Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Memphora search my memories for my current project's technical requirements"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
This MCP server connects your AI assistant to Memphora, giving it the ability to:
Remember information across conversations
Search your personal knowledge base
Extract insights from conversations automatically
Recall your preferences, facts, and context
Related MCP server: AGI MCP Server
# Using pip
pip install memphora-mcp
# Or using uvx (recommended for Claude Desktop)
uvx memphora-mcpGo to memphora.ai/dashboard
Create an account or sign in
Copy your API key from the dashboard
Add to your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"memphora": {
"command": "uvx",
"args": ["memphora-mcp"],
"env": {
"MEMPHORA_API_KEY": "your_api_key_here",
"MEMPHORA_USER_ID": "your_unique_user_id"
}
}
}
}Close and reopen Claude Desktop. You should see the Memphora tools available!
Just tell Claude something about yourself:
You: "I work at Google as a software engineer"
Claude: [stores memory] "Got it! I'll remember that you work at Google as a software engineer."
You: "My favorite programming language is Python"
Claude: [stores memory] "Noted! I'll remember that Python is your favorite programming language."Ask Claude about things you've told it before:
You: "Where do I work?"
Claude: [searches memories] "You work at Google as a software engineer."
You: "What programming languages do I like?"
Claude: [searches memories] "Your favorite programming language is Python."Claude will automatically search your memories when relevant:
You: "Can you help me with some code?"
Claude: [searches memories for context]
"Sure! Since you prefer Python and work at Google, I'll write this in Python
following Google's style guide..."Tool | Description |
| Search memories for relevant information |
| Store new information for future recall |
| Extract memories from a conversation |
| List all stored memories |
| Delete a specific memory |
Environment Variable | Description | Default |
| Your Memphora API key | Required |
| Unique identifier for your memories |
|
Add to your Cursor settings:
{
"mcp": {
"servers": {
"memphora": {
"command": "uvx",
"args": ["memphora-mcp"],
"env": {
"MEMPHORA_API_KEY": "your_api_key_here"
}
}
}
}
}Add to your Windsurf MCP configuration:
{
"mcpServers": {
"memphora": {
"command": "python",
"args": ["-m", "memphora_mcp"],
"env": {
"MEMPHORA_API_KEY": "your_api_key_here"
}
}
}
}# Clone the repo
git clone https://github.com/Memphora/memphora-mcp.git
cd memphora-mcp
# Install dependencies
pip install -e ".[dev]"
# Set your API key
export MEMPHORA_API_KEY="your_key"
# Run the server
python -m memphora_mcppytest tests/Your memories are stored securely in Memphora's cloud
Each user has isolated memory storage
API keys are stored locally on your machine
All communication is encrypted via HTTPS
Documentation: memphora.ai/docs
Issues: GitHub Issues
Email: support@memphora.ai
MIT License - see LICENSE for details.
Delete a specific memory by its ID. Use this when the user wants to forget something or correct incorrect information.
| Name | Required | Description | Default |
|---|---|---|---|
| memory_id | Yes | The ID of the memory to delete |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. While it correctly indicates this is a destructive operation ('Delete'), it lacks important details such as whether deletion is permanent, what permissions are required, if there are confirmation prompts, or what happens to associated data. For a mutation tool with zero annotation coverage, this is insufficient.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two concise sentences with zero waste. The first sentence states the core functionality, and the second provides usage context, making it efficiently front-loaded and appropriately sized for this tool.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a destructive mutation tool with no annotations and no output schema, the description is incomplete. It lacks critical behavioral details (e.g., permanence, error handling) and does not describe what the tool returns upon success or failure, leaving significant gaps for an AI agent to understand the tool fully.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, with the single parameter 'memory_id' fully documented in the schema. The description adds no additional parameter information beyond what the schema provides, so it meets the baseline score of 3 where the schema does the heavy lifting.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Delete') and the resource ('a specific memory by its ID'), making the purpose explicit. It distinguishes this tool from its siblings (list, search, store, extract) by focusing on deletion rather than retrieval or creation operations.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context for when to use this tool ('when the user wants to forget something or correct incorrect information'), which helps differentiate it from read-only siblings. However, it does not explicitly state when NOT to use it or name specific alternatives for related tasks.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Extract and store memories from a conversation. Use this to save important information from a longer discussion. The system will automatically identify and store relevant facts.
| Name | Required | Description | Default |
|---|---|---|---|
| conversation | Yes | List of messages in the conversation |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It mentions that 'the system will automatically identify and store relevant facts', which adds some context about automation, but doesn't cover critical aspects like whether this is a read/write operation (implied write from 'store'), permission requirements, rate limits, or what happens if extraction fails. For a tool that appears to perform memory storage with zero annotation coverage, this is inadequate.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately concise with two sentences that directly address purpose and automation. It's front-loaded with the core function, though the second sentence about automatic identification could be integrated more tightly. There's minimal waste, but slight room for improvement in flow.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no annotations, no output schema, and a tool that performs memory extraction and storage (implied mutation), the description is incomplete. It lacks details on what 'memories' entail, how they're stored, error handling, or return values. For a tool with potential side effects and no structured safety hints, this leaves significant gaps for an AI agent.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, with the single parameter 'conversation' fully documented in the schema as 'List of messages in the conversation'. The description doesn't add any meaningful parameter semantics beyond what the schema provides, such as format examples or constraints on conversation length. Baseline 3 is appropriate when the schema does the heavy lifting.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'extract and store' and the resource 'memories from a conversation', with the specific purpose of 'saving important information from a longer discussion'. However, it doesn't explicitly differentiate from sibling tools like memphora_store, which might also store memories but potentially from different sources.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage context ('from a longer discussion') and suggests when to use it ('to save important information'), but doesn't provide explicit guidance on when to use this vs. alternatives like memphora_store or memphora_search, nor does it mention any exclusions or prerequisites for usage.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List all stored memories for the user. Use this to see what information has been remembered.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Maximum number of memories to return (default: 20) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It mentions listing memories but doesn't describe key behaviors such as pagination (implied by the 'limit' parameter), sorting order, whether it returns all fields or a summary, or any rate limits or authentication requirements. This leaves significant gaps for an agent to understand how to use it effectively.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized with two concise sentences that directly state the purpose and usage. It's front-loaded with the core functionality ('List all stored memories for the user'), and the second sentence adds practical guidance. There's no wasted text, making it efficient for an agent to parse.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity of a list operation with no annotations and no output schema, the description is incomplete. It doesn't explain what the return values look like (e.g., format, fields), behavioral aspects like pagination or sorting, or how it interacts with siblings. For a tool that lists user data, more context is needed to ensure the agent can use it correctly without relying on trial and error.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 100% description coverage, with the 'limit' parameter well-documented in the schema itself (including type, description, and default). The description doesn't add any meaning beyond what the schema provides, as it doesn't mention parameters at all. With high schema coverage, the baseline score of 3 is appropriate, as the schema does the heavy lifting.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose with a specific verb ('List') and resource ('all stored memories for the user'), making it easy to understand what the tool does. It distinguishes from siblings by focusing on listing rather than deleting, extracting, searching, or storing memories. However, it doesn't explicitly differentiate from 'memphora_search' which might also list memories but with filtering.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage by stating 'Use this to see what information has been remembered,' which suggests it's for viewing stored memories. However, it doesn't provide explicit guidance on when to use this tool versus alternatives like 'memphora_search' (which might list with filters) or 'memphora_extract_conversation' (which might list conversation-specific memories). No exclusions or prerequisites are mentioned.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search your personal memories for relevant information. Use this when the user asks about something they may have mentioned before, their preferences, past experiences, or any personal information. Examples: 'What's my favorite food?', 'Where do I work?', 'What projects am I working on?'
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | What to search for in memories | |
| limit | No | Maximum number of results (default: 5) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It effectively communicates this is a read/search operation (not destructive) and implies it returns relevant information from personal memories. However, it doesn't disclose important behavioral aspects like authentication requirements, rate limits, or how results are ranked/returned.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is efficiently structured with two sentences: the first states the purpose and usage context, the second provides concrete examples. Every element serves a clear purpose with zero wasted words, making it easy to parse and understand quickly.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a search tool with no annotations and no output schema, the description does well by clearly explaining purpose, usage context, and providing examples. However, it lacks information about what the search returns (format, structure) and doesn't mention authentication or access considerations, which would be helpful given the personal nature of the data.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
With 100% schema description coverage, the schema already documents both parameters thoroughly. The description doesn't add any parameter-specific information beyond what's in the schema descriptions. It mentions searching 'for relevant information' which aligns with the query parameter but provides no additional syntax or format details.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose with specific verb ('Search') and resource ('your personal memories'), and distinguishes it from siblings by focusing on retrieval of personal information rather than deletion, extraction, listing, or storage. It provides concrete examples that illustrate the scope.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly states when to use this tool ('when the user asks about something they may have mentioned before, their preferences, past experiences, or any personal information') and provides three specific example queries. This gives clear context for selection over alternatives like memphora_list_memories or memphora_store.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Store important information about the user for future recall. Use this when the user shares personal details, preferences, facts about themselves, or explicitly asks you to remember something. Examples: 'I work at Google', 'My favorite color is blue', 'Remember that I'm allergic to peanuts'
| Name | Required | Description | Default |
|---|---|---|---|
| content | Yes | The information to remember (should be a complete, self-contained fact) | |
| category | No | Optional category (e.g., 'preference', 'work', 'health', 'relationship') |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It describes the tool's purpose and usage context but lacks details on behavioral traits such as storage limitations, persistence duration, privacy implications, or error handling. The description doesn't contradict any annotations, but it misses key operational details for a storage tool.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is front-loaded with the core purpose, followed by usage guidelines and examples. Every sentence adds value without redundancy, making it efficient and well-structured for quick understanding.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's moderate complexity (storage operation with 2 parameters) and no output schema, the description covers purpose and usage well but lacks details on return values or behavioral aspects like success confirmation. It's mostly complete but could benefit from information on what happens after storage.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 100% description coverage, providing clear documentation for both parameters. The description adds value by emphasizing that the content should be 'a complete, self-contained fact' and giving examples that illustrate the semantics, though it doesn't explicitly detail parameter usage beyond what the schema covers.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description explicitly states the verb ('Store') and resource ('important information about the user for future recall'), making the purpose specific and clear. It distinguishes this tool from its siblings by focusing on storing user information, unlike deletion, extraction, listing, or searching operations.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides explicit guidance on when to use this tool ('when the user shares personal details, preferences, facts about themselves, or explicitly asks you to remember something') and includes concrete examples. This clearly differentiates it from alternative tools like memphora_search or memphora_list_memories.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
memphora_deletememphora_extract_conversationmemphora_list_memoriesmemphora_searchmemphora_storeEach tool has a clearly distinct purpose with no overlap: delete removes memories, extract processes conversations, list shows all memories, search finds specific memories, and store adds new memories. The descriptions reinforce these distinct roles, making misselection unlikely.
All tools follow a consistent 'memphora_verb_noun' pattern (e.g., memphora_delete, memphora_extract_conversation). This uniform naming convention makes the tool set predictable and easy to understand at a glance.
With 5 tools, this server is well-scoped for a memory management system. Each tool earns its place by covering core operations (store, search, list, delete, and conversation extraction), avoiding bloat while providing complete functionality.
The tool set offers complete CRUD/lifecycle coverage for the memory domain: store (create), search and list (read), delete (delete), and extract_conversation (a specialized create/update). There are no obvious gaps, enabling agents to handle all expected workflows without dead ends.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Gives your AI assistant persistent memory and intelligence about your work patterns.
Persistent personal memory for AI assistants — save, search, and recall across every MCP client.
Persistent memory and knowledge management for AI agents with semantic search and 50+ tools.
Long-term memory for AI assistants. Isolated per-user storage, recall across conversations.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Memphora/memphora-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
