{
  "markdown": "<p align=\"center\">\n  <img src=\"docs/assets/ontorata.webp\" alt=\"Ontorata\" width=\"240\" />\n</p>\n\n<h1 align=\"center\">Ratary</h1>\n\n<p align=\"center\">\n  <strong>Build AI that remembers.</strong>\n</p>\n\n<p align=\"center\">\n  Ratary is the open-source <strong>AI Brain Platform</strong> — persistent memory,<br/>\n  structured knowledge, and intelligent retrieval for every model and agent you run.\n</p>\n\n<p align=\"center\">\n  <a href=\"#quick-start\"><strong>Get started →</strong></a>\n  &nbsp;·&nbsp;\n  <a href=\"https://github.com/ontorata/ratary\">Star on GitHub</a>\n  &nbsp;·&nbsp;\n  <a href=\"https://ontorata.com\">Website</a>\n  &nbsp;·&nbsp;\n  <a href=\"docs/GUIDE.md\">Docs</a>\n</p>\n\n<p align=\"center\">\n  <img src=\"https://img.shields.io/badge/license-MIT-blue\" alt=\"MIT License\"/>\n  <img src=\"https://img.shields.io/badge/MCP-native-indigo\" alt=\"MCP native\"/>\n  <img src=\"https://img.shields.io/badge/self--host-ready-brightgreen\" alt=\"Self-host\"/>\n  <img src=\"https://img.shields.io/badge/node-24.x-green\" alt=\"Node 24\"/>\n  <a href=\"https://www.npmjs.com/org/ratary\"><img src=\"https://img.shields.io/npm/v/@ratary/sdk?label=%40ratary%2Fsdk&color=cb3837\" alt=\"@ratary/sdk on npm\"/></a>\n  <a href=\"https://www.npmjs.com/package/@ratary/cli\"><img src=\"https://img.shields.io/npm/v/@ratary/cli?label=%40ratary%2Fcli&color=cb3837\" alt=\"@ratary/cli on npm\"/></a>\n  <a href=\"https://www.npmjs.com/package/@ratary/mcp-server\"><img src=\"https://img.shields.io/npm/v/@ratary/mcp-server?label=%40ratary%2Fmcp-server&color=cb3837\" alt=\"@ratary/mcp-server on npm\"/></a>\n</p>\n\n<p align=\"center\"><sub><em>Ratary is where AI remembers.</em> · Built by <a href=\"https://ontorata.com\">Ontorata</a></sub></p>\n\n<p align=\"center\"><sub>\n<a href=\"#quick-start\">Quick start</a> ·\n<a href=\"#ecosystem\">Ecosystem</a> ·\n<a href=\"#visual-architecture\">Architecture</a> ·\n<a href=\"#how-ratary-works\">How it works</a> ·\n<a href=\"#core-capabilities\">Capabilities</a> ·\n<a href=\"#documentation\">Docs</a>\n</sub></p>\n\n---\n\n## The problem\n\n*Why does AI forget between sessions?*\n\nEvery AI session starts from zero.\n\nYour model forgets yesterday's architecture decisions. Your agent drops customer context between runs. Your coding assistant can't recall why you chose Postgres over DynamoDB. Teams paste the same background into Cursor, Claude, ChatGPT, and custom bots — and knowledge still drifts.\n\nVector databases store chunks. RAG pipelines retrieve documents. Agent frameworks orchestrate tools.\n\n**None of them give AI a durable brain.**\n\n---\n\n## Why Ratary exists\n\n*Why build a brain layer now?*\n\nAI models are getting cheaper. Context windows are getting larger. Agents are getting capable.\n\n**But AI still forgets.**\n\nThe bottleneck is no longer reasoning. It's **memory** — durable, structured, retrievable, and owned by you.\n\nEvery serious application eventually needed a database. Every serious AI system will need a **brain layer**: persistent intelligence that sits between your models and your storage — independent of any single vendor, IDE, or agent framework.\n\n> **AI should remember.**  \n> **Developers should own that memory.**\n\nRatary exists to be that layer. Applications bring models. **Ratary brings the brain.**\n\n---\n\n## What Ratary is\n\n*What is Ratary?*\n\nRatary is an **AI Brain Platform** — infrastructure that gives AI:\n\n- **Persistent memory** — durable, owner-scoped, versioned\n- **Structured knowledge** — metadata, relations, graph traversal\n- **Intelligent retrieval** — hybrid search + bounded context assembly\n- **Protocol access** — Ratary MCP, REST, optional gRPC\n\nIt sits **between** AI clients and storage. One brain, many surfaces — Cursor, Claude Code, custom agents, enterprise APIs, and remote MCP hosts.\n\nThe runnable deployment is **[Ratary Server](#quick-start)** — this repository. **Ratary** is the product; **Ratary Server** is what you clone and run.\n\n**Bring your model. Ratary brings the memory.**\n\n---\n\n## Quick start\n\n*How do I run Ratary Server locally?*\n\n**Ratary** is the product. **Ratary Server** is the open-source deployment you run — [ontorata/ratary](https://github.com/ontorata/ratary) (this repository). [`@ratary/sdk`](#ecosystem), [`@ratary/cli`](#ecosystem), and [**Ratary MCP**](#ecosystem) connect to it; sibling [Ontorata products](#ecosystem) use the same source of truth.\n\n**Prerequisites:** Node.js 24 · **SQL metadata store** ([pick one](docs/CONFIGURATION.md#sql-metadata-store-choose-one) — **Postgres is the template default**)\n\n### Path A — PostgreSQL (npm + local or Docker)\n\n```bash\ngit clone https://github.com/ontorata/ratary.git\ncd ratary && npm install\ncp .env.example .env   # Set AUTH_SECRET + DATABASE_URL — see .env.example QUICK START\nnpm run db:apply-postgres-schema\nnpm run setup          # wire Ratary MCP for Cursor, Claude Code, …\nnpm run dev\n```\n\nOr use Docker: `docker compose --profile postgres up --build` — see [docs/DOCKER.md](docs/DOCKER.md).\n\n### Path B — Cloudflare D1\n\n```bash\ngit clone https://github.com/ontorata/ratary.git\ncd ratary && npm install\ncp .env.example .env   # Set AUTH_SECRET + SQL_PROVIDER=d1 + CLOUDFLARE_* / D1_*\nnpm run db:migrate\nnpm run setup\nnpm run dev\n```\n\nDetails: **[docs/GUIDE.md](docs/GUIDE.md#1-setup)** · **[docs/CONFIGURATION.md](docs/CONFIGURATION.md)**\n\n→ API `http://localhost:9876` · Swagger `/docs`\n\n**First REST call:** bootstrap once to get an API key (`aic_...`) — see [GUIDE — First REST API key](docs/GUIDE.md#first-rest-api-key-optional).\n\n```bash\n# Save your first memory\ncurl -X POST http://localhost:9876/api/v1/memory \\\n  -H \"Authorization: Bearer aic_YOUR_KEY\" \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\"title\":\"Hello brain\",\"project\":\"demo\",\"content\":\"Ratary remembers this.\"}'\n```\n\nFull guide: **[docs/GUIDE.md](docs/GUIDE.md)** · SDK & MCP examples in [docs/examples/](docs/examples/)\n\n### npm packages (`@ratary`)\n\nClient libraries ship on npm under the **`@ratary`** scope — product name, published by [Ontorata](https://ontorata.com). No server clone required for SDK, CLI, or hosted MCP.\n\n```bash\nnpm install @ratary/sdk@1.1.0\nnpx @ratary/mcp-server@1.1.3          # remote REST → stdio MCP\nnpm install -g @ratary/cli@1.1.0      # operator CLI\n```\n\n| Package | Version | Install | Role |\n|---------|---------|---------|------|\n| [`@ratary/sdk`](https://www.npmjs.com/package/@ratary/sdk) | **1.1.0** | `npm install @ratary/sdk` | Typed REST client + `client.admin.*` |\n| [`@ratary/cli`](https://www.npmjs.com/package/@ratary/cli) | **1.1.0** | `npm install -g @ratary/cli` | Operator commands (`admin`, `connectors`) |\n| [`@ratary/mcp-server`](https://www.npmjs.com/package/@ratary/mcp-server) | **1.1.3** | `npx @ratary/mcp-server` | IDE MCP → hosted API |\n\nSet `RATARY_BASE_URL` and `RATARY_API_KEY` (`aic_...`). Details: **[packages/README.md](packages/README.md)** · [remote MCP install](docs/install/remote.md).\n\n---\n\n## Ecosystem\n\n*Which repository owns what?*\n\nThe [Visual architecture](#visual-architecture) diagram shows **logical layers inside Ratary Server**. **This diagram** shows **repository and product relationships** — what ships in this repo, what connects to it, and what lives in sibling Ontorata repositories. Both views describe the same platform from different angles.\n\nThroughout this README, **Ratary MCP** means the official memory MCP implementation (stdio in this repo · npm `@ratary/mcp-server` for hosted REST). It is not the same as **Ontorata MCP** (ecosystem gateway — separate repo).\n\n```text\n┌─────────────────────────────────────────────────────────────┐\n│                      Ratary Cloud (opt-in)                  │\n│         optional hosted deployment · not this repo          │\n└──────────────────────────────┬──────────────────────────────┘\n                               │\n┌──────────────────────────────▼──────────────────────────────┐\n│  Ratary Server          ← ontorata/ratary (this repo)       │\n└──────────────────────────────┬──────────────────────────────┘\n         │              │              │\n         ▼              ▼              ▼\n   @ratary/sdk    @ratary/cli   @ratary/mcp-server\n   (SDK)          (CLI)         (Ratary MCP · npm)\n         │              │              │\n         └──────────────┴──────────────┘\n                               │\n         ┌─────────────────────┴─────────────────────┐\n         ▼                     ▼                     ▼\n   Ontorata MCP         Ontorata Studio          Ontory Runtime\n   ontorata/ontorata-mcp ontorata/Ontorata-Studio  ontorata/ontory-runtime\n   ecosystem product     ecosystem product         (ADR-2119; was ontorata/ontory)\n```\n\n**Infrastructure** (ships from `ontorata/ratary` — server plus client packages):\n\n| Component | Repository | Role |\n|-----------|------------|------|\n| **Ratary Server** | [ontorata/ratary](https://github.com/ontorata/ratary) | Memory engine — REST, persistence, Ratary MCP stdio. **You run this.** |\n| **Ratary SDK** | [npm `@ratary/sdk`](https://www.npmjs.com/package/@ratary/sdk) · `packages/sdk` | Typed REST client for Ratary Server. |\n| **Ratary CLI** | [npm `@ratary/cli`](https://www.npmjs.com/package/@ratary/cli) · `packages/cli` | Operator commands; delegates to Ratary SDK. |\n| **Ratary MCP** | [npm `@ratary/mcp-server`](https://www.npmjs.com/package/@ratary/mcp-server) · stdio in repo | Memory MCP — full stdio in clone · npm proxy for hosted REST. |\n\n**Ecosystem products** (separate repositories — connect to Ratary Server; not bundled here):\n\n| Product | Repository | Role |\n|---------|------------|------|\n| **Ontorata MCP** | [ontorata/ontorata-mcp](https://github.com/ontorata/ontorata-mcp) | Ecosystem MCP gateway — Ratary MCP plus additional Ontorata tools. |\n| **Ontorata Studio** | [ontorata/Ontorata-Studio](https://github.com/ontorata/Ontorata-Studio) | Operator UI — uses `@ratary/sdk` only. |\n| **Ontory** | [ontorata/ontory-runtime](https://github.com/ontorata/ontory-runtime) | Ontory Runtime — AI orchestration (ADR-2119). Legacy URL `ontorata/ontory` redirects. |\n\nRatary Server **does not depend** on ecosystem product repositories.\n\n### Ratary MCP vs Ontorata MCP\n\n*Which MCP should I install?*\n\n| | **Ratary MCP** | **Ontorata MCP** |\n|---|----------------|------------------|\n| **Layer** | Ratary infrastructure | Ontorata ecosystem product |\n| **What it is** | Official memory protocol for Ratary Server | Ecosystem gateway for Ontorata products |\n| **Scope** | Memory — CRUD, search, context, graph | Ratary memory plus additional Ontorata tools |\n| **Repository** | [ontorata/ratary](https://github.com/ontorata/ratary) · npm `@ratary/mcp-server` | [ontorata/ontorata-mcp](https://github.com/ontorata/ontorata-mcp) |\n| **Typical `mcp.json` key** | `ratary` | `ontorata` |\n\nUse **Ratary MCP** for direct memory access. Use **Ontorata MCP** for one MCP entry point across the Ontorata stack. Both use **Ratary Server** as source of truth.\n\n---\n\n## What Ratary is not\n\n*How is Ratary different from alternatives?*\n\n| | Vector DB | Memory API | RAG | Agent framework | **Ratary** |\n|---|:---:|:---:|:---:|:---:|:---:|\n| **Primary job** | Similarity search | Key-value recall | Document Q&A | Tool orchestration | **Durable AI memory** |\n| **Structured knowledge & graph** | ❌ | ⚠️ | ❌ | ⚠️ | ✅ |\n| **MCP-native IDE integration** | ❌ | ⚠️ | ❌ | ⚠️ | ✅ |\n| **Token-efficient context assembly** | ❌ | ⚠️ | ❌ | ⚠️ | ✅ |\n| **Self-host & data sovereignty** | ✅ | ⚠️ | ⚠️ | ⚠️ | ✅ |\n| **Clear agent boundary** | N/A | ⚠️ | N/A | ❌ bundled | ✅ substrate only |\n\nRatary **complements** your stack — it does not replace pgvector, LangGraph, or your agent of choice. See the **[Capability matrix](#capability-matrix)** for a feature-level comparison.\n\n<details>\n<summary><strong>Why not just pgvector, Mem0, Letta, LangGraph, or RAG?</strong></summary>\n\n| If you only use… | You get… | What you miss |\n|------------------|----------|---------------|\n| **pgvector** | Embedding similarity | Structured memory, graph, MCP, context packing |\n| **Mem0** | Fast hosted memory API | Full self-host, hybrid retrieval, enterprise adapters |\n| **Letta** | Agent + memory bundled | Your agent stays yours — Ratary is substrate, not runtime |\n| **LangGraph** | Workflow & tool routing | Shared durable memory across sessions and clients |\n| **RAG** | Document chunks | Evolving memory — decisions, handoffs, relations |\n\n</details>\n\n---\n\n## Visual architecture\n\n*How is Ratary Server structured internally?*\n\nThis diagram shows the **logical internal architecture** of Ratary — how memory, knowledge, retrieval, and storage layers compose inside the platform. It is **not** a repository or product map.\n\n```\n        ┌─────────────────────────────────────────┐\n        │           Your AI applications           │\n        │  Cursor · Claude · Agents · REST · MCP   │\n        └────────────────────┬────────────────────┘\n                             │\n                    MCP · REST · gRPC\n                             │\n        ┌────────────────────▼────────────────────┐\n        │     Ratary Server (logical layers)       │\n        │  ┌─────────┐ ┌──────────┐ ┌───────────┐ │\n        │  │ Memory  │ │Knowledge │ │ Retrieval │ │\n        │  └────┬────┘ └────┬─────┘ └─────┬─────┘ │\n        │       └───────────┴─────────────┘       │\n        │         Context · Learning · Protocols   │\n        └────────────────────┬────────────────────┘\n                             │\n        ┌────────────────────▼────────────────────┐\n        │     Pluggable storage (your choice)      │\n        │  Postgres · Supabase · MariaDB · D1 · pgvector · Neo4j · │\n        │  R2/S3/MinIO · OpenSearch · ClickHouse · …        │\n        └─────────────────────────────────────────┘\n```\n\n**Search** browses. **Retrieval** injects context. **Embedding** enriches asynchronously — never on the CRUD hot path.\n\nDetails: **[docs-ai](https://github.com/ontorata/docs-ai)** → `products/ratary/roadmap/ARCHITECTURE.md`\n\nFor **repository and product relationships** (SDK, CLI, Ratary MCP, Ontorata ecosystem repos), see **[Ecosystem](#ecosystem)** — a separate diagram, same platform, different perspective.\n\n---\n\n## How Ratary works\n\n*What happens to a memory after you save it?*\n\n```\n   Write          Enrich         Retrieve        Learn          Reuse\n     │               │               │              │              │\n     ▼               ▼               ▼              ▼              ▼\n  Save via       Summarize,      Rank & pack    Signals,       Same memory\n  Ratary MCP/REST embed, link     context for    consolidate,   powers every\n                 relations       your prompt    evolve         client\n```\n\n1. **Write** — Persist memory through Ratary MCP or REST.\n2. **Enrich** — Summarize, embed, and relate — asynchronously.\n3. **Retrieve** — Assemble the smallest useful context slice.\n4. **Learn** — Optional signals and consolidation improve recall over time.\n5. **Reuse** — One brain across IDEs, agents, and APIs.\n\n---\n\n## Core capabilities\n\n*What can Ratary Server do today?*\n\n### Memory intelligence\nDurable memories with summaries, codenames, favorites, archives, and handoffs. Version history with restore and merge — built for long-running work, not chat logs.\n\n### Knowledge\nSemantic enrichment, relation linking, and graph traversal. Memory becomes navigable knowledge — not a flat pile of notes.\n\n### Retrieval\nHybrid search across SQL, vectors, lexical index, and graph. Separate **browse** from **inject**. Optional precision modes (hybrid, semantic, fulltext, title) when you need more control.\n\n### Reasoning support\nProgressive retrieval, token budgets, and summary-first context assembly — typically **~85% fewer tokens** than dumping full memory bodies into prompts.\n\n### Learning\nQuality signals, consolidation, and compression — optional pipelines that improve the brain over time without retraining your model.\n\n### External agent support\nCapability manifests, workspace scoping, and **28 Ratary MCP tools**. External agents discover what the brain can do; Ratary never embeds agent reasoning — see [What Ratary is not](#what-ratary-is-not).\n\n### Platform\nPluggable adapters: choose **SQL metadata** (Postgres, Supabase, MariaDB/MySQL, D1, TiDB/Cockroach) plus optional pgvector, R2/S3/MinIO, Azure Blob, GCS, Meilisearch, OpenSearch, Neo4j, Redis, DuckDB, ClickHouse. Same application code for every backend.\n\n**Self-host stacks:** [docs/DOCKER.md](docs/DOCKER.md) — `postgres` or `enterprise` (MariaDB + MinIO + Redis) profiles.\n\n### Cloud & enterprise\nSelf-host, deploy to Vercel, or run a control plane with metering and federation. RBAC workspaces, audit trails, SSO, and policy hooks — opt-in when you need them.\n\n### Observability\nOpenTelemetry, Prometheus metrics, SLO dashboards, and cost visibility for production brains.\n\n### Developer experience\nOpenAPI, npm **`@ratary/sdk@1.1.0`** · **`@ratary/cli@1.1.0`** · **`@ratary/mcp-server@1.1.3`** — and one-command IDE setup (`npm run setup`).\n\n### Knowledge fabric (opt-in)\nIngest from external systems of record — **Notion** (live on [hosted prod](https://ratary.ontorata.com)), **Confluence**, **Google Drive**, **SharePoint**, and **Teams** connectors (code complete · enable per connector via env). Webhook HMAC, incremental sync, provenance on memories. Flags: `KNOWLEDGE_FABRIC_ENABLED` + `CONNECTOR_SYNC_ENABLED`. Guides: [Knowledge fabric](docs/GUIDE.md#12-knowledge-fabric-live-connectors) · [Production enable](docs/PRODUCTION-ENABLE.md) · Phases 32–34 in [docs-ai](https://github.com/ontorata/docs-ai).\n\n---\n\n## Use cases\n\n*Who is Ratary for?*\n\n| | What you build | What Ratary provides |\n|---|----------------|---------------------|\n| **Developer AI** | Coding assistants across IDEs and sessions | Persistent project memory, MCP tools, handoffs |\n| **Enterprise search** | Internal knowledge discovery | Hybrid retrieval over structured memory, not just files |\n| **Customer support** | AI that handles tickets | Durable customer context without re-prompting every thread |\n| **Knowledge management** | Team second brain | Graph-linked memories, codenames, relations, summaries |\n| **Autonomous agents** | Multi-agent systems | Shared memory layer with workspace and agent scoping |\n| **Personal AI** | Private assistant you own | Self-hosted, exportable, sovereign data |\n\n---\n\n## Capability matrix\n\n*How does Ratary compare feature-by-feature?*\n\nFor category positioning, see **[What Ratary is not](#what-ratary-is-not)**.\n\n| Capability | Ratary | Vector DB | Memory API | RAG | Agent framework |\n|------------|:------:|:---------:|:----------:|:---:|:---------------:|\n| Persistent structured memory | ✅ | ❌ | ⚠️ | ❌ | ⚠️ |\n| MCP-native | ✅ | ❌ | ⚠️ | ❌ | ⚠️ |\n| Hybrid SQL + vector + graph | ✅ | ⚠️ | ⚠️ | ⚠️ | ⚠️ |\n| Token-efficient context assembly | ✅ | ❌ | ⚠️ | ❌ | ⚠️ |\n| Knowledge graph & relations | ✅ | ❌ | ⚠️ | ❌ | ⚠️ |\n| Self-host sovereignty | ✅ | ✅ | ⚠️ | ⚠️ | ⚠️ |\n| Agent boundary (bring your agent) | ✅ | N/A | ⚠️ | N/A | ❌ |\n| Enterprise storage adapters | ✅ | ✅ | ⚠️ | ⚠️ | ⚠️ |\n\n---\n\n## Documentation\n\n*Where do I read next?*\n\n### Engineering knowledge (internal)\n\nArchitecture, ADRs, roadmap, phases, design history, and reviews live in the Ontorata **Knowledge OS**:\n\n**https://github.com/ontorata/docs-ai** → `products/ratary/`\n\n### Operator / developer guides (this repository)\n\n**Ratary Server** (`ontorata/ratary` — this repository):\n\n| | |\n|---|---|\n| [docs/GUIDE.md](docs/GUIDE.md) | Setup, daily usage, Ratary MCP configuration |\n| [docs/install/README.md](docs/install/README.md) | Per-harness MCP / plugin installation |\n| [docs/DOCKER.md](docs/DOCKER.md) | Container & Compose self-host |\n| [docs/CONFIGURATION.md](docs/CONFIGURATION.md) | Environment variables — what each flag does |\n| [docs/examples/](docs/examples/) | MCP and IDE config templates |\n| [MCP/README.md](MCP/README.md) | Ratary MCP — stdio and `@ratary/mcp-server` |\n| [packages/README.md](packages/README.md) | npm packages — install, env, publish |\n| [.env.example](.env.example) | Env template — meanings in [docs/CONFIGURATION.md](docs/CONFIGURATION.md) |\n| [docs/PRODUCTION-ENABLE.md](docs/PRODUCTION-ENABLE.md) | Hosted deploy — knowledge fabric on Vercel |\n| [docs/OPS-PRODUCTION-VERIFY.md](docs/OPS-PRODUCTION-VERIFY.md) | Production ops checklist — `npm run ops:verify-production` |\n| [docs/MCP-CHATGPT-OAUTH.md](docs/MCP-CHATGPT-OAUTH.md) | ChatGPT MCP OAuth + Keycloak IdP runbook |\n| [MCP/submission/directory-status.md](MCP/submission/directory-status.md) | MCP directory listing status |\n| [docs/ENTERPRISE-MODULES.md](docs/ENTERPRISE-MODULES.md) | Enterprise flags (opt-in) |\n| [CHANGELOG.md](CHANGELOG.md) | Release notes and version map |\n| [SECURITY.md](SECURITY.md) | Vulnerability reporting |\n| [ontorata/docs-ai](https://github.com/ontorata/docs-ai) | Architecture · ADR · phases · evidence (Knowledge OS) |\n\n**Canonical hosted API:** `https://ratary.ontorata.com` (self-host uses your own base URL).\n\n**Ontorata ecosystem** (separate repositories — not in this tree):\n\n| | |\n|---|---|\n| [ontorata/ontorata-mcp](https://github.com/ontorata/ontorata-mcp) | Ontorata MCP — ecosystem gateway |\n| [ontorata/Ontorata-Studio](https://github.com/ontorata/Ontorata-Studio) | Ontorata Studio — operator UI ([setup](docs/GUIDE.md#ontorata-studio-web-console)) |\n\n---\n\n## Roadmap\n\n*What is shipping when?*\n\n**Canonical roadmap / phase history:** [ontorata/docs-ai](https://github.com/ontorata/docs-ai) → `products/ratary/roadmap/` · `products/ratary/phases/`\n\nOrganized by direction — not sprints. Phases **1–31** are implemented in code (gates PASS); platform modules stay **opt-in via env** unless noted.\n\n| | Themes | Code | Ops (prod) | Primary repository |\n|---|--------|:----:|:----------:|-------------------|\n| **Today (v1.0)** | Ratary MCP + REST, hybrid/graph retrieval, peer SQL, Docker, npm [`@ratary/sdk@1.1.0`](https://www.npmjs.com/package/@ratary/sdk) · [`@ratary/mcp-server@1.1.3`](https://www.npmjs.com/package/@ratary/mcp-server), remote MCP, [Ontorata Studio](https://github.com/ontorata/Ontorata-Studio). **Platform (opt-in):** knowledge fabric (Notion/Confluence/Drive/SharePoint/Teams live), universal memory fabric (Phase 32), Neptune traversal (Phase 33), federation, global intelligence | ✅ | Partial | `ontorata/ratary` |\n| **Ops (now)** | Prod connector creds · MCP directory follow-ups · ChatGPT OAuth IdP (Keycloak) | ✅ | ⏳ | `ontorata/ratary` |\n\n**Ops status (2026-07-19):** Notion + fabric/federation flags on Vercel ✅ · D1 `db:migrate` ✅ · SDK codegen CI ✅ ([workflow](https://github.com/ontorata/ratary/actions/workflows/sdk-codegen.yml)) · Confluence/Drive/SharePoint/Teams creds ⏳ (owner secrets) · Keycloak IdP ⏳ (`auth.ontorata.com` → deploy Render; prod uses Smithery API-key mode until OAuth enabled) · MCP listings mostly **Listed** ([directory-status.md](MCP/submission/directory-status.md)). Verify: `npm run ops:verify-production` · [OPS-PRODUCTION-VERIFY.md](docs/OPS-PRODUCTION-VERIFY.md)\n\nEnterprise modules ship **opt-in via environment flags** on Ratary Server — defaults stay lean. See [ENTERPRISE-MODULES.md](docs/ENTERPRISE-MODULES.md) and [CONFIGURATION.md](docs/CONFIGURATION.md).\n\n---\n\n## Vision\n\n*What is Ratary building toward?*\n\nToday every application has a database.\n\nTomorrow every AI will have a brain.\n\n**Ratary is building that layer** — open, portable, self-hostable, and protocol-native. Not another chat wrapper. Not another vector dump. Infrastructure for persistent intelligence.\n\nKnowledge should **accumulate**. Boundaries should be **respected**. Agents should stay **coherent across time**.\n\nIf you're building AI that lasts longer than a single prompt — **build on Ratary**.\n\n<p align=\"center\"><br/>\n<strong>Bring your model. Bring your agent. Ratary brings the brain.</strong><br/><br/>\n<sub>Ratary · where AI remembers · <a href=\"https://ontorata.com\">Ontorata</a> · MIT License</sub>\n</p>\n\n---\n\n## Contributing\n\n*How do I contribute?*\n\n**Ratary Server** (this repo): fork [ontorata/ratary](https://github.com/ontorata/ratary) → branch → `npm run lint && npm run build && npm test` → PR to `ontorata/ratary`.\n\n**Standard PR gate:**\n\n```bash\nnpm run lint && npm run build && npm test\n```\n\n**Org-memory / recall changes** (maintainers with `.ai/` workspace): run the full gate before merge — see [docs/RATARY-VALIDATION-RUNBOOK.md](docs/RATARY-VALIDATION-RUNBOOK.md):\n\n```bash\nnpm run ci:ratary-validation\n```\n\n**Production ops changes** (Vercel env, connectors, OAuth): see [docs/OPS-PRODUCTION-VERIFY.md](docs/OPS-PRODUCTION-VERIFY.md):\n\n```bash\nnpm run ops:verify-production\n```\n\nExtended governance (`.ai/` phases, ADRs) lives in the [development mirror](https://github.com/lutfi04/ai-brain) — optional for contributors; **docs-only and standard PRs to `ontorata/ratary` are welcome** without the mirror.\n\n**Ontorata MCP** and **Ontorata Studio** accept contributions in their own repositories — not via this repo.\n\nQuestions: [hello@ontorata.com](mailto:hello@ontorata.com)\n",
  "bytes": 24923,
  "sha": "0026b9276b864efab0e9b519f9e3a99df8eb8914cb846affe082355056dc474b",
  "repo_slug": "ontorata/ratary",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_ontorata_ratary_21614a7a/readme"
}