Opportunities › Extensible AI code assistant platform › Proposed app
Loomdesk
A coding assistant you reshape by asking it to write its own plugins
- Score
- 21
- confidence 46
- 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
- 2021
- Emerging market · AI estimate
ScreensConcept
Desktop app: This coding assistant with plugin authoring and permission review is used at a desk during active development work requiring sustained keyboard interaction. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
New plugin
Ready to review- RequestPlan before edits
About this appAI proposal
Loomdesk is a browser-based AI coding assistant built around one idea: when you need it to behave differently, you tell it, and it writes the plugin itself into a folder it watches. No waiting for a vendor release, no switching tools.
It is for developers who already use AI code assistants and keep hitting the edge of what the tool allows. Use it to add a custom plan mode, run cleanup after each edit, or assess pull requests your way. Plugins are plain files you can read, edit, version and share.
- Tagline
- Ask for the behavior you want. Loomdesk builds the plugin.
- Voice
- DirectPracticalCalm
- Why this name
- A loom weaves threads into the cloth you want. Plugins are the threads, and the desk is where a developer works.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
Plugins written by the assistant
Describe a behavior in plain words. Loomdesk writes the plugin into your plugins folder, loads it and shows you the diff.
No member yet
Plain files in designated folders
Plugins are readable files in project or personal folders. Commit them, review them and carry them between machines.
No member yet
Hooks for the whole session
Attach plugins to events such as before a plan, after an edit, on commit or when a pull request opens.
No member yet
Permission preview
Each plugin lists the files, commands and network calls it asks for. You approve once per project.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#F6F1E7Warm Paper · surface#1F2A2EInk · text#1F5C4DSpruce · primary#D9622BEmber · accent#D8CDB6Sand · borders
Typography
Aa
Loomdesk
ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789
Ask for the behavior you want. Loomdesk builds the plugin.
PricingAI proposal
The material gives no competitor prices. These tiers are a starting point to test, and the marketplace fee idea from the opportunity is left for later.
Free
Free
- One project
- 3 active plugins
- Bring your own model key
Pro
$20 a month
- Unlimited projects and plugins
- Dry runs and rollback
- Share plugins by link
- Permission previews
Team
$40 a month
- Everything in Pro
- Shared team plugin folder
- Approval rules per project
Would you pay Pro · $20 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
- Shared outputs. Users publish plugins as links. Recipients read the permissions and install into their own folder, so each shared plugin shows Loomdesk to other developers.
- Do they become customers?
- Recipients are developers who could become paying users. Test how many who open a shared link create an account and install a plugin.
- First customers
- Solo developers and small-team engineers who already use a terminal AI assistant and write their own hooks or scripts for it.
- Where to reach them
- Hacker News: Show a post with three real plugins built live, such as plan mode and pull request assessment.
- Developer Discords: Join assistant communities and offer free setup for people sharing customization pain.
- First milestone
- Ten developers each build and keep three plugins for four weeks, and five of them pay for Pro.
Why people already pay for this
- Public claimDevelopers switch away from competing code assistants due to limitations in customization capabilities
- Public claimSelf-service plugin extensibility drives user satisfaction and retention
- Public claimUsers implement significant customizations including workflow automation, cleanup automation, and integration features
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 model accessModel API gateway, bring-your-own keyThe assistant needs a language model to write plugins; user keys keep early costs low.
- This app · Plugin sandboxIsolated runtime for untrusted codeAssistant-written plugins run code, so they need strict file, command and network limits.
RisksAI inference
- TechnicalRunning assistant-written code safely in a browser product is hard. A weak sandbox could leak files or secrets.
- CommercialThe evidence is a few Hacker News claims from users of one rival tool. Demand beyond enthusiasts is unproven.
- PlatformExisting assistants may add plugin support, which would remove the main difference.
- AIGenerated plugins may be wrong or unsafe, so users must trust the review step and dry runs.
- TechnicalA browser app may not match the local terminal workflow where these developers work now.
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.