{"slug": "show-hn-i-am-getting-genuine-questions", "title": "Show HN: I am getting genuine questions", "summary": "DocBrain, a self-hosted, read-only memory layer for humans and AI agents, captures undocumented knowledge from tickets, threads, incidents, and code changes with per-claim provenance, and its client source is open under MIT while the server remains closed under BSL 1.1. The project, which missed its original first-half-2026 server publication target, offers free production Docker images and Helm charts for self-hosting, and aims to solve the problem of organizational knowledge loss by ingesting historical systems on day one.", "body_md": "**The memory layer for humans and AI agents.**\n\nEvery other tool retrieves what someone remembered to write down. DocBrain captures what nobody did — from tickets, threads, incidents and code changes — cites every individual claim to its source, and says so plainly when your record has no answer. Self-hosted. Read-only. Zero data egress.\n\n[ Website](https://docbrainapi.com) •\n\n[•](https://docbrainapi.com/docs)\n\n**Docs**[Quickstart](#quickstart)•\n\n[The Problem](#the-problem)•\n\n[How It Works](#how-docbrain-works)•\n\n[Architecture](#architecture)•\n\n[Security](#security)\n\n[\n](https://youtu.be/zZ7WdjmOXHU)\n\n*real recording, no edits — watch the full 90-second quickstart ▶*\n\nProject Status: client source open, server closed.The source for everything DocBrain runs onyourside of the network boundary — the`docbrain`\n\nCLI and the IDE MCP server — is in this repo under[, MIT-licensed, built and tested in public CI — as is everything else in this repository. Audit exactly what runs in your environment and what leaves it. The server ships as free production Docker images (BSL 1.1 permits production use) with full Helm charts, complete configuration, the]`crates/`\n\n[threat model], and all the docs to self-host in production. The server source stays closed for now — we originally targeted the first half of 2026 to publish it, we missed that date, and we won't post a new date until we're certain we can hit it. If a closed server is a dealbreaker for you, that's a rational position and we respect it:[how DocBrain earns trust]. Contributions: code PRs for the client crates, plus documentation, configuration, and bug reports. When the server source publishes, it will be published under the Business Source License 1.1.\n\nEvery organization runs on knowledge that never gets written down: the decision from a meeting, the fix someone found at 2am, the workaround only one person knows. It lives in PRs, chat threads, tickets, and people's heads. When that person changes teams or quits, years of context walk out the door with them.\n\nTools that \"index your docs and add a chatbot\" solve the wrong half of the problem: they retrieve your stale, incomplete wiki slightly faster. The knowledge that actually runs your organization was never captured in the first place. And it's getting worse now that AI produces code, changes, and fluent documentation faster than any human can absorb — your agents read those docs too. More documentation is easy. Documentation your organization can trust is the scarce thing.\n\nDocBrain captures knowledge **at the source**, the moment it's created:\n\n```\n  Someone merges a change      ──→  decisions, caveats, procedures extracted\n  A team works through chat    ──→  the answer, distilled from the thread\n  A deploy goes out            ──→  what changed and why\n  On-call resolves an incident ──→  the fix and the root cause\n  Any other system you run     ──→  ingested via the Connector SDK\n```\n\n**On day one it also reads backwards.** Point it at systems you have been using for years — archived Slack channels, closed tickets, merged pull requests, the wiki nobody has opened since 2022 — and it ingests them in place, read-only. Your first answer can come from a thread nobody remembers writing. Nothing is migrated and nobody has to refile anything.\n\nCaptured fragments are confidence-scored, connected into one memory, and composed into documentation with **per-claim provenance**. Drafts route through human review before anything publishes. Then DocBrain keeps the result honest: freshness tracking, contradiction detection, and staleness alerts as reality changes. Ask a question, get a cited answer — or an honest \"I don't know\" instead of a guess.\n\n```\ngit clone https://github.com/docbrain-ai/docbrain.git && cd docbrain\n./scripts/setup.sh    # interactive wizard: picks provider, sets keys, starts services\n```\n\nOr manually:\n\n```\ncp .env.example .env   # set LLM_PROVIDER and API keys\ndocker compose up -d\n# Get the auto-generated admin API key\ndocker compose exec server cat /app/admin-bootstrap-key.txt\n\n# Open the web dashboard\nopen http://localhost:3001\n\n# Or ask a question via API\ncurl -H \"Authorization: Bearer <key>\" \\\n     -H \"Content-Type: application/json\" \\\n     -d '{\"question\":\"How do I deploy to production?\"}' \\\n     http://localhost:3001/api/v1/ask\n```\n\nFull setup guide: [docs/quickstart.md](/docbrain-ai/docbrain/blob/main/docs/quickstart.md)\n\nIf your team uses Claude Code or Cursor, the [ docbrain-mcp](/docbrain-ai/docbrain/blob/main/crates/docbrain-mcp) server already gives your agent capture tools — most teams just never tell the agent to use them. Three lines in your\n\n`CLAUDE.md`\n\nturn every debugging session into documentation:\n\n```\nWhen we resolve an error or discover non-obvious behavior, call\ndocbrain_suggest_capture for the files involved. If a gap exists, draft a\n3–5 line capture and ask me to approve it before calling docbrain_capture.\n```\n\nYour agent fixes something, checks whether the org already knows it, and — with your approval — files what's missing into the review queue. The knowledge gets captured at the only moment it's free: seconds after the fix. Full guide, including Cursor setup and the privacy model: [docs/agents.md](/docbrain-ai/docbrain/blob/main/docs/agents.md)\n\n**Capture from 13 built-in sources**— Confluence, Slack, Teams, GitHub, GitLab, Jira, PagerDuty, Linear, OpsGenie, Rootly, Zendesk, Intercom, and local files, plus a language-agnostic[Connector SDK](/docbrain-ai/docbrain/blob/main/docs/connectors.md)for anything else.[Ingestion guide →](/docbrain-ai/docbrain/blob/main/docs/ingestion.md)**Ask, with citations**— hybrid vector + keyword search, confidence-scored answers; low confidence asks clarifying questions instead of guessing.[API →](/docbrain-ai/docbrain/blob/main/docs/api-reference.md)— on-demand docs grounded in your own runbooks, incidents, threads, and PRs, with per-claim provenance and honest`docbrain generate`\n\n`needs_input`\n\nfor what the knowledge can't answer.[Generate guide →](/docbrain-ai/docbrain/blob/main/docs/generate.md)**Quality gates on every doc**— structural, style (your style guide, enforced automatically), and semantic scoring; nothing unscored enters the system.[Style policy →](/docbrain-ai/docbrain/blob/main/docs/style-policy.md)**Review workflows and ownership**— multi-stage approvals, space owners, SLAs, and governance dashboards, so documentation has accountability.[Governance →](/docbrain-ai/docbrain/blob/main/docs/governance.md)·[Reviews →](/docbrain-ai/docbrain/blob/main/docs/reviews.md)**Autopilot**— clusters unanswered questions into gaps, drafts grounded fixes, and routes them to human review. Nothing publishes without oversight.[Autopilot →](/docbrain-ai/docbrain/blob/main/docs/autopilot.md)**Live intelligence**— reads connected systems*at the moment you ask*, then reconciles that with indexed history in one cited answer. Write-capable tools are dropped at discovery, so it never holds the capability to change anything.[MCP tools →](/docbrain-ai/docbrain/blob/main/docs/mcp-tools.md)**Freshness and contradiction detection**— stale docs flagged, conflicting docs surfaced, cascade staleness traced across dependent docs.[Knowledge intelligence →](/docbrain-ai/docbrain/blob/main/docs/knowledge-intelligence.md)**Learning pipeline (optional, off by default)**— feedback on answers can fine-tune the embedding model on*your*corpus, so retrieval learns what \"similar\" means in your codebase. Versions are gated on quality before promotion and roll back automatically if they regress. Runs entirely on your infrastructure.[Learning →](/docbrain-ai/docbrain/blob/main/docs/learning.md)**Predictive intelligence**— onboarding-gap detection from what new joiners ask in their first 30 days, seasonal query forecasting, and code-change-triggered doc review.[Knowledge intelligence →](/docbrain-ai/docbrain/blob/main/docs/knowledge-intelligence.md)**Source-system ACL mirroring**— Confluence restrictions, Slack channel membership, and repo visibility enforced at query time.[Access control →](/docbrain-ai/docbrain/blob/main/docs/access-control.md)**RBAC, SSO, audit logging**— 4-tier roles, GitHub/GitLab/OIDC SSO, per-space isolation.[RBAC →](/docbrain-ai/docbrain/blob/main/docs/rbac.md)** Everywhere your team works**— web dashboard, Slack commands, CLI, CI hooks, and MCP tools for Claude Code, Cursor, and any MCP-compatible editor.[Slack →](/docbrain-ai/docbrain/blob/main/docs/slack.md)\n\n```\ngraph TB\n    subgraph \"Capture Layer\"\n        CI[\"CI/CD Pipelines\"]\n        IDE[\"IDE (MCP)\"]\n        SLACK[\"Slack / Teams\"]\n        WEB[\"Web UI\"]\n        CLI[\"CLI\"]\n        API_EXT[\"External APIs\"]\n    end\n\n    subgraph \"DocBrain Server (Rust / Axum)\"\n        FRAG[\"Fragment Router\"]\n        QUAL[\"Quality Pipeline\"]\n        CLUST[\"Clustering Engine\"]\n        COMP[\"Composition Engine\"]\n        REV[\"Review Workflows\"]\n        RAG[\"RAG Pipeline\"]\n        AUTO[\"Autopilot\"]\n        GOV[\"Governance\"]\n        EVT[\"Event Bus + Webhooks\"]\n    end\n\n    subgraph \"Storage\"\n        PG[\"PostgreSQL\"]\n        OS[\"OpenSearch<br/><i>vector + keyword</i>\"]\n        RD[\"Redis\"]\n    end\n\n    CI & IDE & SLACK & WEB & CLI & API_EXT --> FRAG\n    FRAG --> QUAL --> CLUST --> COMP --> REV\n    WEB & CLI & SLACK --> RAG\n    RAG & AUTO & GOV --> PG & OS\n    EVT --> PG\n```\n\nRust server, PostgreSQL, OpenSearch, Redis. Full design: [docs/architecture.md](/docbrain-ai/docbrain/blob/main/docs/architecture.md)\n\nDocBrain runs entirely in your infrastructure, **read-only** against your sources. You choose where the model runs: fully local via Ollama (zero egress), your own cloud account (Bedrock, Azure, Vertex — your KMS, your audit trail), or a provider API. Documents, embeddings, and indexes never leave your network; only the query and the relevant chunks reach the LLM you chose.\n\nAPI keys are Argon2-hashed, every endpoint enforces RBAC, rate limits are per-key, and admin actions are audit-logged. The client code you install is open source in [ crates/](/docbrain-ai/docbrain/blob/main/crates). The full threat model — 11 analyzed attack vectors and an operator checklist — is published:\n\n[THREAT_MODEL.md](/docbrain-ai/docbrain/blob/main/THREAT_MODEL.md)\n\n**LLM providers (14):** Anthropic, OpenAI, AWS Bedrock, Ollama, Google Gemini, Vertex AI, Azure OpenAI, DeepSeek, Groq, Mistral, xAI, OpenRouter, Together AI, Cohere. [Provider setup →](/docbrain-ai/docbrain/blob/main/docs/providers.md)\n\n```\n# Docker Compose — everything behind a single origin at localhost:3001\ndocker compose up -d\n\n# Kubernetes\nhelm install docbrain ./helm/docbrain \\\n  --set llm.provider=anthropic \\\n  --set llm.anthropicApiKey=sk-ant-...\n```\n\n[Kubernetes guide →](/docbrain-ai/docbrain/blob/main/docs/kubernetes.md) · [Configuration →](/docbrain-ai/docbrain/blob/main/docs/configuration.md)\n\n|\n\n[Configuration](/docbrain-ai/docbrain/blob/main/docs/configuration.md)[Provider Setup](/docbrain-ai/docbrain/blob/main/docs/providers.md)[Architecture](/docbrain-ai/docbrain/blob/main/docs/architecture.md)[Ingestion Guide](/docbrain-ai/docbrain/blob/main/docs/ingestion.md)[External Connectors](/docbrain-ai/docbrain/blob/main/docs/connectors.md)[Governance](/docbrain-ai/docbrain/blob/main/docs/governance.md)[Review Workflows](/docbrain-ai/docbrain/blob/main/docs/reviews.md)[Knowledge Intelligence](/docbrain-ai/docbrain/blob/main/docs/knowledge-intelligence.md)[Autopilot](/docbrain-ai/docbrain/blob/main/docs/autopilot.md)[Generate](/docbrain-ai/docbrain/blob/main/docs/generate.md)[Coding Agents](/docbrain-ai/docbrain/blob/main/docs/agents.md)[API Reference](/docbrain-ai/docbrain/blob/main/docs/api-reference.md)[RBAC](/docbrain-ai/docbrain/blob/main/docs/rbac.md)[Slack Integration](/docbrain-ai/docbrain/blob/main/docs/slack.md)[Kubernetes](/docbrain-ai/docbrain/blob/main/docs/kubernetes.md)**▶ The quickstart, recorded unedited** — install → ingest → cited answer →\n\n`generate`\n\nturning raw on-call notes into a runbook that cites its sources. 90 seconds, shipped images, 100% local models, nothing staged.\n|\n\n[Deep Dive Podcast](https://youtu.be/GN4SC6L8YmI), 20-min deep dive[MCP Preview](https://youtu.be/9mZLoQnGLl8), 30-sec IDE demo[Full Proof Demo](https://youtu.be/yqj5BCVOLHw), Downvote → Gap → DraftWe opened Atlassian's own AI assistant, asked it to compare itself with DocBrain, and published the full response unedited — including where it wins.\n\n\"DocBrain's 'capture knowledge that was never written down' is solving a problem I fundamentally can't.\"— Rovo, answering a direct comparison prompt\n\nWhere Rovo wins: native Atlassian integration, zero setup for existing Atlassian Cloud teams, broad work execution, and 3M+ users. [Full transcript →](https://docbrainapi.com/rovo-comparison.html)\n\nA project built on refusing to overclaim shouldn't overclaim about itself.\n\n**The server is closed source.** The client that runs inside your network is MIT and auditable, in. The server isn't published. We targeted the first half of 2026, missed it, and won't name a new date until we're certain of it.`crates/`\n\n**We publish no measured accuracy benchmark.** Grounding is measured internally and gates every model promotion, but no number goes in this README until it's measured across real customer corpora and we can publish the methodology with it.**Self-hosting isn't unique to us.** Open-source alternatives exist and several are genuinely good at retrieval — some more permissively licensed than we are. What they don't do is capture what was never written down. Judge us on that.**Our evidence base is strongest for software teams.** The mechanism generalises across support, operations and the rest of the business; the published statistics haven't been measured everywhere, and we won't imply otherwise.\n\n**GitHub Issues:**[Bug reports and feature requests](https://github.com/docbrain-ai/docbrain/issues)** GitHub Discussions:**[Questions and community conversation](https://github.com/docbrain-ai/docbrain/discussions)** Email:**[hello@docbrainapi.com](mailto:hello@docbrainapi.com)\n\nWe welcome contributions. The client tooling source ([ crates/](/docbrain-ai/docbrain/blob/main/crates)) accepts code PRs; server-side contributions land best as documentation, configuration, and bug reports. See\n\n[Contributing Guide](/docbrain-ai/docbrain/blob/main/CONTRIBUTING.md).\n\nTo report a security vulnerability, see [SECURITY.md](/docbrain-ai/docbrain/blob/main/SECURITY.md). Do **not** file a public issue.\n\nThis repository is [MIT licensed](/docbrain-ai/docbrain/blob/main/LICENSE) — the `docbrain`\n\nCLI, the MCP server, the Helm\ncharts, configuration, examples and documentation.\n\nThe **DocBrain server binaries and container images** are distributed under the\n[Business Source License 1.1](/docbrain-ai/docbrain/blob/main/LICENSE-SERVER). Production use is permitted, except offering\nDocBrain as a hosted service. For alternative licensing:\n[licensing@docbrainapi.com](mailto:licensing@docbrainapi.com).\n\n[Contributor Covenant Code of Conduct](/docbrain-ai/docbrain/blob/main/CODE_OF_CONDUCT.md).\nReport concerns to [hello@docbrainapi.com](mailto:hello@docbrainapi.com).", "url": "https://wpnews.pro/news/show-hn-i-am-getting-genuine-questions", "canonical_source": "https://github.com/docbrain-ai/docbrain", "published_at": "2026-08-19 16:16:33+00:00", "updated_at": "2026-08-19 16:44:03.272702+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-products", "ai-infrastructure", "ai-agents", "developer-tools"], "entities": ["DocBrain", "DocBrain API", "MIT", "BSL 1.1", "Docker", "Helm", "GitHub"], "alternates": {"html": "https://wpnews.pro/news/show-hn-i-am-getting-genuine-questions", "markdown": "https://wpnews.pro/news/show-hn-i-am-getting-genuine-questions.md", "text": "https://wpnews.pro/news/show-hn-i-am-getting-genuine-questions.txt", "jsonld": "https://wpnews.pro/news/show-hn-i-am-getting-genuine-questions.jsonld"}}