
# MeshOS
**The Memory & Knowledge Engine for Multi-Agent Systems**
MeshOS is a **developer-first framework** for building **multi-agent AI-driven operations** with structured memory, knowledge retrieval, and real-time collaboration. Unlike generic memory stores, MeshOS is purpose-built for:
- **Autonomous Agents & Teams** – Agents and humans evolve a shared memory over time.
- **Graph-Based Memory** – Track relationships, dependencies, and evolving knowledge.
- **Fast Semantic Search** – Vector-based retrieval with **pgvector**.
- **Event-Driven Execution** – Automate workflows based on evolving context.
- **Versioned Knowledge** – Track updates, past decisions, and historical context.
- **Open & Portable** – Runs on **PostgreSQL + Hasura** with no vendor lock-in.
## Why MeshOS?
Most frameworks give you a **blob of memories**—MeshOS gives you **structured, evolving intelligence** with deep relationships and versioning.
| Feature | MeshOS | Mem0 / Letta / Zep |
| ---------------------------- | ------ | ------------------ |
| **Multi-Agent Memory** | ✅ Yes | ❌ No |
| **Structured Taxonomy** | ✅ Yes | ❌ No |
| **Versioned Knowledge** | ✅ Yes | ❌ No |
| **Graph-Based Relationships** | ✅ Yes | ❌ No |
| **Semantic & Vector Search** | ✅ Yes | ✅ Partial |
| **Event-Driven Execution** | ✅ Yes | ❌ No |
| **Open-Source & Portable** | ✅ Yes | ✅ Partial |
### **Who is MeshOS for?**
✅ **Builders of AI-powered operations** – Structured memory and decision-making for AI-driven systems.
✅ **Multi-agent system developers** – AI agents that need to store, process, and evolve shared knowledge.
✅ **Developers & engineers** – Wanting an **open-source, PostgreSQL-powered framework** with no lock-in.
---
```mermaid
flowchart LR
%% Main System
subgraph MeshOS[MeshOS System]
direction LR
%% Taxonomy Details
subgraph Taxonomy[Memory Classification]
direction TB
subgraph DataTypes[Data Types]
direction LR
knowledge[Knowledge Type]
activity[Activity Type]
decision[Decision Type]
media[Media Type]
end
subgraph Subtypes[Example Subtypes]
direction LR
k_types[Research/Mission/Vision]
a_types[Conversations/Logs/Events]
d_types[Policies/Strategies]
m_types[Documents/Images]
knowledge --> k_types
activity --> a_types
decision --> d_types
media --> m_types
end
subgraph Relations[Edge Types]
direction LR
basic[related_to/version_of]
semantic[influences/depends_on]
temporal[follows_up/precedes]
end
end
%% Memory Operations
subgraph MemoryEngine[Memory Operations]
direction LR
rememberAction[Store/Remember]
recallAction[Search/Recall]
linkAction[Link Memories]
versioning[Version Control]
rememberAction --> recallAction
recallAction --> linkAction
linkAction --> versioning
end
end
%% Organization & Agents
subgraph Organization[Organization & Agents]
direction TB
%% Company Memory
subgraph CompanyMemory[Company-Wide Memory]
direction LR
corpVision[Company Vision]
corpMission[Company Mission]
corpData[Knowledge Base]
end
%% Agents
subgraph Agent1[Research Agent]
a1Mem[Research Memories]
end
subgraph Agent2[Service Agent]
a2Mem[Service Memories]
end
end
%% System Connections
Taxonomy --> MemoryEngine
MemoryEngine --> Organization
%% Memory Connections
corpVision -.->|influences| a1Mem
corpMission -.->|guides| a2Mem
a1Mem -.->|shares| a2Mem
a2Mem -.->|feedback| corpData
a1Mem -.->|versions| corpData
%% Styling
classDef system fill:#dfeff9,stroke:#333,stroke-width:1.5px
classDef engine fill:#fcf8e3,stroke:#333
classDef taxonomy fill:#e7f5e9,stroke:#333
classDef types fill:#f8f4ff,stroke:#333
classDef org fill:#f4f4f4,stroke:#333
class MeshOS system
class MemoryEngine engine
class Taxonomy,DataTypes,Subtypes,Relations taxonomy
class Organization org
```
---
## Getting Started
### Install & Create a New Instance
```bash
pip install mesh-os
mesh-os create my-project && cd my-project
mesh-os up
```
## Usage
```python
from mesh_os import MeshOS
# Initialize MeshOS
os = MeshOS()
# Register an agent
agent = os.register_agent(name="AI_Explorer")
# Store structured knowledge
memory = os.remember(
content="The Moon has water ice.",
agent_id=agent.id,
metadata={
"type": "knowledge",
"subtype": "fact",
"tags": ["astronomy", "moon"],
"version": 1
}
)
# Retrieve similar knowledge
results = os.recall(query="Tell me about the Moon.")
```
---
## 🏗️ Core Features
✅ **Memory for Multi-Agent Systems** – Let agents store, retrieve, and link structured knowledge.
✅ **Fast Semantic Search** – pgvector-powered similarity matching across all memories.
✅ **Graph-Based Knowledge** – Build evolving relationships between facts, ideas, and actions.
✅ **Versioning Built-In** – Track updates, past decisions, and context shifts.
✅ **Event-Driven Execution** – Automate workflows based on new knowledge.
✅ **Open & Portable** – Runs anywhere PostgreSQL does. No black-box infrastructure.
---
## 🔗 Structured Taxonomy & Memory Graph
MeshOS **enforces structured knowledge** with **memory classification** and **versioning**:
| **Memory Type** | **Examples** |
| --------------- | -------------------------------------------- |
| **Knowledge** | Research reports, datasets, concepts |
| **Activity** | Agent workflows, logs, system events |
| **Decision** | Policy updates, business strategy |
| **Media** | Documents, images, AI-generated content |
Memories **evolve** over time, with full versioning and relationship tracking.
---
## 🛠️ Development & Configuration
### **Configuration**
```ini
# Required
OPENAI_API_KEY=your_api_key_here
# Optional (defaults shown)
POSTGRES_PASSWORD=mysecretpassword
HASURA_ADMIN_SECRET=meshos
POSTGRES_PORT=5432
HASURA_PORT=8080
HASURA_ENABLE_CONSOLE=true
```
### **Development**
```bash
git clone https://github.com/yourusername/mesh-os.git
cd mesh-os
poetry install
poetry run pytest
```
### **Contributing**
Contributions are welcome! Please submit a Pull Request.
---
## ⚖️ License
This project is licensed under the Apache 2.0 License – see [LICENSE](./LICENSE) for details.
Raw data
{
"_id": null,
"home_page": null,
"name": "mesh-os",
"maintainer": null,
"docs_url": null,
"requires_python": "<4.0,>=3.9",
"maintainer_email": null,
"keywords": "ai, agents, memory, semantic-search, vector-database",
"author": "Calvin Hoenes",
"author_email": "calvin@props.app",
"download_url": "https://files.pythonhosted.org/packages/b2/1d/06dc224ba868fb77586422420977e65543f3928e9fbb3902c5ce46f0a3d5/mesh_os-0.1.11.tar.gz",
"platform": null,
"description": "\n\n# MeshOS\n\n**The Memory & Knowledge Engine for Multi-Agent Systems**\n\nMeshOS is a **developer-first framework** for building **multi-agent AI-driven operations** with structured memory, knowledge retrieval, and real-time collaboration. Unlike generic memory stores, MeshOS is purpose-built for:\n\n- **Autonomous Agents & Teams** \u2013 Agents and humans evolve a shared memory over time.\n- **Graph-Based Memory** \u2013 Track relationships, dependencies, and evolving knowledge.\n- **Fast Semantic Search** \u2013 Vector-based retrieval with **pgvector**.\n- **Event-Driven Execution** \u2013 Automate workflows based on evolving context.\n- **Versioned Knowledge** \u2013 Track updates, past decisions, and historical context.\n- **Open & Portable** \u2013 Runs on **PostgreSQL + Hasura** with no vendor lock-in.\n\n## Why MeshOS?\n\nMost frameworks give you a **blob of memories**\u2014MeshOS gives you **structured, evolving intelligence** with deep relationships and versioning.\n\n| Feature | MeshOS | Mem0 / Letta / Zep |\n| ---------------------------- | ------ | ------------------ |\n| **Multi-Agent Memory** | \u2705 Yes | \u274c No |\n| **Structured Taxonomy** | \u2705 Yes | \u274c No |\n| **Versioned Knowledge** | \u2705 Yes | \u274c No |\n| **Graph-Based Relationships** | \u2705 Yes | \u274c No |\n| **Semantic & Vector Search** | \u2705 Yes | \u2705 Partial |\n| **Event-Driven Execution** | \u2705 Yes | \u274c No |\n| **Open-Source & Portable** | \u2705 Yes | \u2705 Partial |\n\n### **Who is MeshOS for?**\n\n\u2705 **Builders of AI-powered operations** \u2013 Structured memory and decision-making for AI-driven systems. \n\u2705 **Multi-agent system developers** \u2013 AI agents that need to store, process, and evolve shared knowledge. \n\u2705 **Developers & engineers** \u2013 Wanting an **open-source, PostgreSQL-powered framework** with no lock-in. \n\n---\n\n```mermaid\nflowchart LR\n %% Main System\n subgraph MeshOS[MeshOS System]\n direction LR\n\n %% Taxonomy Details\n subgraph Taxonomy[Memory Classification]\n direction TB\n \n subgraph DataTypes[Data Types]\n direction LR\n knowledge[Knowledge Type]\n activity[Activity Type]\n decision[Decision Type]\n media[Media Type]\n end\n\n subgraph Subtypes[Example Subtypes]\n direction LR\n k_types[Research/Mission/Vision]\n a_types[Conversations/Logs/Events]\n d_types[Policies/Strategies]\n m_types[Documents/Images]\n\n knowledge --> k_types\n activity --> a_types\n decision --> d_types\n media --> m_types\n end\n\n subgraph Relations[Edge Types]\n direction LR\n basic[related_to/version_of]\n semantic[influences/depends_on]\n temporal[follows_up/precedes]\n end\n end\n\n %% Memory Operations\n subgraph MemoryEngine[Memory Operations]\n direction LR\n rememberAction[Store/Remember]\n recallAction[Search/Recall]\n linkAction[Link Memories]\n versioning[Version Control]\n\n rememberAction --> recallAction\n recallAction --> linkAction\n linkAction --> versioning\n end\n end\n\n %% Organization & Agents\n subgraph Organization[Organization & Agents]\n direction TB\n\n %% Company Memory\n subgraph CompanyMemory[Company-Wide Memory]\n direction LR\n corpVision[Company Vision]\n corpMission[Company Mission]\n corpData[Knowledge Base]\n end\n\n %% Agents\n subgraph Agent1[Research Agent]\n a1Mem[Research Memories]\n end\n\n subgraph Agent2[Service Agent]\n a2Mem[Service Memories]\n end\n end\n\n %% System Connections\n Taxonomy --> MemoryEngine\n MemoryEngine --> Organization\n\n %% Memory Connections\n corpVision -.->|influences| a1Mem\n corpMission -.->|guides| a2Mem\n a1Mem -.->|shares| a2Mem\n a2Mem -.->|feedback| corpData\n a1Mem -.->|versions| corpData\n\n %% Styling\n classDef system fill:#dfeff9,stroke:#333,stroke-width:1.5px\n classDef engine fill:#fcf8e3,stroke:#333\n classDef taxonomy fill:#e7f5e9,stroke:#333\n classDef types fill:#f8f4ff,stroke:#333\n classDef org fill:#f4f4f4,stroke:#333\n\n class MeshOS system\n class MemoryEngine engine\n class Taxonomy,DataTypes,Subtypes,Relations taxonomy\n class Organization org\n```\n\n---\n\n## Getting Started\n\n### Install & Create a New Instance\n```bash\npip install mesh-os\nmesh-os create my-project && cd my-project\nmesh-os up\n```\n\n## Usage\n```python\nfrom mesh_os import MeshOS\n\n# Initialize MeshOS\nos = MeshOS()\n\n# Register an agent\nagent = os.register_agent(name=\"AI_Explorer\")\n\n# Store structured knowledge\nmemory = os.remember(\n content=\"The Moon has water ice.\",\n agent_id=agent.id,\n metadata={\n \"type\": \"knowledge\",\n \"subtype\": \"fact\",\n \"tags\": [\"astronomy\", \"moon\"],\n \"version\": 1\n }\n)\n\n# Retrieve similar knowledge\nresults = os.recall(query=\"Tell me about the Moon.\")\n```\n\n---\n\n## \ud83c\udfd7\ufe0f Core Features\n\n\u2705 **Memory for Multi-Agent Systems** \u2013 Let agents store, retrieve, and link structured knowledge. \n\u2705 **Fast Semantic Search** \u2013 pgvector-powered similarity matching across all memories. \n\u2705 **Graph-Based Knowledge** \u2013 Build evolving relationships between facts, ideas, and actions. \n\u2705 **Versioning Built-In** \u2013 Track updates, past decisions, and context shifts. \n\u2705 **Event-Driven Execution** \u2013 Automate workflows based on new knowledge. \n\u2705 **Open & Portable** \u2013 Runs anywhere PostgreSQL does. No black-box infrastructure. \n\n---\n\n## \ud83d\udd17 Structured Taxonomy & Memory Graph\n\nMeshOS **enforces structured knowledge** with **memory classification** and **versioning**:\n\n| **Memory Type** | **Examples** |\n| --------------- | -------------------------------------------- |\n| **Knowledge** | Research reports, datasets, concepts |\n| **Activity** | Agent workflows, logs, system events |\n| **Decision** | Policy updates, business strategy |\n| **Media** | Documents, images, AI-generated content |\n\nMemories **evolve** over time, with full versioning and relationship tracking.\n\n---\n\n## \ud83d\udee0\ufe0f Development & Configuration\n\n### **Configuration**\n```ini\n# Required\nOPENAI_API_KEY=your_api_key_here\n\n# Optional (defaults shown)\nPOSTGRES_PASSWORD=mysecretpassword\nHASURA_ADMIN_SECRET=meshos\nPOSTGRES_PORT=5432\nHASURA_PORT=8080\nHASURA_ENABLE_CONSOLE=true\n```\n\n### **Development**\n```bash\ngit clone https://github.com/yourusername/mesh-os.git\ncd mesh-os\npoetry install\npoetry run pytest\n```\n\n### **Contributing**\nContributions are welcome! Please submit a Pull Request.\n\n---\n\n## \u2696\ufe0f License\nThis project is licensed under the Apache 2.0 License \u2013 see [LICENSE](./LICENSE) for details.\n\n\n",
"bugtrack_url": null,
"license": null,
"summary": "MeshOS - A lightweight multi-agent memory system with semantic search",
"version": "0.1.11",
"project_urls": {
"Repository": "https://github.com/Props-Labs/mesh-os"
},
"split_keywords": [
"ai",
" agents",
" memory",
" semantic-search",
" vector-database"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "a8a0e0b585f61005d3926559a7def8fba402246d3f3443ce0e4a5973739662d1",
"md5": "2688c1fb0c797d086c4a6524b6118ea1",
"sha256": "246dd31fd5ea3084595c5e1bc8f80db506d95781063a93ea91f9881159af050f"
},
"downloads": -1,
"filename": "mesh_os-0.1.11-py3-none-any.whl",
"has_sig": false,
"md5_digest": "2688c1fb0c797d086c4a6524b6118ea1",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<4.0,>=3.9",
"size": 44486,
"upload_time": "2025-02-06T03:03:32",
"upload_time_iso_8601": "2025-02-06T03:03:32.849303Z",
"url": "https://files.pythonhosted.org/packages/a8/a0/e0b585f61005d3926559a7def8fba402246d3f3443ce0e4a5973739662d1/mesh_os-0.1.11-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "b21d06dc224ba868fb77586422420977e65543f3928e9fbb3902c5ce46f0a3d5",
"md5": "4d875e179886fce6e3e8aa39bab2b16a",
"sha256": "bc572f38db26addc39fb3ac15d57664dea45055efc35d3bb60135e61bba06d62"
},
"downloads": -1,
"filename": "mesh_os-0.1.11.tar.gz",
"has_sig": false,
"md5_digest": "4d875e179886fce6e3e8aa39bab2b16a",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<4.0,>=3.9",
"size": 35265,
"upload_time": "2025-02-06T03:03:35",
"upload_time_iso_8601": "2025-02-06T03:03:35.164015Z",
"url": "https://files.pythonhosted.org/packages/b2/1d/06dc224ba868fb77586422420977e65543f3928e9fbb3902c5ce46f0a3d5/mesh_os-0.1.11.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-02-06 03:03:35",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "Props-Labs",
"github_project": "mesh-os",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "mesh-os"
}