Opportunities › Word document editing for AI agents › Proposed app
Paperhand
Let AI agents edit Word files like HTML, with formatting kept intact
- Score
- 41
- confidence 52
- Evidence
- 11
- items on the page
- Rivals
- 8
- named in sources
- First version
- 35 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 2010
- Mature market · AI estimate
ScreensConcept
Desktop app: AI developers configure agents, review diffs, manage budgets, and monitor document processing workflows from a development environment. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
Run #2841
Saved3 calls
About this appAI proposal
Paperhand is a hosted MCP service for teams whose AI agents have to change Word documents. The agent sees a clean HTML view of the DOCX, sends plain find-and-replace edits, and Paperhand writes those changes back into the original file. Styles, numbering, tables and tracked structure stay where they were.
It is built for developers at legal tech, life sciences and govtech companies, and for AI teams running document pipelines. Agents spend their context on the task instead of on OOXML. There is no new editing language to learn and no dozens of tools to choose from: a handful of calls reads, edits and saves.
You get a dashboard to watch edits, replay failed ones, and see what was changed in each file. Edits are metered, so you pay for what your agents do. A free tier lets you test on your own documents before you commit.
- Tagline
- Your agent edits. The document stays itself.
- Voice
- PlainPreciseCalm
- Why this name
- A short, plain word for work done carefully by hand on paper, which is what the service lets an agent do to a Word file.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
HTML view of any DOCX
Agents read a compact, HTML-like version of the document they already understand, not raw OOXML or a custom syntax.
No member yet
Find-and-replace edits
Edits are plain before/after text on the HTML view. No new DSL, no code written against low-level Word libraries.
No member yet
Write-back to the original file
Changes are reconciled into the source DOCX so styles, numbering and layout are not lost the way a Markdown round trip loses them.
No member yet
Small tool surface
A few MCP tools instead of hundreds, so agents pick the right one and finish a task in few calls.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#1E5B53Pine · primary#F6F1E7Paper · surface#1F2A2EInk · text#D9622BMarker · accent#3B6EA5Slate blue · data
Typography
Aa
Paperhand
ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789
Your agent edits. The document stays itself.
PricingAI proposal
The material gives no competitor prices. Only the free tier of 500 edits a month and the idea of paid usage tiers come from it. The Team price is a starting point to test.
Free
Free
- 500 edits per month
- All MCP tools
- Edit log for 7 days
Team
$99 a month
- 10,000 edits per month
- Per-key budgets
- Edit replay and diff
- Log kept 90 days
Regulated
Contact us
- Zero data retention
- Self-hosted option
- Higher edit volumes
- Support agreement
Would you pay Team · $99 a month?
Your answer tells PWAS which prices people accept. It changes no score.
No answers yet.
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
- Partners and integrations. Developers add Paperhand to their agents through MCP directories and framework guides. Each agent project that needs Word editing becomes a connection, and teams share a key across their agents.
- Do they become customers?
- Those reached are developers who can sign up and pay themselves. Test whether free-tier users with real documents move to paid within 30 days, and whether directory listings bring any.
- First customers
- AI engineers at small legal tech and life sciences startups (5 to 30 people) whose agents already draft or revise contracts or regulatory documents in Word.
- Where to reach them
- Hacker News and developer forums: Share a benchmark on public sample documents with call counts and cost, and invite readers to rerun it.
- MCP directories: List the server in public MCP registries with a one-line install and a free key.
- Direct outreach to legal tech teams: Message engineers building document agents and offer to run their worst DOCX files through a trial.
- First milestone
- Five teams in legal tech or life sciences paying for the Team plan after testing on their own documents.
Why people already pay for this
- VerifiedHacker News post received 38 points and 18 comments
- Public claimAI agents waste significant time and tokens on Word document mechanics, degrading performance on downstream tasks
- Public claimDocument editing tasks with existing solutions are time-consuming and produce low-quality results
- AI inferenceMultiple companies across different industries depend on AI-powered document editing
- Public claim
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 · DOCX processingOOXML parser and writer serviceConverts DOCX to simplified HTML and writes edits back without breaking styles.
- This app · Edit reconciliation modelFine-tuned model on GPU inferenceMaps HTML edits back to the right OOXML runs. Needs hosted inference with cost tracking.
RisksAI inference
- TechnicalMapping HTML edits back into OOXML is hard on real files with tracked changes, fields, nested tables and odd styles. Failures that corrupt a file would end trust quickly.
- CommercialRunning a fine-tuned reconciliation model has significant operating costs, per a public claim. Usage prices may not cover inference at free and low tiers.
- CommercialCompetitors such as SuperDoc and Adeu already offer MCP editing, and Claude's own DOCX skill may improve. The speed and call-count advantage is a public claim and must be re-tested.
- RegulatoryLegal, pharma and govtech buyers will ask for data retention limits, audit trails and self-hosting before they send documents to a hosted service.
- AIThe reconciliation model can place an edit in the wrong spot while appearing to succeed. The fidelity check must catch this reliably.
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.