Opportunities › Code review and change comprehension for AI-generated code › Proposed app
Readthrough
Read AI-written code changes by intent, one chapter at a time
- Score
- 35
- confidence 45
- Evidence
- 4
- items on the page
- Rivals
- 2
- named in sources
- First version
- 25 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 1976
- Mature market · AI estimate
ScreensConcept
Desktop app: Developers read and review AI-written code changes chapter by chapter, marking and annotating their understanding during focused code review sessions. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
Chapters
22 files, 840 lines6 chapters
About this appAI proposal
Readthrough takes the changes your coding agent made and sorts them into chapters by purpose, such as a new data model, the code that uses it, then tests and config. You read them in a browser one chapter at a time, instead of walking the repository tree file by file.
It is for developers and small teams who let AI agents write large changes and then have to check them. Point it at local changes before you open a pull request, or at a branch. It reads a plain git diff, so it works with whatever agent produced the code.
Each chapter has a short summary of its intent, the hunks that belong to it, and a way to mark it understood, flag it or leave a note.
- Tagline
- Understand what your agent changed before you merge it.
- Voice
- CalmClearPlain
- Why this name
- A readthrough is how you go through a draft in order, which is what the product does with a change.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
Chapters by intent
Changes are grouped by purpose, not by folder. One idea that touches six files is read in one place.
No member yet
Step-by-step reading
Move through chapters in a suggested order, with a progress bar showing how much review is left.
No member yet
Works with any agent
Reads a plain git diff, so it does not matter which coding agent or editor produced the change.
No member yet
Review local changes first
Run it on uncommitted work and read it before you open a pull request.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#12372AInk Green · primary#F6F3EAPaper · surface#1E2420Charcoal · text#E0A32EAmber · accent#B5482FBrick · flag
Typography
Aa
Readthrough
ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789
Understand what your agent changed before you merge it.
PricingAI proposal
The material gives no competitor prices. It only says the model is a subscription with a free open-source CLI. These tiers and prices are a starting point to test.
Open CLI
Free
- Open-source command line tool
- Chapter view of local changes
- Runs on your machine
Solo
$12 a month
- Hosted browser reader
- Saved review history
- Shareable review links
- Notes to agent prompts
Team
$24 a month
- Per seat, everything in Solo
- Shared reviews and comments
- Team review progress view
Would you pay Solo · $12 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
$190,000
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
- Shared outputs. A developer sends a chapter-ordered review link to a teammate or puts it in a pull request description. The reader sees the product name on the page and the CLI install line.
- Do they become customers?
- Link readers are developers who may install the free CLI, but that is not payment. Test how many readers install it and how many of those later pay for sharing.
- First customers
- Small engineering teams of 3 to 15 developers who already use AI coding agents daily and open pull requests with large agent-written changes.
- Where to reach them
- Hacker News and Show HN: Launch the open-source CLI there, where the topic already drew 46 points and 32 comments, and ask for feedback.
- Open-source CLI on GitHub: Publish the CLI with a clear README and example chapters from public repositories.
- Agent user communities: Share examples in forums and chats where people discuss coding agents and their review pain.
- First milestone
- Five small teams pay for the hosted service within 60 days of the CLI launch.
Why people already pay for this
- VerifiedHacker News post received 46 points and 32 comments
- Public claimDevelopers find reading changes organized by logical chapters easier than reading them in repository tree order
- Public claimUsers want the structured chapter reading experience for local changes before opening a PR
- Public claimThe tool works with any coding agent
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 · AI groupingHosted LLM API for chapter grouping and summariesGrouping hunks by intent and writing chapter summaries needs a language model.
- This app · CLIOpen-source Node command line packageThe free CLI reads local git diffs and opens the browser view or uploads to the hosted service.
RisksAI inference
- AIWrong or misleading chapter grouping and summaries could make reviewers trust code they have not understood. Hunks must stay fully visible and ungrouped hunks must be shown.
- CommercialA free open-source CLI may satisfy most solo users, and the evidence only shows interest, not willingness to pay. Teams must be tested separately.
- PlatformCode hosts and IDEs or agent vendors could add intent-grouped diffs to their own review views.
- DataTeams may refuse to upload private source code to a hosted service and its model provider. A local-only mode may be needed.
- TechnicalVery large diffs may exceed model context limits and make grouping slow or costly.
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.