Open Index: A Deterministic Memory Layer for Your AI Agents

Summary

Open Index is a powerful tool for building domain-specific, accurate, and structured data that AI agents can effectively operate on. It enables the creation of a "brain," a searchable and continuously improving context graph tailored to any domain. This system ensures agents have access to reliable, up-to-date information, enhancing their capabilities and decision-making processes.

Use at your own risk

OSRepos shares public repositories for knowledge and discovery only. Any installation, execution, configuration, or use of code from these repositories is the user's own responsibility. Always review the repository, source code, dependencies, licenses, and security implications before running or installing anything. OSRepos is not responsible for issues, damages, or losses resulting from third-party repositories.

Introduction

Open Index is a powerful tool designed for building domain-specific, accurate, and structured data that AI agents can effectively operate on. It also ensures this data remains correct as circumstances evolve. At its core, Open Index helps you construct a "brain," a searchable and continuously improving context graph tailored to your specific domain. This brain is domain-agnostic, capable of modeling diverse scenarios, from a support organization to a sales pipeline or infrastructure.

The brain is built upon four fundamental primitives:

doc_type: A concept you wish to track and maintain, such as service, customer, or issue.
doc_schema: Defines the fields stored for a given doc_type.
entity: An instance of a doc_type, stored according to its schema. Entities can link to others via related_to and relationship_edge_meaning.
connector: An optional source for extracting entities, for example, an MCP server.

Why Use Open Index and Key Benefits

Open Index provides a robust context layer for domain-specialized agents, enabling them to access structured knowledge and maintain it effectively. Its MCP server publishes dynamic, brain-specific instructions, allowing agents to navigate the domain before tool calls. Agents can read context using search_brain() and get_entity(), and write back learnings via put_entity() or create_doc_type(), fostering a continuously improving loop.

Key benefits include:

Structured Context for AI Agents: Provides a reliable, validated way for agents to interact with domain knowledge.
Domain Agnostic Flexibility: Model any domain, from customer support to infrastructure, with custom concepts and relationships.
Continuous Improvement: Agents can update and refine the knowledge graph, ensuring data accuracy over time.
Powerful Search and Exploration: Features like per-field boosting, semantic search, and a UI explorer with map and analytics views.
Flexible Storage Options: Supports SQLite for local development and OpenSearch for scalable, multi-writer production environments.
Granular Data Control: Distinguish between file-backed (git-tracked) and index-backed (DB-owned) entities.
Easy Deployment: Options for local setup, Docker, and HTTP serving for cloud agents.

Installation

To get started with Open Index, you can install it using pip. The .[all] extra includes the core functionality, the explorer UI, and the MCP server.

pip install -e '.[all]'

Examples

Open Index ships with several runnable examples to help you understand its capabilities. You can try the bundled support-brain, infra-brain, or personal-brain.

To try an example:

open-index index --brain examples/support-brain
open-index ui    --brain examples/support-brain      # open the Map tab, pick an anchor

Alternatively, you can start your own brain from scratch:

open-index init my-brain
open-index add-doc-type customer --brain my-brain
# ...add entities under my-brain/entities/**/*.json...
open-index index --brain my-brain
open-index ui    --brain my-brain