Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion src/config/navigation.yml
Original file line number Diff line number Diff line change
Expand Up @@ -235,6 +235,7 @@ sidebar:
items:
- docs/community/tools/strands-deepgram
- docs/community/tools/strands-hubspot
- docs/community/tools/strands-memmachine
- docs/community/tools/strands-teams
- docs/community/tools/strands-telegram
- docs/community/tools/strands-telegram-listener
Expand Down Expand Up @@ -274,4 +275,4 @@ github:
- label: strands-agents
href: https://github.com/strands-agents
- label: strands-labs
href: https://github.com/strands-labs
href: https://github.com/strands-labs
122 changes: 122 additions & 0 deletions src/content/docs/community/tools/strands-memmachine.mdx
Original file line number Diff line number Diff line change
@@ -0,0 +1,122 @@
---
title: strands-memmachine
community: true
description: Persistent episodic and semantic memory for Strands agents via the MemMachine Platform.
integrationType: tool
languages: Python
sidebar:
label: "memmachine"
project:
pypi: https://pypi.org/project/strands-memmachine/
github: https://github.com/MemMachine/MemMachine/tree/main/integrations/strands-memmachine
maintainer: MemMachine
service:
name: MemMachine
link: https://memmachine.ai
---

[strands-memmachine](https://github.com/MemMachine/MemMachine/tree/main/integrations/strands-memmachine) is a persistent memory tool for Strands agents powered by the [MemMachine Platform](https://memmachine.ai) by MemVerge. It supports episodic and semantic memory with full project management.

## Installation

```bash
pip install strands-memmachine
```

## Usage

```python
from strands import Agent
from strands_memmachine import memmachine_memory

agent = Agent(tools=[memmachine_memory])

# Store a memory
agent.tool.memmachine_memory(
action="store",
content="User prefers aisle seats on flights",
metadata={"user_id": "alice", "category": "travel"},
)

# Search memories
agent.tool.memmachine_memory(
action="search",
query="What are the flight preferences?",
top_k=5,
)

# List memories with filter
agent.tool.memmachine_memory(
action="list",
filter="metadata.user_id=alice",
page_size=20,
)

# Delete a memory
agent.tool.memmachine_memory(
action="delete",
memory_type="episodic",
memory_id="mem-123",
)

# Bulk delete memories
agent.tool.memmachine_memory(
action="delete",
memory_type="semantic",
memory_ids=["sem-1", "sem-2", "sem-3"],
)

# Create a project namespace
agent.tool.memmachine_memory(
action="create_project",
project_id="my-project",
description="Travel preferences project",
)

# Get a project
agent.tool.memmachine_memory(
action="get_project",
project_id="my-project",
)

# List all projects
agent.tool.memmachine_memory(
action="list_projects",
)

# Get episode count for a project
agent.tool.memmachine_memory(
action="get_episode_count",
project_id="my-project",
)

# Delete a project
agent.tool.memmachine_memory(
action="delete_project",
project_id="my-project",
)
```

## Key Features

- **Episodic Memory**: Conversational, event-based memories tied to interactions
- **Semantic Memory**: Factual, structured knowledge extracted from conversations
- **Project Management**: Create, retrieve, list, and delete isolated memory namespaces
- **Metadata Filtering**: Filter memories using key-value expressions
- **Bulk Operations**: Single and bulk memory deletion support
- **Pagination**: Control result sets with page size and page number

## Configuration

```bash
MEMMACHINE_API_KEY=your_api_key # Required
```

Get your API key at: [console.memmachine.ai](https://console.memmachine.ai)

## Resources

- [PyPI Package](https://pypi.org/project/strands-memmachine/)
- [GitHub Repository](https://github.com/MemMachine/MemMachine/tree/main/integrations/strands-memmachine)
- [MemMachine Platform](https://memmachine.ai)
- [MemMachine API Docs](https://api.memmachine.ai/docs)