Skip to content
Back to search

beever-atlas

Your First LLM-Wiki Conversation Knowledge Base

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/beever-ai--beever-atlas/badge.svg)](/listings/beever-ai--beever-atlas)

HTML

<a href="/listings/beever-ai--beever-atlas"><img src="/api/listings/beever-ai--beever-atlas/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Beever Atlas ships as a Docker Compose stack (backend + bot + web + 4 datastores). You can try a seeded demo in 30 seconds with zero keys, then pick one of **three deployment options** to install it for real.

1. Get the code

git clone https://github.com/beever-ai/beever-atlas.git
cd beever-atlas

2. Try the demo first (optional, no keys needed for seeding)

make demo

`make demo` brings up the full stack pre-loaded with a public Wikipedia corpus (Ada Lovelace + Python history). Seeding uses pre-computed fixtures — no API keys required. Asking questions via `/api/ask` needs a free-tier `GOOGLE_API_KEY` because the QA agent calls Gemini. See [demo/README.md](demo/README.md) for curl examples.

Skip this step if you're ready to install for real.

3. Before you start: get your API keys

Two free keys are required before installing. Both offer generous free tiers — enough to sync a small team's channels for testing.

| Key | Purpose | Where to get it | |---|---|---| | `GOOGLE_API_KEY` | Gemini — extraction, entity graph, answers | [aistudio.google.com/apikey](https://aistudio.google.com/apikey) | | `JINA_API_KEY` | Jina v4 embeddings (2048-dim) for semantic search | [jina.ai/api-dashboard](https://jina.ai/api-dashboard/) |

Optional (skip unless you know you need them):

| Key | What it enables | |---|---| | `TAVILY_API_KEY` | External web search when QA retrieval confidence is low — [tavily.com](https://tavily.com/) | | `OLOSTEP_API_KEY` | Olostep web search (alternative to Tavily). Set `WEB_SEARCH_PROVIDER=olostep` in your `.env` — [olostep.com/dashboard](https://www.olostep.com/dashboard) | | Slack / Discord / Teams bot tokens | **Configured via the web UI after setup**, not `.env` — the bot stores platform credentials encrypted in MongoDB |

> **Tip:** Keep the two required keys handy before you start. Option 1 prompts for them interactively; Options 2 and 3 need them pasted into `.env`.

4. Choose a deployment option

| Option | When to use | Time to "up" | |---|---|---| | **1. One-line install** (recommended) | You want the fastest path to a running stack. | ~2 min first run | | **2. Manual Docker** | CI/CD, ops environments, or when you want explicit control over every step. | ~3 min first run | | **3. Local development** | Active contributors who need hot-reload on backend and frontend. | varies |

Option 1 — One-line install (recommended)

./atlas

The `atlas` installer walks you through a guided 5-step checklist:

1. **Embedding model** — pick a provider (Jina / OpenAI / Cohere / Voyage / Gemini / Mistral / Ollama), then its API key. 2. **Agent LLM provider** — pick a provider for the 16 ADK agents (Google Gemini / OpenAI / Anthropic / Mistral / DeepSeek / Groq / MiniMax / Ollama / Custom); optional second provider for hybrid setups. 3. **Graph backend** — Neo4j (default) or skip. 4. **Optional integrations** — Tavily web search, MCP server for Claude Code / Cursor. 5. **Auth tokens** — keep dev defaults or rotate now.

Under the hood it verifies `docker` + `docker compose`, copies `.env.example` → `.env` (preserves your values on re-run, `chmod 600`), auto-generates `CREDENTIAL_MASTER_KEY` (64 hex) and `WEAVIATE_API_KEY` (32 hex), runs a port-conflict preflight, launches the stack via `docker compose up -d --build --force-recreate --remove-orphans`, and polls `/api/health` before printing the ready card.

When you see **"Beever Atlas is ready"**, open **[http://localhost:3000](http://localhost:3000)** — then **Settings → AI Setup** to manage providers, assign LLMs per-agent, run Test Connection, or discover models. For CI / Docker / GitOps, configure declaratively: `BEEVER_LLM_API_KEY=...` (single-provider shortcut), `BEEVER_ENDPOINTS='[...]'` + `BEEVER_PRESET=...`, or commit an `atlas.yaml` and run `atlas apply` — see [`docs/runbooks/ai-setup.md`](docs/runbooks/ai-setup.md) and [`docs/runbooks/atlas-yaml.md`](docs/runbooks/atlas-yaml.md).

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 397 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <h1 align="center">
  2. <picture>
  3. <source media="(prefers-color-scheme: dark)" srcset="web/public/logo-white.svg" />
  4. <img src="web/public/logo-primary.svg" alt="" height="48" align="absmiddle" />
  5. </picture>
  6. &nbsp;Beever Atlas
  7. </h1>
  8. <div align="center">
  9. <picture>
  10. <source media="(prefers-color-scheme: dark)" srcset="assets/banner-dark.png" />
  11. <img src="assets/banner.png" alt="Beever Atlas — LLM-first Wiki Knowledge Base" width="85%" />
  12. </picture>
  13. </div>
  14. <h3 align="center">
  15. Turn your team's Slack, Discord, Teams &amp; Mattermost chats<br>
  16. into a self-maintaining wiki — automatically.
  17. </h3>
  18. <p align="center">
  19. <a href="https://docs.beever.ai/atlas"><img src="https://img.shields.io/badge/DOCS-docs.beever.ai/atlas-FFC107?style=for-the-badge&labelColor=4A4A4A" alt="Docs" /></a>
  20. <a href="LICENSE"><img src="https://img.shields.io/badge/LICENSE-Apache_2.0-7CB342?style=for-the-badge&labelColor=4A4A4A" alt="License Apache 2.0" /></a>
  21. <a href="https://beever.ai"><img src="https://img.shields.io/badge/BUILT_BY-BEEVER.AI-15404E?style=for-the-badge&labelColor=4A4A4A" alt="Built by Beever.ai" /></a>
  22. <a href="https://google.github.io/adk-docs/"><img src="https://img.shields.io/badge/BUILT_WITH-Google_ADK-FF6F00?style=for-the-badge&labelColor=4A4A4A" alt="Built with Google ADK" /></a>
  23. <a href="https://glama.ai/mcp/servers/Beever-AI/beever-atlas"><img src="https://img.shields.io/badge/MCP-Listed_on_Glama-A855F7?style=for-the-badge&labelColor=4A4A4A" alt="MCP server on Glama" /></a>
  24. </p>
  25. <p align="center">
  26. <a href="https://discord.gg/VshBCUUX"><img src="https://img.shields.io/badge/DISCORD-Join_Community-5865F2?style=for-the-badge&labelColor=4A4A4A&logo=discord&logoColor=white" alt="Join our Discord" /></a>
  27. <a href="https://x.com/Beever_AI"><img src="https://img.shields.io/badge/X-@Beever__AI-000000?style=for-the-badge&labelColor=4A4A4A&logo=x&logoColor=white" alt="Follow us on X" /></a>
  28. <a href="https://beever.ai/"><img src="https://img.shields.io/badge/WEBSITE-beever.ai-15404E?style=for-the-badge&labelColor=4A4A4A" alt="beever.ai" /></a>
  29. </p>
  30. <div align="center">
  31. <a href="https://glama.ai/mcp/servers/Beever-AI/beever-atlas">
  32. <img width="380" height="200" src="https://glama.ai/mcp/servers/Beever-AI/beever-atlas/badge" alt="Beever Atlas MCP server on Glama" />
  33. </a>
  34. </div>
  35. ---
  36. Beever Atlas pulls the conversations your team already has on Slack, Discord, Microsoft Teams, and Mattermost, extracts atomic facts, deduplicates them, and clusters them into topic pages with citations. A graph store links the people, decisions, and projects mentioned across channels. Ask questions in natural language and get answers cited back to the source messages — through the dashboard, or through MCP into Claude Code and Cursor.
  37. If you want a knowledge base that grows on its own from the chats your team already has, this is it.
  38. ---
  39. ## ✨ Features in action
  40. Six short clips — connect a workspace, sync history, watch memory build, browse the auto-generated wiki, ask questions, plug external AI agents in via MCP.
  41. <table>
  42. <tr>
  43. <td width="33%" align="center" valign="top">
  44. <strong>Multi-Platform</strong><br><br>
  45. <img src="assets/clips/multi-platform.gif" alt="Multi-platform connections demo" width="100%"><br>
  46. Connect Slack, Discord, Teams, Mattermost, or file imports. One bot, every workspace.
  47. </td>
  48. <td width="33%" align="center" valign="top">
  49. <strong>Message Sync</strong><br><br>
  50. <img src="assets/clips/sync.gif" alt="Channel sync demo" width="100%"><br>
  51. Pull channel history on demand or on a schedule. Resumable and rate-limit aware.
  52. </td>
  53. <td width="33%" align="center" valign="top">
  54. <strong>Memory Ingestion</strong><br><br>
  55. <img src="assets/clips/memory.gif" alt="Memory ingestion pipeline demo" width="100%"><br>
  56. 6-stage ADK pipeline distils messages into atomic facts, entities, and relationships.
  57. </td>
  58. </tr>
  59. <tr>
  60. <td width="33%" align="center" valign="top">
  61. <strong>LLM Wiki</strong><br><br>
  62. <img src="assets/clips/wiki.gif" alt="LLM wiki browsing demo" width="100%"><br>
  63. Auto-maintained wiki per channel — overview, topics, people, decisions, citations.
  64. </td>
  65. <td width="33%" align="center" valign="top">
  66. <strong>QA Agent</strong><br><br>
  67. <img src="assets/clips/qa.gif" alt="QA agent answering demo" width="100%"><br>
  68. Streams cited answers over SSE. Smart router picks semantic or graph per question.
  69. </td>
  70. <td width="33%" align="center" valign="top">
  71. <strong>MCP Server</strong><br><br>
  72. <img src="assets/clips/mcp.gif" alt="MCP server querying from Claude Code demo" width="100%"><br>
  73. Plug Claude Code / Cursor into your knowledge base — 28 tools, per-agent auth.
  74. </td>
  75. </tr>
  76. </table>
  77. ---
  78. ## 🏗️ Architecture
  79. Conversations from any supported platform flow into a unified ingestion pipeline that produces two complementary memory systems — a **3-tier semantic store** (channel / topic / atomic fact) for fast hybrid search, and a **graph store** that extracts entities and their relationships. Those memories fuel two consumer surfaces: the **LLM Wiki** (distilled, auto-maintained) and **QA Agents** (served through the dashboard directly, or through **MCP** into Claude Code / Cursor).
  80. <p align="center">
  81. <picture>
  82. <source media="(prefers-color-scheme: dark)" srcset="assets/architecture-dark.png" />
  83. <img src="assets/architecture.png" alt="Beever Atlas architecture — chat platforms → memory ingestion → 3-tier semantic memory + graph memory → LLM Wiki and QA Agent → Dashboard and MCP clients" width="60%" />
  84. </picture>
  85. </p>
  86. <p align="center"><em>From chat platforms to MCP agents — one ingestion path, two memory systems, two delivery surfaces.</em></p>
  87. Under the hood, three services (backend, bot, frontend) are backed by four data stores (Weaviate, Neo4j, MongoDB, Redis). See the [architecture overview](https://docs.beever.ai/atlas/concepts/architecture) on the documentation site for the full design — component responsibilities, dual-memory internals, and the smart query router.
  88. ---
  89. ## 💡 Why Wiki-First RAG?
  90. Most RAG systems answer questions by retrieving raw message snippets and feeding them straight to an LLM. Beever Atlas takes a different approach: it continuously distils conversations into a structured, auto-maintained wiki — with topic pages, entity graphs, decisions, and citations — before any query is issued. When you ask a question, the retrieval layer works against clean, deduplicated knowledge rather than noisy chat history. This means answers are more consistent, citations are traceable to source messages, and the wiki itself becomes a useful artifact your team can browse independently of the Q&A interface. The dual-memory architecture (semantic + graph) lets the query router pick the right retrieval strategy per question, keeping latency low and context precise.
  91. <div align="center">
  92. <img src="assets/wiki-preview.png" alt="Beever Atlas wiki view — auto-generated overview with concept map, topics, FAQ, glossary, and resources, built from Slack messages" width="85%" />
  93. </div>
  94. <p align="center"><em>A live auto-generated channel wiki: overview, concept map, topics, FAQ, glossary — distilled from 246 Slack messages, not hand-written.</em></p>
  95. ### The inspiration: LLMs read wikis, not chat logs
  96. The per-channel wiki concept is directly inspired by [Andrej Karpathy's observation](https://x.com/karpathy/status/2039805659525644595) that LLMs are far better at reasoning over curated, encyclopedic content (books, docs, wikis) than over raw conversational transcripts. Chat history is noisy, redundant, temporally scattered, and full of implicit context that only humans resolve. A wiki, by contrast, is the *already-distilled* form of that knowledge — deduplicated, structured, citation-bearing, and organised by topic rather than by timestamp.
  97. Beever Atlas operationalises this insight: every synced channel gets its own **auto-generated, continuously-updated wiki** — sections for topics, entities, decisions, open questions, and timelines — rebuilt incrementally as new messages arrive. The QA agent retrieves against this wiki first, falling back to raw messages only when a fact hasn't been distilled yet.
  98. ### What this unlocks in practice
  99. - **Better answers, fewer hallucinations** — retrieval operates on fact-dense prose with explicit entity relationships, not on fragmented turn-by-turn chat.
  100. - **Traceable citations** — every wiki claim links back to the source messages that produced it, so answers are auditable all the way down to the original Slack/Discord/Teams thread.
  101. - **A browsable artifact, not just a Q&A box** — the wiki is useful *on its own*. New teammates onboarding to a channel can read the distilled wiki instead of scrolling three months of history.
  102. - **Cheaper inference at query time** — the expensive distillation work happens once, at ingestion. Queries hit compact, pre-digested context instead of re-summarising raw logs on every request.
  103. - **Graph-aware reasoning** — the entity graph built alongside the wiki lets the query router answer relational questions ("who worked on X with Y?") that pure vector RAG struggles with.
  104. For a detailed comparison with other LLM knowledge tools, see [the comparison page](https://docs.beever.ai/atlas/comparison) on the documentation site.
  105. ---
  106. ## 🚀 Quick Start
  107. Beever Atlas ships as a Docker Compose stack (backend + bot + web + 4 datastores). You can try a seeded demo in 30 seconds with zero keys, then pick one of **three deployment options** to install it for real.
  108. ### 1. Get the code
  109. ```bash
  110. git clone https://github.com/beever-ai/beever-atlas.git
  111. cd beever-atlas
  112. ```
  113. ### 2. Try the demo first (optional, no keys needed for seeding)
  114. ```bash
  115. make demo
  116. ```
  117. `make demo` brings up the full stack pre-loaded with a public Wikipedia corpus (Ada Lovelace + Python history). Seeding uses pre-computed fixtures — no API keys required. Asking questions via `/api/ask` needs a free-tier `GOOGLE_API_KEY` because the QA agent calls Gemini. See [demo/README.md](demo/README.md) for curl examples.
  118. Skip this step if you're ready to install for real.
  119. ### 3. Before you start: get your API keys
  120. Two free keys are required before installing. Both offer generous free tiers — enough to sync a small team's channels for testing.
  121. | Key | Purpose | Where to get it |
  122. |---|---|---|
  123. | `GOOGLE_API_KEY` | Gemini — extraction, entity graph, answers | [aistudio.google.com/apikey](https://aistudio.google.com/apikey) |
  124. | `JINA_API_KEY` | Jina v4 embeddings (2048-dim) for semantic search | [jina.ai/api-dashboard](https://jina.ai/api-dashboard/) |
  125. Optional (skip unless you know you need them):
  126. | Key | What it enables |
  127. |---|---|
  128. | `TAVILY_API_KEY` | External web search when QA retrieval confidence is low — [tavily.com](https://tavily.com/) |
  129. | `OLOSTEP_API_KEY` | Olostep web search (alternative to Tavily). Set `WEB_SEARCH_PROVIDER=olostep` in your `.env` — [olostep.com/dashboard](https://www.olostep.com/dashboard) |
  130. | Slack / Discord / Teams bot tokens | **Configured via the web UI after setup**, not `.env` — the bot stores platform credentials encrypted in MongoDB |
  131. > **Tip:** Keep the two required keys handy before you start. Option 1 prompts for them interactively; Options 2 and 3 need them pasted into `.env`.
  132. ### 4. Choose a deployment option
  133. | Option | When to use | Time to "up" |
  134. |---|---|---|
  135. | **1. One-line install** (recommended) | You want the fastest path to a running stack. | ~2 min first run |
  136. | **2. Manual Docker** | CI/CD, ops environments, or when you want explicit control over every step. | ~3 min first run |
  137. | **3. Local development** | Active contributors who need hot-reload on backend and frontend. | varies |
  138. ### Option 1 — One-line install (recommended)
  139. ```bash
  140. ./atlas
  141. ```
  142. The `atlas` installer walks you through a guided 5-step checklist:
  143. 1. **Embedding model** — pick a provider (Jina / OpenAI / Cohere / Voyage / Gemini / Mistral / Ollama), then its API key.
  144. 2. **Agent LLM provider** — pick a provider for the 16 ADK agents (Google Gemini / OpenAI / Anthropic / Mistral / DeepSeek / Groq / MiniMax / Ollama / Custom); optional second provider for hybrid setups.
  145. 3. **Graph backend** — Neo4j (default) or skip.
  146. 4. **Optional integrations** — Tavily web search, MCP server for Claude Code / Cursor.
  147. 5. **Auth tokens** — keep dev defaults or rotate now.
  148. Under the hood it verifies `docker` + `docker compose`, copies `.env.example` → `.env` (preserves your values on re-run, `chmod 600`), auto-generates `CREDENTIAL_MASTER_KEY` (64 hex) and `WEAVIATE_API_KEY` (32 hex), runs a port-conflict preflight, launches the stack via `docker compose up -d --build --force-recreate --remove-orphans`, and polls `/api/health` before printing the ready card.
  149. When you see **"Beever Atlas is ready"**, open **[http://localhost:3000](http://localhost:3000)** — then **Settings → AI Setup** to manage providers, assign LLMs per-agent, run Test Connection, or discover models. For CI / Docker / GitOps, configure declaratively: `BEEVER_LLM_API_KEY=...` (single-provider shortcut), `BEEVER_ENDPOINTS='[...]'` + `BEEVER_PRESET=...`, or commit an `atlas.yaml` and run `atlas apply` — see [`docs/runbooks/ai-setup.md`](docs/runbooks/ai-setup.md) and [`docs/runbooks/atlas-yaml.md`](docs/runbooks/atlas-yaml.md).
  150. For CI or unattended installs — skip prompts, pre-seed keys from shell env:
  151. ```bash
  152. GOOGLE_API_KEY=... JINA_API_KEY=... ./atlas --non-interactive
  153. ```
  154. Re-running `./atlas` on an existing stack is idempotent.
  155. ### Option 2 — Manual Docker
  156. Full control, step-by-step.
  157. ```bash
  158. cp .env.example .env
  159. ```
  160. Open `.env` and fill in the two required keys:
  161. ```env
  162. GOOGLE_API_KEY=your_gemini_key
  163. JINA_API_KEY=your_jina_key
  164. ```
  165. Generate two required secrets and paste them into `.env`:
  166. ```bash
  167. # CREDENTIAL_MASTER_KEY — AES-256-GCM key for stored platform credentials (64 hex chars)
  168. python -c "import secrets; print(secrets.token_hex(32))"
  169. # WEAVIATE_API_KEY — auth between backend and Weaviate (required by docker-compose)
  170. python -c "import secrets; print(secrets.token_hex(16))"
  171. ```
  172. Launch:
  173. ```bash
  174. docker compose up -d --build
  175. ```
  176. Open **[http://localhost:3000](http://localhost:3000)**.
  177. **Services started:**
  178. | Service | Port | Description |
  179. |---|---|---|
  180. | Web (nginx) | `:3000` | React dashboard |
  181. | Backend | `:8000` | FastAPI + ADK agents |
  182. | Bot | `:3001` | Platform bridge (Slack / Discord / Teams) |
  183. | Weaviate | `:8080` | Semantic memory |
  184. | Neo4j | `:7474` / `:7687` | Graph memory |
  185. | MongoDB | `:27017` | State + wiki cache |
  186. | Redis | `:6380` | Sessions (internal `:6379`) |
  187. First run takes 2–3 minutes while images build and databases initialize. Subsequent runs start in seconds.
  188. ### Option 3 — Local development
  189. Databases in Docker, app services native for hot-reload.
  190. **Prerequisites:** Python 3.12+ with [uv](https://docs.astral.sh/uv/), Node.js 20+
  191. ```bash
  192. cp .env.example .env
  193. # Fill in GOOGLE_API_KEY, JINA_API_KEY, CREDENTIAL_MASTER_KEY, WEAVIATE_API_KEY (same as Option 2)
  194. # Start just the databases
  195. docker compose up -d weaviate neo4j mongodb redis
  196. # Backend (terminal 1)
  197. uv sync
  198. uv run uvicorn beever_atlas.server.app:app --reload --port 8000
  199. # Bot (terminal 2)
  200. cd bot && npm install && npm run dev
  201. # Web (terminal 3) — Vite dev server with HMR
  202. cd web && npm install && npm run dev
  203. ```
  204. Open **[http://localhost:5173](http://localhost:5173)** (the Vite dev port — **not** `:3000`).
  205. The Vite dev server proxies `/api/*` to `http://localhost:8000` (configured via `VITE_API_URL`).
  206. ### Before going to production
  207. `.env.example` defaults are tuned for local testing. Before any real deploy, rotate the secrets that ship with placeholder values and flip the environment flag:
  208. | What to change | Why | How |
  209. |---|---|---|
  210. | `BEEVER_API_KEYS`, `BEEVER_ADMIN_TOKEN` | Ship as `dev-key-change-me` / `dev-admin-change-me` — public placeholders | `python -c "import secrets; print(secrets.token_hex(24))"` per token |
  211. | `BRIDGE_API_KEY` | Shared secret between backend and bot; blank by default, required outside local dev | Same `secrets.token_hex(24)` |
  212. | `VITE_BEEVER_API_KEY`, `VITE_BEEVER_ADMIN_TOKEN` | Vite bakes these into the web bundle at build time — must mirror the rotated backend values above | Copy the rotated `BEEVER_API_KEYS` / `BEEVER_ADMIN_TOKEN` values |
  213. | `NEO4J_PASSWORD` + password half of `NEO4J_AUTH` | Dev password is public in this repo | Pick a strong password; both values must match |
  214. | `BEEVER_ENV=production` | Enables fail-fast startup that rejects every dev default above | Flip the value in `.env` |
  215. Option 1 (`./atlas`) handles all of this through the **"Rotate auth tokens"** prompt in step 4 of the checklist — answer **Y** and the installer generates random tokens and mirrors the VITE_* values for you. If you used Option 2 or 3, you can re-run `./atlas` on the existing `.env`, skip every other prompt with Enter, and only accept the rotation prompt.
  216. ### 5. Open the dashboard
  217. Navigate to the URL for your chosen option:
  218. - **Options 1 & 2** → **[http://localhost:3000](http://localhost:3000)**
  219. - **Option 3** → **[http://localhost:5173](http://localhost:5173)**
  220. From there:
  221. - **Real mode** (default, `ADAPTER_MOCK=false`): connect a workspace in **Settings → Connections** — Slack / Discord / Teams tokens are entered through the UI, not `.env`.
  222. - **Mock mode** (`ADAPTER_MOCK=true`): uses fixture data — opt in for local UI iteration without platform credentials.
  223. ### 6. Sync a channel
  224. From the dashboard: **Connections → Add Workspace → Select channels → Sync**.
  225. Or via API (auto-extracts your bearer token from `.env`):
  226. ```bash
  227. curl -X POST http://localhost:8000/api/channels/C12345/sync \
  228. -H "Authorization: Bearer $(grep -E '^BEEVER_API_KEYS=' .env | cut -d= -f2 | cut -d, -f1)"
  229. ```
  230. Media shared in synced channels (images, PDFs, video) is persisted durably so it
  231. keeps rendering after the platform CDN link expires. It defaults to in-database
  232. storage with zero extra infra, and can use MinIO/S3 at scale. See
  233. [docs/media-persistence.md](docs/media-persistence.md) for the mechanism,
  234. `CHANNEL_MEDIA_*` configuration, the MinIO/S3 backend, and backfilling existing
  235. channels.
  236. ### MCP server (for external AI agents)
  237. Beever Atlas exposes a curated MCP (Model Context Protocol) server at `/mcp` for AI agents like Claude Code and Cursor. This allows external code assistants to query your team's knowledge base without using the dashboard.
  238. See [docs/mcp-server.md](docs/mcp-server.md) for:
  239. - **Tool catalog** — 28 tools for discovery, retrieval, wiki reading, graph traversal, and long-running operations
  240. - **Auth setup** — generating and managing `BEEVER_MCP_API_KEYS`
  241. - **Client configuration** — ready-to-use `.mcp.json` templates for Claude Code and Cursor
  242. - **Rate limits** — principal-keyed limits to prevent one agent from throttling others
  243. It also ships a **standalone stdio mode** (`python -m beever_atlas.api.mcp_server` / `beever-atlas-mcp`) that exposes the same tool catalog with no HTTP server or backing stores — handy for MCP registries (Glama.ai) and local introspection. See [docs/mcp-server.md](docs/mcp-server.md#standalone-introspection-mode-stdio).
  244. Quick example (Claude Code):
  245. ```json
  246. {
  247. "mcpServers": {
  248. "beever-atlas": {
  249. "url": "https://atlas.example.com/mcp",
  250. "transport": "streamable-http",
  251. "headers": {
  252. "Authorization": "Bearer ${BEEVER_MCP_KEY}"
  253. }
  254. }
  255. }
  256. }
  257. ```
  258. ### Common commands
  259. ```bash
  260. docker compose up -d # Start in background
  261. docker compose logs -f beever-atlas # Tail backend logs
  262. docker compose down # Stop (keeps data)
  263. docker compose down -v # Stop and DELETE all indexed data
  264. make demo # Full stack + seeded demo corpus
  265. make docker-up # Shortcut for `docker compose up -d`
  266. ```
  267. ---
  268. ## 🔒 Privacy & Telemetry
  269. Beever Atlas collects no telemetry. No usage data, error reports, or analytics are sent anywhere by default. All LLM calls go through API keys you configure in your own `.env`, and all data stays in the databases you control.
  270. ---
  271. ## 📐 API Stability
  272. All `/api/*` endpoints are **UNSTABLE** in 0.1.0. v0.2.0 will introduce a `/api/v1/*` prefix; clients pinning current paths will break. See [SECURITY.md](SECURITY.md).
  273. ---
  274. ## 💬 Community & Contact
  275. - **Discord**: [discord.gg/VshBCUUX](https://discord.gg/VshBCUUX) — get help, share what you're building, talk to the team
  276. - **X / Twitter**: [@Beever_AI](https://x.com/Beever_AI) — release notes, posts, announcements
  277. - **Website**: [beever.ai](https://beever.ai/) — about the company and other projects
  278. - **GitHub Discussions**: [github.com/Beever-AI/beever-atlas/discussions](https://github.com/Beever-AI/beever-atlas/discussions) — longer-form questions and ideas
  279. Commercial support, partnerships, or press: `tech@beever.ai`.
  280. ---
  281. ## 📜 License
  282. [Apache License 2.0](LICENSE) © 2026 Beever Atlas contributors. Third-party attributions in [NOTICE](NOTICE).
  283. Security policy: [SECURITY.md](SECURITY.md) | Community standards: [CODE_OF_CONDUCT.md](CODE_OF_CONDUCT.md)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.