# Open Index

## Docs

- [Open Index](https://open-index.io/index.md): Build domain-specific, structured context that agents can actually operate on — and keep it correct as things change.
- [Why Open Index](https://open-index.io/why-open-index.md): Why file- and markdown-based context management breaks down for autonomous agents — and what structured context management solves.
- [Quickstart](https://open-index.io/quickstart.md): Install Open Index, run the bundled example brain, and open the explorer.
- [Core concepts](https://open-index.io/concepts.md): The four primitives a brain is built from, and how entities link into a graph.
- [Use cases](https://open-index.io/use-cases.md): Example domains people model as a brain.
- [Creating a brain](https://open-index.io/guides/creating-a-brain.md): Scaffold a brain, define doc_types, and author entities step by step.
- [Populating entities](https://open-index.io/guides/populating-entities.md): Four ways to add entities to a brain, all landing in one validated store.
- [Entity management](https://open-index.io/guides/entity-management.md): How often to create entities, how many, and how to keep a brain from filling with noise.
- [Search configuration](https://open-index.io/guides/search-configuration.md): Where entities live, which engine stores them, and how search behaves.
- [Open Index as your agent's context layer](https://open-index.io/agents/mcp.md): Expose a brain over MCP so agents can read domain context and keep it current.
- [Connectors](https://open-index.io/agents/connectors.md): Pull entities from someone else's MCP server into your brain on a schedule.
- [Deployment](https://open-index.io/deployment.md): Deploy a brain and connect your agent — local, remote, Docker, TLS, and the exact config to paste.
- [CLI reference](https://open-index.io/reference/cli.md): Every open-index command and what it does.
- [Building autonomous agents](https://open-index.io/guide/index.md): A practical guide to the failure modes of autonomous agents and the context-engineering practices that fix them.
- [What are autonomous agents?](https://open-index.io/guide/what-are-autonomous-agents.md): A working definition of autonomous agents, the human-in-the-loop spectrum, and what it actually takes to remove the human.
- [The harness for autonomous agents](https://open-index.io/guide/the-harness.md): The control-flow scaffolding around the model — soft gates, critic sub-agents, in-session evals — and where the context layer fits.
- [How to reduce context poisoning in your agents](https://open-index.io/guide/context-poisoning.md): Why markdown/file-based context makes agents drift, and how a structured, discoverable context model contains the blast radius.
- [How to improve context search accuracy](https://open-index.io/guide/context-search-accuracy.md): Why agentic search with navigation beats pre-injected RAG, and how field boosts, hybrid ranking, and schemas make retrieval precise.
- [How to avoid staleness and conflicts in your knowledge window](https://open-index.io/guide/conflicts-and-staleness.md): Why distributed markdown knowledge self-contradicts over time, and how per-concept structured updates and time-decay keep a knowledge base correct.
- [How to ensure provenance in your AI agents](https://open-index.io/guide/provenance.md): Why users can't trust context they can't trace, and how per-fact attribution and citable retrieval build a proof-of-work agents can defend.
- [What is human steering, and how to reduce it](https://open-index.io/guide/human-steering.md): Human steering is the hidden tax on agent autonomy. What causes it, why it caps value, and the context and harness changes that reduce it.
- [How to increase determinism in your AI agents](https://open-index.io/guide/determinism.md): You can't make an LLM deterministic, but you can push variance out of everything around it — retrieval, control flow, and grounding — so behavior becomes repeatable.
- [How to build self-learning loops in your AI agents](https://open-index.io/guide/self-learning-loops.md): Agents that get better with use by writing what they learn back into a structured, validated knowledge base — without the drift of appending to prompts or markdown.