[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_121":122},"routes/modules.$creator.$moduleId",{"_14":15,"_16":17,"_18":19,"_20":21},"routes/modules.$creator.$moduleId.endpoints",{"_11":12,"_13":-5},"error","Module not found","tools","module",{"_20":22,"_23":24,"_25":26,"_27":28,"_29":30,"_31":32,"_33":34,"_35":26,"_36":37,"_38":-5,"_39":-5,"_40":-5,"_41":42,"_43":44,"_45":46,"_47":48,"_49":48,"_50":51,"_52":53,"_54":55,"_56":-5,"_57":48,"_58":59,"_60":61,"_62":-5,"_63":-5,"_64":-5,"_65":-5,"_66":67,"_68":69,"_70":71,"_72":73,"_74":75,"_76":-5,"_77":-5,"_78":-5,"_79":80,"_81":-5,"_82":44,"_83":84,"_85":86,"_87":88,"_89":90,"_91":92,"_93":94},"apiUrl","https://odel.app/api","creator","xiaojiou176-open","moduleId","provenote-mcp","@xiaojiou176-open/provenote-mcp","displayName","Provenote MCP","tagline","First-party Provenote MCP server for drafts, research threads, auditable runs, and knowledge search.","iconKey","https://avatars.githubusercontent.com/u/270514762?v=4","category","local","status","published","visibility","public","metaDescription","deploymentType","external","workerName","bindingName","mcpConfig","pricingModel","free","priceCredits",0,"creatorUserId","system:external-import","createdAt",1790466471000,"updatedAt","sourceRegistry","official-mcp-registry","sourceUrl","https://registry.modelcontextprotocol.io/v0/servers?search=io.github.xiaojiou176-open%2Fprovenote-mcp","sourceVerified",true,"sourceMissingSince","importedAt","upstreamStars",1,"upstreamUpdatedAt",1787623866000,"detectedTransport","detectedAuthType","probeStatus","lastProbedAt","githubRepoUrl","https://github.com/xiaojiou176-open/notebooklab","githubRepoFullName","xiaojiou176-open/notebooklab","releaseId","ext-xiaojiou176-open-provenote-mcp","version","1.8.5","descriptionMarkdown","# Notebooklab\n\n

\n Teach agents and operators to turn messy long context into structured outcomes they can carry into notes, research threads, drafts, and inspectable results.\n

\n $\"GitHub$ \n $\"License\"$ \n $\"Quick$ \n $\"Public$ \n

\n Quick Result Path\n ·\n Long Context\n ·\n Public Proof\n ·\n Project Status\n ·\n Docs\n

\n Second ring:\n MCP & Integrations\n ·\n Companion Host Bundles\n ·\n Distribution\n ·\n FAQ\n ·\n Discussions\n

\n Star Notebooklab if you want a source-heavy AI workbench that stays inspectable after the chat scrollback is gone.\n

\n\n![Notebooklab hero showing grounded sources, auditable markdown, and reusable outputs in one workbench](https://raw.githubusercontent.com/xiaojiou176-open/notebooklab/main/docs/assets/hero/notebooklab-hero.png)\n\n![Notebooklab quick-result overview showing the shortest repo-documented path from source import to auditable markdown download](https://raw.githubusercontent.com/xiaojiou176-open/notebooklab/main/docs/assets/demo/notebooklab-quick-result-overview.png)\n\nThis illustrated overview is a repo-authored summary of the shortest documented path. It is intentionally not presented as a live product recording.\n\n> Canonical product path:\n> `messy long context -> structured insight -> note / research thread / draft -> inspectable outcome`\n\nThat is the first door. MCP, starter bundles, distribution pages, and promotion assets are valuable second-ring surfaces, but they should not outrank the product path.\n\nAgent-facing truth comes first: Notebooklab teaches an agent to read messy\ncontext, structure it, move it into note / research-thread / draft lanes, and\nonly then carry that outcome workflow forward through the first-party\n`notebooklab-mcp` server. Public skills, host bundles, and registry packs are\ncompanion surfaces around that workbench, not the product root.\n\n## At A Glance\n\nIf you only want the shortest truthful filter before reading deeper, use this\ntable first:\n\n| What you need to know | Current answer |\n| --- | --- |\n| Product thesis | turn messy long context into structured insight and inspectable outcomes you can carry into notes, research threads, and drafts |\n| Fastest result path | import one source -> run Auditable Markdown -> download one inspectable result |\n| First proof | the quick-result overview plus the public proof page |\n| Second ring only | MCP, host bundles, public skills, and distribution surfaces |\n| What it must never be reduced to | a hosted one-minute trial or a generic chat wrapper |\n\n## Start Here In 10 Seconds\n\nIf you only want the fastest honest map, use this:\n\n| Question | Open this first | Why |\n| --- | --- | --- |\n| \"Can this help with messy long context?\" | [Long Context](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/use-cases/long-context-to-structured-notes.md) | This is the product center, not a side use case. |\n| \"Can I get one real result quickly?\" | [Quick Result Path](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/quickstart.md) | This is the shortest repo-documented local proof loop. |\n| \"Is this real or just copywriting?\" | [Public Proof](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/proof.md) | This is the evidence layer. |\n| \"What is still intentionally unclaimed?\" | [Project Status](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/project-status.md) | This is the boundary page. |\n\nJudge the workbench before you judge the side doors. MCP pages, starter bundles, distribution packs, and promotion assets matter, but they are second-layer surfaces around the main product path.\n\n## Why Notebooklab Exists\n\nMost AI note tools make it easy to generate words and hard to verify where those words came from.\n\nThat gets even worse when the raw material is long and messy: a huge chat log, a copied forum thread, a meeting recap, or a web page pile you do not want to flatten into one more throwaway summary.\n\nNotebooklab is built for the opposite direction:\n\n- collect notes, documents, audio, and web content in one place\n- structure long context before it disappears into another chat turn\n- search and ask across that material without losing the source trail\n- turn high-value source work into auditable markdown with integrity counters\n- keep the whole workflow repository-documented and reproducible\n\nThink of it like moving from a loose pile of research tabs to a workbench with labeled drawers, measuring tools, and a clean export lane.\n\nIf your problem starts as \"I have too much messy context,\" the strongest current repo-backed answer is not another empty chat box. It is the path from long context to structured notes and then into reusable outcome objects.\n\n## Fastest Result Path\n\nIf you want one visible outcome before learning the whole workbench, keep the\norder this short:\n\n1. import or open one source\n2. open the source detail surface\n3. run **Auditable Markdown**\n4. download the inspectable result\n\nThat is the shortest repo-documented result path today.\nMCP, starter bundles, podcasts, and distribution surfaces stay second ring\nuntil this outcome path already makes sense.\n\n## If You Start With Messy Long Context\n\nThe strongest current first-entry path is:\n\n```text\nImport the source\n-> Run Chat Knowledgeization\n-> Inspect the structured insight\n-> Continue it as a note or notebook research thread\n-> Keep draft work inside the notebook lane\n```\n\nIn plain language: first turn the pile into labeled folders, then decide whether it belongs in your notes, your active research lane, or your notebook draft workflow.\n\n## What You Can Do In One Workbench\n\n- **Collect grounded context**: bring in text, files, audio, and web sources instead of starting every session from a blank chat box.\n- **Structure long context before chatting more**: use built-in transformations such as **Chat Knowledgeization** when the material starts as long chat logs, forum threads, meeting notes, or copied web discussions.\n- **Keep structured results moving**: after long-context structuring, continue the result into a note, a notebook research thread, or the notebook draft lane instead of leaving the insight stranded.\n- **Search and think with structure**: move between notebooks, ask/search flows, transformations, and model settings without hopping across disconnected tools.\n- **Create outputs with receipts**: use the auditable markdown lane when you need stronger traceability than ordinary chat alone, or create notebook-level drafts plus export bundles when you need a reusable outcome object you can hand off.\n- **Hand outcomes off cleanly**: download draft markdown when you only need the text, or export a richer draft bundle when you need metrics, claim/section review data, PID summaries, and source manifest context together.\n- **Go beyond text-only workflows**: generate podcast-ready outputs and reusable transformations from the same source base.\n- **Bring it into coding agents through MCP**: expose notebooks, sources, drafts, research threads, and auditable runs to Claude Code, OpenAI Codex, Cursor, and other MCP-capable hosts through a first-party MCP server.\n- **Run the same outcome lanes from a terminal/operator surface**: use the first-party `notebooklab` CLI when you want notebook outcome inspection, auditable markdown, or research-thread-to-draft handoffs without treating MCP host setup as the only operator path.\n\nIf you only remember one first-entry rule, make it this:\n\n```text\nmessy long context\n-> structured insight\n-> note / seeded research lane / notebook research thread\n-> draft-adjacent notebook work\n```\n\nThat is the product center. MCP host pages come after that path, not before it.\n\n## Ecosystem Boundary\n\nThink of these surfaces like rooms around the main workshop:\n\n- host guides are the side doors\n- the CLI is the tool cart\n- the main workbench is still the path from messy context to reusable outcomes\n\n| Surface | What it is today | What it is not |\n| --- | --- | --- |\n| Claude Code / Codex / Cursor / OpenCode pages | repo-backed compatibility guides through the first-party MCP server, with public-ready starter bundles under `examples/hosts/` | official partnership, bundled integration, plugin, or marketplace listing |\n| `notebooklab` CLI | a first-party local operator surface for outcome inspection and `research_thread -> draft -> verify/download` workflows | a separate distributed product line or a renamed MCP server |\n| public skills surface | tracked public-ready skill packets now exist under `public-skills/` for host-specific submission flows | a public skills catalog, live marketplace listing, or host-specific skills program endorsement |\n| OpenClaw | a live ClawHub skill listing now exists at `https://clawhub.ai/xiaojiou176/notebooklab-mcp-outcome-workflows`, and public-ready OpenClaw-compatible bundles still live under `examples/hosts/openclaw` | official OpenClaw partnership, vendor endorsement, or every other marketplace surface being live |\n| Official MCP Registry | a live websiteUrl-backed `notebooklab-mcp` entry already points to the repo-owned MCP docs/install surface | a package-backed public artifact, official host marketplace listing, or vendor endorsement |\n| release / listing / domain / trademark / partnership | external decision and publication work | completed repo-side truth |\n\nIf you want the shortest public page that keeps those boundaries honest, start with [docs/project-status.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/project-status.md).\n\nIf you want repo-owned install artifacts instead of prose-only setup pages, start with [examples/hosts/README.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/examples/hosts/README.md). That index points to the public-ready Claude Code, Codex, Cursor, and OpenCode starter bundles plus the OpenClaw-compatible bundles and the ClawHub submission pack.\n\nIf you want the standalone host-facing skill folders used for OpenHands/extensions\nor ClawHub-style submissions, start with [public-skills/README.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/public-skills/README.md).\n\nIf you want the full claim ladder instead of inferring it from scattered host pages, use [docs/distribution.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/distribution.md).\n\n## Second-Layer Surfaces\n\nOnce the product path is clear, use one second-ring page instead of reading every host, registry, and distribution page in order.\n\n| If you need... | Open this first | Why it stays second-ring |\n| --- | --- | --- |\n| a host-compatible MCP path | [docs/mcp.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/mcp.md) | coding-agent carry-forward matters, but it is not the product center |\n| checked-in bundle artifacts | [examples/hosts/README.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/examples/hosts/README.md) and [examples/hosts/packet-index.json](https://github.com/xiaojiou176-open/notebooklab/blob/main/examples/hosts/packet-index.json) | bundle installs are packaging truth, not the first doorway |\n| public claim ladder and listing boundary | [docs/distribution.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/distribution.md) | it is a truth ledger, not the main product story |\n| what is still intentionally unclaimed | [docs/project-status.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/project-status.md) | this is the boundary page, not the first impression |\n| terminal/operator workflows | [docs/runbooks/operator-cli.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/runbooks/operator-cli.md) | the CLI is a carry-forward surface after the workbench story is clear |\n\nThe ordering is intentional: first the workbench, then the proof lane, then the carry-forward surfaces.\n\n## Product Truth Boundary\n\nOrdinary chat and ask flows are a fast assistant surface for exploration and iteration.\n\nWhen you need stronger traceability, the stricter product truth lane lives in auditable markdown and auditable-runs rather than ordinary chat alone.\n\n## Quick Result Path\n\nThis path optimizes for the first visible result, not full system mastery.\n\nThink of it as the fastest repo-documented local proof loop, not a one-minute hosted trial.\n\nIn plain language: this is a **5 to 10 minute local proof loop**, not a hosted one-minute trial.\n\n1. Copy the local environment template and add the two fast-path values.\n\n ```bash\n cp .env.example .env\n ```\n\n Fast-path values:\n\n ```bash\n OPEN_NOTEBOOK_ENCRYPTION_KEY=change-me-to-a-secret-string\n GEMINI_API_KEY=your-google-ai-studio-key\n ```\n\n2. Start the default local stack.\n\n ```bash\n docker compose -f ops/compose/docker-compose.yml up -d --build\n ```\n\n On this Docker fast path, Compose injects the SurrealDB connection defaults and a known-good fast-path Gemini model for you. That is why the first visible result path only asks you to set the encryption key and Gemini key up front.\n\n3. Open `http://localhost:8502`, let the app route you into the `/sources` workbench, then create or import a source and move into the source detail view.\n\n4. Run **Auditable Markdown** to generate a downloadable markdown report with integrity counters.\n5. Open a notebook and create a **Draft** if you want a reusable notebook-level outcome instead of a single-source artifact.\n\nIf you want the full walkthrough, go straight to [docs/quickstart.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/quickstart.md).\nIf you want a zero-setup hosted product, this repository is not promising that experience today.\n\nIf you prefer a terminal-first operator path once the local stack is up, use the narrow first-party operator runbook in [docs/runbooks/operator-cli.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/runbooks/operator-cli.md). That surface is intentionally outcome-first and does not replace the workbench UI or the MCP compatibility pages.\n\nIf you want the ecosystem boundary before you read host pages, use [docs/project-status.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/project-status.md).\n\n## Runtime Cleanup Operator Path\n\nNotebooklab treats disk cleanup as a governed operator path, not an ad-hoc \"delete big folders\" exercise.\n\n- Repo-local dependency root: `apps/web/node_modules`\n - This path is a repo-owned rebuildable dependency tree.\n - It is **not** a shared machine cache.\n - Clearing it is allowed only through the repo-local cleanup contract, and the rebuild command is `cd apps/web && npm ci`.\n- Repo-related machine cache examples:\n - `${HOME}/.cache/notebooklab/...`\n - `${HOME}/.cache/notebooklab/python/uv-cache`\n - `${HOME}/.cache/notebooklab/playwright/ms-playwright`\n - `${HOME}/.cache/notebooklab/ci-host/npm-cache`\n - these are the only repo-specific machine-level download caches that should remain outside the checkout.\n - entrypoint wrappers now auto-run the repo-owned machine-cache cleanup lane so stale cache roots and oversized download caches do not silently accumulate forever.\n- Repo-local CI/runtime residue:\n - `.runtime-cache/venv/default`\n - `.runtime-cache/ci-host/...`\n - these are repo-owned runtime or bootstrap surfaces and must stay inside the checkout rather than leaking into machine-wide cache roots.\n- Shared tool caches remain advisory-only:\n - `${HOME}/.cache/uv`\n - `${HOME}/.npm`\n - `${HOME}/Library/Caches/ms-playwright`\n - `${HOME}/Library/Containers/com.docker.docker`\n - `${HOME}/.docker`\n - these are related to the repo, but they are not repo-exclusive and must not be auto-cleared by Notebooklab entrypoints.\n- Docker runtime surfaces:\n - local `open-notebook-ci:*` images\n - buildx builders / BuildKit containers\n - Docker build cache reported by `docker system df -v`\n - these belong to the Docker operator path, not the repo-local `cleanup_runtime_cache.sh` lane\n\nUse the explicit operator flow:\n\n```bash\nmake cleanup-operator-audit\nmake cleanup-operator-apply\n```\n\nThat operator flow is intentionally split:\n\n- `cleanup_runtime_cache.sh` handles repo-local runtime/cache surfaces\n- `cleanup_machine_cache.sh` handles the remaining repo-related download caches under `~/.cache/notebooklab`\n- entrypoint wrappers (`run_uv_managed.sh`, `run-playwright-managed.sh`, and `run_in_consistent_container.sh`) invoke that machine-cache lane in apply mode so repo-specific external caches obey TTL/cap/root-cap policy by default\n- `docker-buildx-clean` / `docker-runtime-audit` cover repo-related Docker builder and image surfaces\n\n## Real Result, Not Just Positioning\n\nThis is the fastest concrete proof path in the current repo surface:\n\n| Evidence lane | What goes in | What comes out | What you can inspect |\n| --- | --- | --- | --- |\n| Auditable Markdown | one imported source from text, file, audio, or web content | a downloadable markdown report | coverage-oriented integrity counters plus a direct markdown download path |\n| Notebook Drafts | multiple notebook-linked sources | a downloadable notebook-level markdown draft plus export bundle | draft versions, section/claim inspection, bundle export, and notebook outcome state |\n\nThe current UI and API evidence behind that lane is already public:\n\n- the workbench exposes an **Auditable Markdown** panel with **Run** and **Download Markdown** actions\n- the panel surfaces **Coverage Rate**, **Missing**, **Duplicate**, **Uncited Claims**, **Unknown PID**, and **Unclassified**\n- the API publishes auditable-run routes behind the same workflow\n\nStart from [docs/proof.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/proof.md) if you want the file-and-route evidence behind those claims.\n\nIf you want a fixed, reproducible local proof loop, use the [public proof pack](https://github.com/xiaojiou176-open/notebooklab/blob/main/examples/public-proof/auditable-markdown/README.md).\n\n### Sanitized result shape\n\n```text\n# Auditable Markdown Report\n\n## Source\n- title: Example imported source\n- lane: auditable markdown\n\n## Integrity Counters\n- coverage_rate: 0.98\n- missing_count: 0\n- duplicate_count: 0\n- uncited_claims_count: 0\n\n## Output\n- downloadable markdown report\n- inspectable sections and claims\n- stronger traceability than a plain chat reply\n```\n\nThis is a sanitized result shape, not a copied user artifact. The concrete product evidence for the lane lives in the current UI panel and auditable-run API surface.\n\n## Why People Star Notebooklab Now\n\n- **You want a repo-documented workbench, not another single chat pane**. Notebooklab already combines sources, notebooks, search, transformations, podcasts, and an auditable output lane in one repo surface.\n- **You care about source-grounded writing that leaves receipts**. The auditable markdown lane is a stronger hook than generic \"AI notes\" because it points people toward inspectable output, not only generated text.\n- **You want a repo worth revisiting, not a one-time curiosity**. Releases, changelog, proof surfaces, and docs make it easier to track visible progress over time.\n- **You want a fork that acts like a real public product surface**. Provenance still matters, but the current support, review, and stewardship boundary is repository-local and documented.\n\n## Why Watch Notebooklab Over Time\n\n- **The repo exposes proof surfaces, not only roadmap language**. README, proof, changelog, release notes, and issue/discussion routing all work together as public evidence layers.\n- **The product shape is already broader than one feature spike**. The current surface spans collection, search, research threads, notebooks, transformations, podcasts, source QA, and auditable outputs.\n- **The current fork boundary is explicit**. That makes progress easier to interpret because support, review, and stewardship are not hidden behind upstream ambiguity.\n\n## Public Proof, Not Vibes\n\n![Notebooklab public proof stack showing product proof, workflow proof, trust proof, and growth proof](https://raw.githubusercontent.com/xiaojiou176-open/notebooklab/main/docs/assets/proof/notebooklab-proof-stack.png)\n\nIf you want proof before commitment, start here:\n\n| What you want to verify | Where to look |\n| --- | --- |\n| The product is wider than a single chat screen | [docs/proof.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/proof.md) |\n| The fastest path to a real result | [docs/quickstart.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/quickstart.md) |\n| How to use it from coding agents | [docs/mcp.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/mcp.md) |\n| The current scope and readiness boundary | [docs/project-status.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/project-status.md) |\n| The current release story | [CHANGELOG.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/CHANGELOG.md) and [GitHub Releases](https://github.com/xiaojiou176-open/notebooklab/releases) |\n| The runtime shape | [docs/architecture.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/architecture.md) |\n| The support and security boundary | [SUPPORT.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/SUPPORT.md) and [SECURITY.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/SECURITY.md) |\n\nRelease visibility is one public signal, not automatic proof that the latest release-event build and asset set are clean.\n\n## Docs By Goal\n\n| If your goal is... | Start here | Then go deeper with... |\n| --- | --- | --- |\n| Evaluate whether Notebooklab is worth your attention | [README.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/README.md) | [docs/proof.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/proof.md), [docs/faq.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/faq.md) |\n| Get to a first visible result quickly | [docs/quickstart.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/quickstart.md) | [docs/installation.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/installation.md), [docs/configuration.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/configuration.md) |\n| Connect it to coding agents through MCP | [docs/mcp.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/mcp.md) | [docs/integrations/claude-code.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/integrations/claude-code.md), [docs/integrations/codex.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/integrations/codex.md), [docs/integrations/cursor.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/integrations/cursor.md), [docs/integrations/opencode.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/integrations/opencode.md) |\n| Understand concrete outcome-first use cases | [docs/use-cases/long-context-to-structured-notes.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/use-cases/long-context-to-structured-notes.md) | [docs/use-cases/source-grounded-ai-research.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/use-cases/source-grounded-ai-research.md), [docs/use-cases/ai-notes-with-receipts.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/use-cases/ai-notes-with-receipts.md), [docs/use-cases/source-grounded-drafts.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/use-cases/source-grounded-drafts.md), [docs/use-cases/source-to-verified-draft.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/use-cases/source-to-verified-draft.md) |\n| Understand the system shape | [docs/architecture.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/architecture.md) | [services/api/main.py](https://github.com/xiaojiou176-open/notebooklab/blob/main/services/api/main.py), [apps/web/src/components/layout/AppSidebar.tsx](https://github.com/xiaojiou176-open/notebooklab/blob/main/apps/web/src/components/layout/AppSidebar.tsx) |\n| Contribute safely | [CONTRIBUTING.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/CONTRIBUTING.md) | [docs/development.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/development.md), [MAINTAINERS.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/MAINTAINERS.md) |\n| Understand the current naming and domain boundary | [docs/brand-domain.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/brand-domain.md) | [docs/faq.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/faq.md), [docs/mcp.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/mcp.md) |\n\n## Use It With Coding Agents\n\nNotebooklab ships a first-party MCP server so you can bring the same outcome objects you use in the web workbench into coding-agent hosts.\n\nIf your starting problem is messy long context, start with [Long Context](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/use-cases/long-context-to-structured-notes.md) first and come back here when you want to carry those outcome objects into a host.\n\nCurrent public fit:\n\n- use Notebooklab with Claude Code through MCP\n- use Notebooklab with OpenAI Codex through MCP\n- use Notebooklab with Cursor through MCP\n- use Notebooklab with OpenCode through MCP\n\nCurrent boundary:\n\n- this is a compatibility surface, not an official vendor partnership\n- vendor names are used descriptively to show host targets\n- the product center is still Notebooklab's auditable/source-grounded outcome path, not the MCP control plane by itself\n\nStart with [docs/mcp.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/docs/mcp.md) if you want the MCP overview, then choose the host-specific page that matches your agent runtime.\n\nIf you want checked-in starter artifacts instead of only setup pages, go straight to [examples/hosts/README.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/examples/hosts/README.md). That index links the public-ready Claude Code, Codex, Cursor, and OpenCode starter bundles plus the OpenClaw-compatible bundles and the ClawHub submission pack.\n\n## Product Surface At A Glance\n\nThe main workbench currently spans:\n\n- **Collect**: sources and notebooks\n- **Process**: ask/search and transformations\n- **Create**: podcasts, notebook drafts, and auditable markdown\n- **Manage**: models, settings, advanced controls, and repo-local governance\n\n![Notebooklab architecture view showing the UI, API, core runtime, and verification surfaces](https://raw.githubusercontent.com/xiaojiou176-open/notebooklab/main/docs/assets/architecture/notebooklab-architecture.png)\n\n## Best Fit And Not The Best Fit\n\n**Notebooklab is a strong fit if you want to:**\n\n- turn notes, documents, audio, and web sources into inspectable markdown outputs\n- run a repository-documented research workbench instead of depending on a hosted SaaS\n- evaluate a NotebookLM-like surface that already includes notebooks, search, podcasts, and auditable output paths\n- follow a repo that publishes proof surfaces, release surfaces, and stewardship boundaries in public\n\n**Notebooklab is probably not the best fit if you want:**\n\n- a zero-setup hosted product with no local runtime or Docker involvement\n- a multi-provider runtime by default today\n- a tiny single-purpose summarizer or export-only utility\n- a repo that hides fork lineage or blurs who currently owns support and review\n\n## Brand And Lineage\n\nNotebooklab is the public identity of this repository.\n\nThis repository is a deep, productized fork of the upstream Open Notebook project.\n\nUpstream lineage still matters for provenance, but the current support, review, release, and collaboration surface is repository-local to this checkout.\n\nThat means two things can both be true at once:\n\n- Notebooklab is the brand people should remember\n- upstream lineage still belongs in the evidence layer, not the headline\n\nThe repo-local stewardship and trust boundary is anchored in [NOTICE.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/NOTICE.md), [MAINTAINERS.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/MAINTAINERS.md), [SUPPORT.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/SUPPORT.md), [SECURITY.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/SECURITY.md), and [CONTRIBUTING.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/CONTRIBUTING.md).\n\n## Local Cleanup Operator Path\n\nWhen disk pressure comes from local Docker builders plus repo-local rebuildables,\nuse one explicit operator path instead of guessing which command owns which\nsurface:\n\n```bash\nmake cleanup-operator-dry-run\nmake cleanup-operator-rebuildable\nmake cleanup-operator-aggressive\n```\n\nThat split is intentional:\n\n- `docker-buildx-clean` handles local Buildx/builder residue\n- `tooling/scripts/ops/cleanup_runtime_cache.sh` handles repo-local runtime and rebuildable surfaces\n- `apps/web/node_modules` remains a repo-local rebuildable dependency root, not a machine cache\n- repo-related machine caches still stay review-first instead of being auto-cleared by repo automation, and they are limited to download-cache surfaces only\n\n## Community\n\n- **Questions and product discussion**: GitHub Discussions\n- **Reproducible bugs and scoped improvements**: GitHub Issues\n- **Contribution flow**: [CONTRIBUTING.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/CONTRIBUTING.md)\n- **Support boundary**: [SUPPORT.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/SUPPORT.md)\n- **Security reporting**: [SECURITY.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/SECURITY.md)\n- **Collaboration conduct**: [CODE_OF_CONDUCT.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/CODE_OF_CONDUCT.md)\n- **Current review routing**: [CODEOWNERS](https://github.com/xiaojiou176-open/notebooklab/blob/main/CODEOWNERS)\n- **Stewardship and provenance**: [NOTICE.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/NOTICE.md) and [MAINTAINERS.md](https://github.com/xiaojiou176-open/notebooklab/blob/main/MAINTAINERS.md)\n\n## License\n\nThis fork continues to distribute upstream-licensed material under [LICENSE](https://github.com/xiaojiou176-open/notebooklab/blob/main/LICENSE).\n","changelogMarkdown","videoKey","requiredSecretsJson","publishedAt",1775685883000,"priceCents","trialDays","iconType","image","upstream",{"_108":109,"_110":111,"_112":113,"_114":115,"_116":117,"_118":119},"requiredSecrets",[],"tags",[97,98,99,100,101,102,103,104,105,106,107],"assets",[],"stats",{"_95":44,"_96":44},"ratingAvg","reviewCount","ai-notes","auditable-markdown","citations","knowledge-management","long-context","notebooks","research-assistant","research-threads","research-workbench","source-grounded-writing","traceable-writing","topics",[97,98,99,100,101,120,102,103,104,105,106,107],"license","MIT","language","Python","registryName","io.github.xiaojiou176-open/provenote-mcp","homepage","https://github.com/xiaojiou176-open/provenote/blob/main/docs/mcp.md","packages",[],"mcp","ENV",{"_123":124,"_125":126,"_127":128,"_129":-7},"ENVIRONMENT","production","TURNSTILE_SITE_KEY","0x4AAAAAAB8-rEI2gNDf_QhU","SENTRY_DSN","https://45fd1229f78089d5c3ee78be128030e1@o4510583261757440.ingest.de.sentry.io/4510584253448272","SENTRY_ENABLE_LOCAL"]