Croft

Croft turns Claude into your full-stack builder. Describe an app in plain language — a CRM, a habit tracker, an expense splitter — and Claude writes the code, deploys it to your private Croft workspace, and gives you a live URL. Your apps run on a server you own, stay private to you and the people you invite, and are versioned on every change so you can roll back anytime. Key features - Build by conversation — Claude scaffolds, writes, and iterates on real apps from a description. - One-step deploy — every change is committed and shipped to a live, - Live status & logs — check whether an app is up and read its runtime logs without leaving the chat. - Secrets management — store API keys and env secrets encrypted, injected at deploy. - Team access — invite people by email and grant per-app access; you stay in control of who sees what. - Version history & rollback — every deploy is a release you can revert.

View in the Claude registry

Integration type: Connector
Verification status: Community connector
Platform: Claude
Category: AI
Primary Subcategory: AI App & Website Builders
Secondary Subcategories: None listed
Brand: Croft
Access: No account required
First tracked: 2026-08-03
Tool count: 19
Geography: US

Claude Discoverability Score

Live · refreshed daily

Croft in AI App & Website Builders

0/100

Invisible

#11of 18competitors

This is the score for the Integration’s Primary Category, AI App & Website Builders. Croft's tools align with this Category, which is why we measure it here. Claude did not find it in connector search, show it in the connector picker, or invoke it in the measured contested Runs. That is why the score is 0.

How this score is measured

Improve your score Talk to our team

What we measure

One core score. Three important factors to discoverability.

Picked
· 0/100
Sets the score
Found
· 0/100
Diagnostic
Positioned
· 0/100
Diagnostic

Primary Subcategory

Competing in Claude AI App & Website Builders

View Category

8 scoredVerified connectors by score · Community connectors A–Z

Primary Subcategory

1Supabase65/100

2Lovable58/100

3v042/100

4Base4423/100

5Replit23/100

—AppDeployOrganic discovery is pending verificationGet notified when verified

—CaffeineOrganic discovery is pending verificationGet notified when verified

—CroftOrganic discovery is pending verificationGet notified when verified

—DeplixoOrganic discovery is pending verificationGet notified when verified

—EmergentOrganic discovery is pending verificationGet notified when verified

—HatchableOrganic discovery is pending verificationGet notified when verified

—icod.aiOrganic discovery is pending verificationGet notified when verified

—PepitaOrganic discovery is pending verificationGet notified when verified

—WyberAiOrganic discovery is pending verificationGet notified when verified

Score explained

What does 0/100 mean?

0 is Croft's rounded picker appearance rate. It appeared in the connector picker, or Claude invoked it directly, in 0% of contested conversations. All three measures are important to discoverability: Picked determines the headline score, Found shows how often Claude found Croft in connector search, and Positioned shows where it appeared in the picker.

InvisibleBuriedEmergingCompetitiveWinning

Croft is 10 points from Buried.

FoundDiagnostic

Whether Claude found your Connector in connector search. It must be Found before it can reach the picker, but the score counts picker appearances—not search results.

PickedMain score

How often your Connector appeared in the picker, or Claude invoked it directly, across contested conversations. This percentage is the Discoverability Score; the headline number is rounded.

PositionedDiagnostic

What position your Connector appeared in when it was shown in the picker. This shows prominence, but it does not affect the score.

Read the full scoring methodology Recommended guideHow to get your Claude Connector organically discoveredFollow the practical steps agents use to find, shortlist, and choose your Connector.

Tools

19 tools agents can invoke

Recent runtime logs for an app (relayed from its box).
logs
Reports an app's deploy and health status. `live` is true only when the latest deploy has fully succeeded; `deploy_state` gives the current stage (queued, running, succeeded, or failed). Read-only and safe to call repeatedly — each call returns the current stage, so it can be polled to follow a deploy's progress.
status
Provisions an empty app on this workspace: a git repo, subdomain, port, and golden-path scaffold files. It contains no feature implementation and is not yet live.
create_app
Permanently deletes an app: removes its containers, data, and source code from the server, then deletes the app record. Irreversible. Does not affect other apps or the server. Deletes only when confirm is true.
delete_app
Commits, builds (Nixpacks), and deploys the app. Returns a deploy job id; the build runs in the background and its deploy_state moves through queued and running before it ends in succeeded or failed. The app is live only once the deploy has succeeded.
deploy
Get an app's file manifest, recent releases, and secret key names.
get_app
Return the Croft golden-path rules that every app on this workspace must follow.
get_platform_conventions
Grant an existing user access to an app.
grant_access
Invite someone by email, set their workspace role, and grant app access. role is member (use apps), creator (can build their own apps), or admin. apps is 'all' or a list of app names.
invite_user
List the apps in this workspace with their URLs, status, and last release.
list_apps
List an app's secret key names (never values).
list_secret_keys
List the people in this workspace and which apps they can access.
list_users
Read one file from an app's own git repo on this Croft workspace. No external API — the path is resolved inside the app's repository.
read_file
Revoke a user's access to an app.
revoke_access
Re-deploy the previous release of an app.
rollback
Sets which of an app's paths are reachable WITHOUT login. By default an app is fully private (workspace login required everywhere). Pass the complete desired list of specific paths (e.g. ["/survey"]); each listed path and everything under it becomes public, everything else stays private. An empty list makes the app fully private again. The whole app cannot be made public — root or wildcard paths are rejected.
set_public_paths
Set (or replace) an env secret for an app. Value is stored encrypted and injected at deploy.
set_secret
Writes files into an app's git repo and commits them (as 'Claude via Croft'). files is a list of {path, content}. Committed files are staged in the repo; they are not built or published until the app is deployed.
write_files
No public description for this tool yet.

Croft FAQ

How do I improve a Community connector's discoverability?

The levers are the listing surface agents actually read: names, descriptions, keywords, tool metadata, and registry health. Which lever matters depends on where discovery breaks, which is what continuous measurement shows.

What are Croft alternatives on Claude?

As of 2026-08-14, Croft competes with Supabase, Lovable, v0, Base44, Replit, Floot, Macaly, Retool, AppDeploy, Caffeine, Deplixo, Emergent, Hatchable, icod.ai, Jotform Apps, Pepita, WyberAi in Claude AI App & Website Builders, ranked by public Discoverability Score.

Where does Croft rank in AI App & Website Builders on Claude?

As of 2026-08-14, Croft ranks #11 of 18 in Claude AI App & Website Builders with a Discoverability Score of 0/100 (Invisible).

Where is this profile measured?

This profile uses the geography attached to the latest public registry snapshot: US. Locale tags are intentionally omitted.