Opportunities › Local knowledge compilation for enterprise teams › Proposed app
Pagewright
Compile raw documents into a wiki your team can read, check and host itself
- Score
- 58
- confidence 35
- Evidence
- 3
- items on the page
- Rivals
- 2
- named in sources
- First version
- 45 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 2023
- Emerging market · AI estimate
ScreensConcept
Desktop app: Teams compile large document collections, edit wikis, resolve conflicts, and review source trails—all knowledge work best done with full text visibility at a desk. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
Compile run
Finished in 14 min412 docs
About this appAI proposal
Pagewright compiles a folder of raw documents (PDFs, notes, slides, exports) into a structured wiki of plain, linked pages. Every page shows the source passages it came from and the step that wrote it, so people can read the knowledge instead of trusting a hidden index.
It is for enterprise teams and knowledge workers who organize large document collections, and for people who feed knowledge to AI agents. It runs with a local model on your own infrastructure, so documents need not leave your network.
Unlike a typical RAG setup, the output is a wiki you can edit and correct. When a page is wrong, you fix it once and the fix carries into later compiles. Agents then query the wiki and get citations back to the source.
- Tagline
- Knowledge you can read, check and keep.
- Voice
- plaincarefultransparent
- Why this name
- A wright is a maker of things, as in wheelwright. Pagewright makes pages from raw documents, and the plain compound is easy to say.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
Readable compile steps
Each run lists what was read, split, merged and written in plain language. Open any step to see its input and output.
No member yet
Pages with source trails
Every claim on a wiki page links to the exact passage in the original document it was drawn from.
No member yet
Runs on your own servers
Deploy on your infrastructure with a local model. No external API is required to compile or search.
No member yet
Edits that stick
Correct a page by hand and Pagewright keeps the correction when sources change and the wiki is recompiled.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#1F5C4APine · primary#F6F1E7Parchment · surface#1F2A24Ink · text#D9822BAmber · accent#3A6EA5Slate blue · data
Typography
Aa
Pagewright
ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789
Knowledge you can read, check and keep.
PricingAI proposal
The material gives no competitor prices. These tiers are a starting point to test, based on the document volume and support options named in the opportunity.
Self-host
Free
- One wiki, up to 200 documents
- Local model support
- Readable compile steps
- Community support
Team
$99 a month
- Up to 5,000 documents
- Review queue and edit history
- Agent read API
- Email support
Business
$399 a month
- Up to 50,000 documents
- Single sign-on
- Gap and staleness reports
- Priority support
What it could be worthEstimate
What a software business making this much sells for today, from real deals. Pick a monthly revenue.
If it makes
Worth about
$194,000
Sale range $132,000 to $263,000
$120,000 a year in revenue
From 87 software businesses listed and sold on Empire Flippers in the last year, revenue vetted by the broker: they sell for about 19× their monthly revenue, most between 13× and 26×.
Measured on profit, the middle deal is 32× monthly net profit: about 2.7 years of profit.
An estimate, not an offer or financial advice. A real price depends on profit, growth, churn, how much of the work is the owner's, and the buyer.
How it would growAI proposal
How the product would reach new customers through its own use, and how the first ones would be found. A plan to test, not a result.
- Built-in loop
- No loop in the product. Wikis are used inside a company and stay private. Some read-only page links could be shared with colleagues, but the product mainly reaches customers through content, open-source communities and direct sales.
- Do they become customers?
- Viewers inside a company are colleagues of the buyer and could ask for their own wiki. Test whether any of them become admins who start a second wiki or pay.
- First customers
- Knowledge or IT leads at 50 to 500 person companies in regulated fields such as legal, finance or healthcare administration, who want AI over internal documents without cloud services.
- Where to reach them
- Open-source communities: Post compile walkthroughs in GitHub discussions and forums where local-first RAG alternatives are discussed.
- LinkedIn outreach: Message knowledge managers and IT leads in regulated companies with a demo wiki built from their public documents.
- Technical blog: Publish guides on moving from RAG to a readable wiki with citations, ending with a free self-host trial.
- First milestone
- Five paying teams in one regulated vertical, each compiling a real wiki of over 500 documents within 30 days of trial.
Why people already pay for this
- VerifiedGitHub repository gained 1559 stars in 180 days
- Public claimSynthadoc is positioned as a transparent, human-readable alternative to traditional RAG that can be self-managed and self-improved without external tools.
- AI inferenceThe repository demonstrates market adoption with 1,559 stars and 158 forks on GitHub.
Tech stack
Every proposal builds on the same PWA base, installable from the browser on any phone or computer. Dashed boxes are what this app adds.
- AppNext.js and TypeScript
- PWAManifest and service worker
- UITailwind CSS
- Data and authSupabase Postgres
- PaymentsStripe
- EmailResend
- HostingVercel
- This app · Local model runtimeSelf-hosted LLM server with open-weight modelsCompiling and answering must work without external APIs, so the model runs on the customer's hardware.
- This app · Document parsingOpen-source PDF and Office text extractionRaw files must become clean text with page positions so each claim can cite its passage.
RisksAI inference
- TechnicalLocal open-weight models may write weaker or less consistent pages than hosted models, and quality may vary by customer hardware.
- TechnicalKeeping hand edits through recompiles is hard when sources change; merge errors could erase corrections and lose trust.
- CommercialThe open-source project in the evidence is free and popular, so teams may self-host it rather than pay. Paid value must come from support, scale or review tools.
- AIThe model may invent claims or attach a wrong source. Citations must be checked automatically, and unsupported sentences flagged.
- DataEnterprise documents hold confidential data. Access rules from the source systems must carry into the wiki, or pages could expose restricted content.
Everything marked AI proposal or AI inference was written by a model from the evidence on the opportunity page. The proposed name must be checked against trademark registers before it is used.