Documents and notes decision
Document.Bot
A capable developer can assemble a functional local-first RAG desktop tool from existing open-source building blocks, but reproducing the official product's polish, enterprise connectors, EU-hosted managed options, and air-gapped packaging is work-heavy and operationally distinct.
Visit website↗Built by Mick.net - Maker: Document.Bot 🤖 BestTime.app 🎉, who ships 3 products in this index
Not priced
No pricing page we fetched carried a figure, so there is nothing to compare against. The build side is still real.
$100one-off100 h to build
$50/mo6 h/mo upkeep
No published price to break even against.
No open-source build does this yet
Nothing published replaces this one, so a replacement starts from an empty file. Here is what it would have to cover.
What a replacement has to do
- Index local folder → retrieve relevant passages with citations → draft an answer from selected sources → inspect and accept/reject edits
What it still won’t have
- Polish and UX quality of the official desktop app (installer, native behaviours, polished editor)
- Managed EU-hosted or enterprise self-hosted deployment options and SLAs
- Packaged offline model hosting and vendor support for air-gapped deployments
- Integrated team features (shared workspaces, connectors to SharePoint/OneDrive/S3) and commercial onboarding
What remains hard
- Compliance and regulation
Stay AI (EU) compliant
- Compliance and regulation
Choose hosted, EU-hosted, or offline models depending on privacy rules, client files, and workspace boundaries.
First-year cost
No published price
Document.Bot does not publish a price we could read, so there is nothing to compare against. What building costs is below.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a cross-platform Electron desktop app (Electron + React) with a local SQLite metadata store and a vector index (use an open vector store library) that: 1) lets a user choose a workspace folder and recursively index PDFs, .docx, .xlsx, .md and text files (extract text and structural anchors like PDF page, Word paragraph, Excel cell ranges); 2) builds embeddings and searchable text index and returns retrieval results with exact file + page/paragraph/cell provenance; 3) provides a chat UI that accepts tagged source selections, runs RAG prompts against either local LLM servers (Ollama/LM Studio) or remote OpenAI-style APIs (API key input), and shows an explicit inspection trace of files/sections opened; 4) includes a reviewable editor that suggests changes to a file and lets the user accept or reject before saving; 5) includes error handling, credential storage encryption, unit/integration tests for indexing and retrieval, and an automated packaging pipeline for macOS/Windows. Out of scope: enterprise connectors (SharePoint/S3), multi-user sync, managed EU cloud hosting, and advanced team features.
How we checked
How the score was reached
- Partly verdict base52
- 2 cited sources+1
- 3/3 assessment runs agreed+4
- Hard moats found in the evidence-3
- Evidence score54
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 2
Every page the run actually retrieved.
- official productDocument.Bot - official product
- official docsDocument.Bot Docs
Integrity checks
What held up, and what did not.
