Opportunities › Documentation context providers for AI coding assistants › Proposed app
Docket Shelf
Private team docs, served to your AI coding assistant over MCP
- Score
- 44
- confidence 47
- Evidence
- 4
- items on the page
- Rivals
- 3
- named in sources
- First version
- 30 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 2025
- New market · AI estimate
ScreensConcept
Desktop app: Developers use this integrated with code editors during long coding sessions where documentation lookup and context matter most. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
Sources
All sources healthy4 sources
About this appAI proposal
Docket Shelf indexes your team's own documentation (internal wikis, API references, markdown in repos, runbooks) and serves the right passages to Cursor, Windsurf and VS Code assistants through an MCP server. Your assistant stops guessing and stops relying on stale training data.
It is built for engineering teams whose important docs are custom or internal. Instead of pasting snippets into a chat window, you connect sources once. The assistant asks, and Docket Shelf answers with short, cited excerpts matched to the code in play.
What is different: every answer shows its source and last-updated date, docs are pinned to the versions your project uses, and a team dashboard shows which questions found nothing, so you know what to write next.
- Tagline
- Give your AI assistant the docs your team actually uses.
- Voice
- PlainPreciseCalm
- Why this name
- A docket is a list of items to be dealt with; a shelf is where reference material sits. Together: your docs, ready to hand.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
One MCP endpoint for every editor
Add a single server URL to Cursor, Windsurf or VS Code. Each teammate gets the same docs without per-person setup.
No member yet
Private and internal sources
Index Git repos, markdown folders and published doc sites. Access follows team membership, so private docs stay private.
No member yet
Version pinning
Link a project's lockfile and Docket Shelf serves docs for the versions you use, not the latest release.
No member yet
Cited, trimmed excerpts
Replies return only the relevant passages with a source link and freshness date, keeping the context window small.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#12343BDeep Spruce · primary#F6F2E9Paper · surface#1B1F23Ink · text#E8A13AAmber · accent#4F8A6BMoss · data
Typography
Aa
Docket Shelf
ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789
Give your AI assistant the docs your team actually uses.
PricingAI proposal
The material gives no competitor prices. These tiers are a starting point to test against the open-source alternatives noted in the evidence.
Solo
Free
- 3 sources
- Public and private repos
- Community support
Team
$20 a month
- Up to 10 seats
- Unlimited sources
- Version pinning
- Gap report
Business
Contact us
- SSO
- Audit log
- Self-host support
- Priority support
Would you pay Team · $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
- Users invite others. A team admin connects sources, then invites teammates to add the MCP URL to their editors. Each added seat and shared source makes the tool more useful to the rest of the team.
- Do they become customers?
- Invited teammates are developers who can become paying seats or push their own teams to adopt. Test how many invited seats connect an editor and how many teams upgrade from Solo.
- First customers
- Engineering teams of 5 to 30 developers who use Cursor or VS Code daily and keep internal API docs in Git or a wiki.
- Where to reach them
- Developer communities: Post setup guides and demos in Cursor, MCP and AI coding forums and Discords.
- GitHub open core: Publish the self-host core so developers try it first, then upsell hosting and team features.
- Direct outreach: Message engineering leads at small software companies with public API docs and private repos.
- First milestone
- Ten teams connect a private source and five pay for Team within 60 days of launch.
Why people already pay for this
- VerifiedGitHub repository gained 1789 stars in 570 days
- Public claimOpen-source alternative to paid services for documentation context in AI coding environments
- AI inferenceActive adoption and community interest in documentation context tools for AI coding
- Public claimMultiple AI coding platforms (Cursor, Windsurf, VS Code) represent the target market for this solution
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 · Vector searchPostgres pgvector with keyword searchHybrid retrieval over indexed docs without adding a separate database.
- This app · MCP serverRemote MCP server over streamable HTTPEditors connect to one hosted URL with per-team tokens.
RisksAI inference
- CommercialFree and open-source alternatives exist, and the evidence says users cite them as alternatives to paid services. Willingness to pay for private-doc support is untested.
- PlatformEditors may add built-in doc indexing or change MCP behaviour, making a separate provider less necessary.
- TechnicalRetrieval quality on messy internal docs decides trust. Poor matches would waste context and hurt the assistant.
- DataCustomers' private docs are sensitive; hosted indexing needs strict tenant isolation and clear retention rules.
- AIAssistants may ignore or misuse returned excerpts, so users may blame the tool for errors it did not cause.
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.