": ["mem0-mcp-server"],
#52	"env": {
#53	"MEM0_API_KEY": "m0-...",
#54	"MEM0_DEFAULT_USER_ID": "your-handle"
#55	}
#56	}
#57	}
#58	}
#59	```
#60	</Accordion>
#61
#62	<Accordion title="Docker container">
#63	Containerized deployment with HTTP endpoint:
#64
#65	```bash
#66	docker build -t mem0-mcp-server https://github.com/mem0ai/mem0-mcp.git
#67	docker run --rm -d -e MEM0_API_KEY="m0-..." -p 8080:8081 mem0-mcp-server
#68	```
#69
#70	Configure for HTTP:
#71	```json
#72	{
#73	"mcpServers": {
#74	"mem0-docker": {
#75	"command": "curl",
#76	"args": ["-X", "POST", "http://localhost:8080/mcp", "--data-binary", "@"],
#77	"env": {
#78	"MEM0_API_KEY": "m0-..."
#79	}
#80	}
#81	}
#82	}
#83	```
#84	</Accordion>
#85
#86	<Accordion title="Smithery">
#87	One-click setup with managed service:
#88
#89	Visit [smithery.ai/server/@mem0ai/mem0-memory-mcp](https://smithery.ai/server/@mem0ai/mem0-memory-mcp) and:
#90
#91	1. Select your AI client (Cursor, Claude Desktop, etc.)
#92	2. Configure your Mem0 API key
#93	3. Set your default user ID
#94	4. Enable graph memory (optional)
#95	5. Copy the generated configuration
#96
#97	Your client connects automatically - no installation required.
#98	</Accordion>
#99	</AccordionGroup>
#100
#101	## Configuration
#102
#103	### Required environment variables
#104	```bash
#105	MEM0_API_KEY="m0-..." # Your Mem0 API key
#106	MEM0_DEFAULT_USER_ID="your-handle" # Default user ID
#107	```
#108
#109	### Optional variables
#110	```bash
#111	MEM0_ENABLE_GRAPH_DEFAULT="true" # Enable graph memories
#112	MEM0_MCP_AGENT_MODEL="gpt-4o-mini" # LLM for bundled examples
#113	```
#114
#115	<AccordionGroup>
#116	<Accordion title="Test your setup with the Python agent">
#117	The included Pydantic AI agent provides an interactive REPL to test memory operations:
#118
#119	```bash
#120	# Install the package
#121	pip install mem0-mcp-server
#122
#123	# Set your API keys
#124	export MEM0_API_KEY="m0-..."
#125	export OPENAI_API_KEY="sk-openai-..."
#126
#127	# Clone and test with the agent
#128	git clone https://github.com/mem0ai/mem0-mcp.git
#129	cd mem0-mcp-server
#130	python example/pydantic_ai_repl.py
#131	```
#132
#133	Testing different server configurations:
#134
#135	- Local server (default): `python example/pydantic_ai_repl.py`
#136
#137	- Docker container:
#138	```bash
#139	export MEM0_MCP_CONFIG_PATH=example/docker-config.json
#140	export MEM0_MCP_CONFIG_SERVER=mem0-docker
#141	python example/pydantic_ai_repl.py
#142	```
#143
#144	- Smithery remote:
#145	```bash
#146	export MEM0_MCP_CONFIG_PATH=example/config-smithery.json
#147	export MEM0_MCP_CONFIG_SERVER=mem0-memory-mcp
#148	python example/pydantic_ai_repl.py
#149	```
#150
#151	Try these test prompts:
#152	- "Remember that I love tiramisu"
#153	- "Search for my food preferences"
#154	- "Update my project: the mobile app is now 80% complete"
#155	- "Show me all memories about project Phoenix"
#156	- "Delete memories from 2023"
#157	</Accordion>
#158	</AccordionGroup>
#159
#160	## How the testing works
#161
#162	1. Configuration loads - Reads from `example/config.json` by default
#163	2. Server starts - Launches or connects to the Mem0 MCP server
#164	3. Agent connects - Pydantic AI agent (Mem0Guide) attaches to the server
#165	4. Interactive REPL - You get a chat interface to test all memory operations
#166
#167	## Example interactions
#168
#169	Once connected, your AI agent can:
#170
#171	```
#172	User: Remember that I'm allergic to peanuts
#173	Agent: [calls add_memory] Got it! I've saved your peanut allergy.
#174
#175	User: What dietary restrictions do I know about?
#176	Agent: [calls search_memories] You have a peanut allergy.
#177	```
#178
#179	The agent automatically decides when to use memory tools based on context.
#180
#181	## Try these prompts
#182
#183	```python
#184	# Multi-task operations
#185	"Generate 5 user personas for our e-commerce app with different demographics, store them all, then search for existing personas"
#186
#187	# Natural context retrieval
#188	"Anything about my work preferences I should remember?"
#189
#190	# Complex information updates
#191	"Update my current project: the mobile app is now 80% complete, we've fixed the login issues, and the launch date is March 15"
#192
#193	# Time-based queries
#194	"What meetings did I have last week about Project Phoenix?"
#195
#196	# Memory cleanup
#197	"Delete all test data and temporary memories from our development phase"
#198
#199	# Personal preferences
#200	"I drink oat milk cappuccino with one sugar every morning, and I prefer standing desks"
#201
#202	# Health and wellness tracking
#203	"I'm allergic to peanuts and shellfish, and I go for 5km runs on weekday mornings"
#204	```
#205
#206	These examples demonstrate how MCP enables natural language memory operations - the AI agent automatically determines when to add, search, update, or delete memories based on context.
#207
#208	## What you can do
#209
#210	The Mem0 MCP server enables powerful memory capabilities for your AI applications:
#211
#212	- Health tracking: "I'm allergic to peanuts and shellfish" - Add new health information
#213	- Research data: "Store these trial parameters: 200 participants, double-blind, placebo-controlled" - Save structured data
#214	- Preference queries: "What do you know about my dietary preferences?" - Search and retrieve relevant memories
#215	- Project updates: "Update my project status: the mobile app is now 80% complete" - Modify existing memory
#216	- Data cleanup: "Delete all memories from 2023" - Bulk remove outdated information
#217	- Topic overview: "Show me everything about Project Phoenix" - List all memories for a subject
#218
#219	## Performance tips
#220
#221	- Enable graph memories for relationship-aware recall
#222	- Use specific filters when searching large memory sets
#223	- Batch operations when adding multiple memories
#224	- Monitor memory usage in the Mem0 dashboard
#225
#226	## Best practices
#227
#228	- Start simple: Use the Python package for development
#229	- Use wildcards: `user_id: "*"` to search across all users
#230	- Test locally: Use the bundled Python agent to verify setup
#231	- Monitor usage: Track memory operations in the dashboard
#232	- Document patterns: Share successful prompt patterns with your team
#233
#234	{/* DEBUG: verify CTA targets */}
#235
#236	<CardGroup cols={2}>
#237	<Card
#238	title="Memory Filters"
#239	description="Refine memory retrieval with powerful filtering capabilities"
#240	icon="scale-balanced"
#241	href="/platform/features/v2-memory-filters"
#242	/>
#243	<Card
#244	title="Gemini 3 with MCP"
#245	description="See MCP in action with Google's Gemini 3 model"
#246	icon="book-open"
#247	href="/cookbooks/frameworks/gemini-3-with-mem0-mcp"
#248	/>
#249	</CardGroup>

z6Mkq5mY3JWtxoxUobWcfNHm7AkRubgSWEZTkBVqZXJviFZ5/my-project-public