v0 vs Floot

Agent access

1. ai-native userPoint an agent at llms.txt or agent-oriented docs

   weight 2 · round to Floot
   v0—none0/10

   No evidence that v0 publishes an llms.txt file or agent-oriented docs endpoint that other agents can be pointed at; the docs.md probe shows a markdown version of a page but not a standardized llms.txt/agent-discovery mechanism referenced anywhere in the pack.

   ⚑ flag

   Floot✓fullprobed8/10

   Floot serves a live llms.txt (HTTP 200) and an llms-full.txt with agent-oriented documentation of its capabilities, confirmed by direct probes. missing for 10: independent third-party confirmation that agents actually consume and act on these docs successfully, and docs.md/openapi endpoints return 404 suggesting the agent-doc surface is limited to llms.txt only.

   • [probe] “PROBE llms.txt: HTTP 200 at https://floot.com/llms.txt # Floot > Floot turns a plain-language description into a real, deployed full-stack …”
   • [claimed-docs] “point at an element or annotate a screenshot for the assistant to pick up”
   • [probe] “PROBE docs-md: HTTP 404 at https://floot.com/docs.md”
   • [probe] “PROBE openapi: all candidate paths 404 (https://floot.com/openapi.json, https://floot.com/swagger.json, https://floot.com/api/openapi.json, …”

   proof ↗⚑ flag

2. ai-native userRun the product headlessly / in CI for automation

   weight 2 · round to v0
   v0~partialclaimed5/10

   v0 provides a programmatic API/SDK (v0-sdk) allowing chat/code-gen calls to be scripted, and GitHub integration runs CI checks and can auto-fix CI failures, suggesting some automation-friendly hooks; but there's no documented headless CLI mode or explicit CI-runner integration pattern (e.g., GitHub Actions example) for orchestrating v0 itself within a pipeline. missing for 10: a documented CLI/headless mode, an official CI/CD pipeline example (e.g., GitHub Actions), and independent confirmation of running v0 fully unattended in automation.

   • [claimed-docs] “The v0 API gives you programmatic access to all of v0's capabilities, letting you integrate v0's AI-powered code generation, chat interface,…”
   • [github] “const response = await v0.chats.create({ message: 'Build me a personal website', })”
   • [github] “Set V0_API_KEY, or deploy server-side code on Vercel with OIDC enabled, then use the default client.”
   • [claimed-docs] “Connect a v0 project to GitHub to keep its code in a repository, work on isolated branches, preview changes, run CI, and publish through pul…”
   • [claimed-docs] “CI checks: See whether checks are running, passing, skipped, or failing.”
   • [claimed-docs] “v0 reads the relevant logs, diagnoses the failure, and applies a fix.”

   proof ↗⚑ flag

   Floot—none0/10

   Floot is a chat-based/MCP-driven app builder with no evidence of a CLI, API, or headless mode for CI/automation pipelines; MCP integration is for interactive AI clients (Cursor, VS Code, Zed), not headless CI execution. No mention of scripted builds, CI/CD triggers, or non-interactive automation. Missing for 10: CLI or SDK for scripted invocation, CI/CD pipeline documentation, headless build/deploy commands, API endpoints for automation.

   ⚑ flag

3. ai-native userPlug MCP servers into this product so it can use their tools

   weight 3 · round drawn
   v0—none0/10

   No evidence that v0 supports connecting to MCP servers or consuming their tools; docs mention integrations with databases, backend services, and AI providers but never reference MCP. missing for 10: any mention of MCP protocol/server support, documented tool-use via MCP, or third-party MCP connector.

   ⚑ flag

   Floot—none0/10

   All evidence describes Floot exposing itself as an MCP server that other clients (Cursor, VS Code, Zed) can connect to and use its tools (get_logs, query_database, etc.) — the reverse direction of this story. There is no evidence that a user can plug external MCP servers into Floot so Floot's own AI can call their tools.

   • [claimed-docs] “Add Floot to Cursor, VS Code, Zed, or any client that supports remote MCP servers”
   • [claimed-docs] “get_logs — the first step for any 500, blank page, or failed request.”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [probe] “official MCP server documented at https://floot.com/docs/integrations/other-mcp-clients”

   proof ↗⚑ flag

4. ai-native userUse an official CLI

   weight 2 · round drawn
   v0—none0/10

   Evidence shows a v0 API and a JavaScript/TypeScript SDK (v0-sdk) for programmatic access, but no official command-line interface tool is documented anywhere in the pack.

   • [claimed-docs] “The v0 API gives you programmatic access to all of v0's capabilities, letting you integrate v0's AI-powered code generation, chat interface,…”
   • [github] “const response = await v0.chats.create({ message: 'Build me a personal website', })”
   • [github] “Set V0_API_KEY, or deploy server-side code on Vercel with OIDC enabled, then use the default client.”

   proof ↗⚑ flag

   Floot—none0/10

   Floot is a chat/browser-based app builder that offers an MCP server for connecting external AI clients, but there is no evidence of an official CLI tool for AI-native workflows; all evidence points to MCP integration and web-based chat, not a CLI. Missing for 10: any mention of a CLI binary, npm/pip install, or command-line docs.

   • [claimed-docs] “Add Floot to Cursor, VS Code, Zed, or any client that supports remote MCP servers”
   • [probe] “official MCP server documented at https://floot.com/docs/integrations/other-mcp-clients”

   proof ↗⚑ flag

5. ai-native userDrive the product through a documented public API

   weight 3 · round to v0
   v0✓fullclaimed8/10

   v0 documents a full public API (v0.app/docs/api/platform) offering programmatic access to code generation, chat, project management, and deployment, backed by an official SDK (vercel/v0-sdk) with concrete code examples (chats.create, streaming reads, API key/OIDC auth). This is corroborated by GitHub SDK evidence, not just marketing docs. Missing for 10: independent third-party developer reports/tutorials building production integrations on the API, and more detail on API rate limits/versioning.

   • [claimed-docs] “The v0 API gives you programmatic access to all of v0's capabilities, letting you integrate v0's AI-powered code generation, chat interface,…”
   • [claimed-docs] “Custom chat interfaces - Build your own UI for v0's AI-powered code generation”
   • [claimed-docs] “Start with existing files (fast, no tokens)”
   • [claimed-docs] “GitHub integration for repository imports”
   • [claimed-docs] “Export capabilities for backup and migration”
   • [github] “const response = await v0.chats.create({ message: 'Build me a personal website', })”
   • [github] “const result = readV0Stream(serverResult.toResponse())”
   • [github] “Set V0_API_KEY, or deploy server-side code on Vercel with OIDC enabled, then use the default client.”

   proof ↗⚑ flag

   Floot~partialprobed7/10

   Floot documents a first-party MCP server exposing concrete tools (get_logs, query_database, execute_sql, create_checkpoint, generate_image) that let an AI client drive the product programmatically, which directly serves the AI-native persona. However, probes show no conventional public REST/OpenAPI API (openapi.json/swagger.json all 404, docs.md 404), so the 'documented public API' story is only satisfied via the MCP protocol rather than a broader API surface. Missing for 10: a traditional REST/OpenAPI-documented API or SDK, and independent confirmation of MCP tool reliability beyond vendor docs.

   • [claimed-docs] “Add Floot to Cursor, VS Code, Zed, or any client that supports remote MCP servers”
   • [claimed-docs] “get_logs — the first step for any 500, blank page, or failed request.”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “create_checkpoint labels a point you can revert to from the Checkpoint panel in your project view.”
   • [claimed-docs] “generate_image writes AI images into the project.”
   • [probe] “official MCP server documented at https://floot.com/docs/integrations/other-mcp-clients”
   • [probe] “PROBE openapi: all candidate paths 404 (https://floot.com/openapi.json, https://floot.com/swagger.json, https://floot.com/api/openapi.json, …”
   • [probe] “PROBE docs-md: HTTP 404 at https://floot.com/docs.md”

   proof ↗⚑ flag

6. ai-native userIssue scoped/least-privilege API credentials for an agent

   weight 2 · round drawn
   v0—none0/10

   Evidence shows only a single API key (V0_API_KEY) or OIDC-based server auth for the v0 SDK, with no mention of scoped, role-based, or least-privilege credential issuance for agents. Missing for 10: any documentation of granular permission scopes, per-agent tokens, or restricted API key capabilities.

   • [github] “Set V0_API_KEY, or deploy server-side code on Vercel with OIDC enabled, then use the default client.”
   • [claimed-docs] “Securely store sensitive information like API keys, tokens, and other credentials.”

   proof ↗⚑ flag

   Floot—none0/10

   No evidence Floot supports issuing scoped or least-privilege API credentials/tokens for agents; MCP integration docs describe tool capabilities (query_database, execute_sql, etc.) but nothing about credential scoping, permission levels, or least-privilege access control for agents. missing for 10: any mention of API keys, token scoping, permission tiers, or credential management for agent access.

   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [probe] “official MCP server documented at https://floot.com/docs/integrations/other-mcp-clients”

   proof ↗⚑ flag

7. ai-native userBuild against official SDKs

   weight 2 · round to v0
   v0✓fullclaimed7/10

   v0 provides an official v0-sdk on GitHub with documented API usage (v0.chats.create, streaming, API key auth) plus a Platform API documented for programmatic access to chat, code generation, and project management. missing for 10: independent/hands-on developer corroboration of SDK reliability, multi-language SDK support, and deeper API reference coverage beyond quickstart snippets.

   • [github] “const response = await v0.chats.create({ message: 'Build me a personal website', })”
   • [github] “const result = readV0Stream(serverResult.toResponse())”
   • [github] “Set V0_API_KEY, or deploy server-side code on Vercel with OIDC enabled, then use the default client.”
   • [claimed-docs] “The v0 API gives you programmatic access to all of v0's capabilities, letting you integrate v0's AI-powered code generation, chat interface,…”
   • [claimed-docs] “Custom chat interfaces - Build your own UI for v0's AI-powered code generation”

   proof ↗⚑ flag

   Floot—none0/10

   Floot exposes an MCP server with tool functions (get_logs, query_database, execute_sql, etc.) but there is no evidence of official language SDKs (e.g., npm/pip packages) or an OpenAPI spec for developers to build against programmatically; probes for openapi.json/swagger.json all 404.

   • [probe] “PROBE openapi: all candidate paths 404 (https://floot.com/openapi.json, https://floot.com/swagger.json, https://floot.com/api/openapi.json, …”
   • [probe] “PROBE docs-md: HTTP 404 at https://floot.com/docs.md”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”

   proof ↗⚑ flag

8. ai-native userSubscribe to events via webhooks

   weight 2 · round drawn
   v0—none0/10

   No evidence of a webhooks subscription/event system in v0's docs or API pack; the API is described as request/response (create chats, streaming) with no mention of webhook endpoints or event subscriptions.

   ⚑ flag

   Floot—none0/10

   No evidence pack item mentions webhooks, event subscriptions, or any outbound notification/event system that AI-native users could subscribe to; the docs cover database, logs, images, mobile apps, and MCP tooling but nothing about webhooks.

   ⚑ flag

Agentic features

1. ai-native userGet AI-generated insights and suggestions from my data inside the product

   weight 2 · round to v0
   v0~partialclaimed4/10

   v0 offers some AI-driven analysis of a user's own data in the sense of reading logs, diagnosing build failures, and applying fixes, and it 'plans, creates tasks, and connects to databases as it builds,' which is a narrow form of AI-generated insight/suggestion from data within the product. However, there is no evidence of broader data-insight features (e.g., analytics, trend/insight surfacing from business or usage data) beyond code/log diagnostics, and no independent corroboration of this specific capability. Missing for 10: dedicated data-insight/analytics features, independent hands-on evidence of 'insights' beyond build-error fixing.

   • [claimed-docs] “v0 reads the relevant logs, diagnoses the failure, and applies a fix.”
   • [claimed-docs] “Automatically fix errors in your code with intelligent diagnostics.”
   • [claimed-docs] “select Fix with v0 or Fix Build. v0 reads the relevant logs, diagnoses the failure, and applies a fix.”
   • [claimed-docs] “v0 plans, creates tasks, and connects to databases as it builds.”

   proof ↗⚑ flag

   Floot—none0/10

   Floot's docs show built-in analytics tracking (page views, traffic sources, trends) and AI tools that can query/execute against the app's database, but there is no evidence the AI surfaces proactive insights or suggestions derived from that data back to the user — the query/execute tools are for development/debugging, not analytical insight generation.

   • [claimed-docs] “Built-in visitor tracking: page views, traffic sources, and trends.”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”

   proof ↗⚑ flag

2. ai-native userSet up automations that run autonomously in the background

   weight 2 · round drawn
   v0—none0/10

   v0's docs describe on-demand code generation, GitHub integration, deployments, and API access, but there is no evidence of scheduled/triggered automations or background agentic tasks that run autonomously without a user-initiated chat or API call (e.g., no cron jobs, webhooks, or persistent background workers are described). The closest features (auto-fix on CI failure, 'Fix with v0') are triggered by user action or CI events within a user-initiated flow, not autonomous background automation.

   • [claimed-docs] “v0 reads the relevant logs, diagnoses the failure, and applies a fix.”
   • [claimed-docs] “select Fix with v0 or Fix Build. v0 reads the relevant logs, diagnoses the failure, and applies a fix.”
   • [claimed-docs] “open the branch menu and select Fix Conflicts, Fix CI, or Fix CI + Conflicts.”
   • [claimed-docs] “v0 plans, creates tasks, and connects to databases as it builds.”

   proof ↗⚑ flag

   Floot—none0/10

   Floot's evidence covers app-building tools (database queries, logs, checkpoints, image generation, MCP integration) but nothing about scheduled tasks, triggers, or autonomous background automations running independently of a chat session. There is no mention of cron-like jobs, webhooks-as-triggers, or agent workflows executing without user interaction.

   ⚑ flag

3. ai-native userDelegate tasks to a built-in AI assistant inside the product

   weight 3 · round drawn
   v0✓fullprobed8/10

   v0 is itself a built-in AI assistant that users delegate coding/build tasks to: docs show it plans, creates tasks, connects to databases, diagnoses and fixes build/CI errors autonomously (v0-docs-35, v0-docs-9/17/20, v0-docs-26), and community reports confirm a full 'dev-to-deploy' delegation experience with an MVP built end-to-end (v0-comm-1). Missing for 10: independent benchmarking of autonomous task delegation beyond anecdote, and no detail on multi-step task planning transparency.

   • [claimed-docs] “v0 plans, creates tasks, and connects to databases as it builds.”
   • [claimed-docs] “v0 reads the relevant logs, diagnoses the failure, and applies a fix.”
   • [claimed-docs] “Automatically fix errors in your code with intelligent diagnostics.”
   • [claimed-docs] “select Fix with v0 or Fix Build. v0 reads the relevant logs, diagnoses the failure, and applies a fix.”
   • [claimed-docs] “open the branch menu and select Fix Conflicts, Fix CI, or Fix CI + Conflicts.”
   • [community] “Built a feature-rich internal app (time tracker w/ auth, roles, export) and had a working MVP in users' hands by Monday. 'The speed was incr…”
   • [probe] “PROBE docs-md: HTTP 200 at https://v0.app/docs.md --- title: What is v0? description: v0 is an AI agent that helps anyone create real code a…”

   proof ↗⚑ flag

   Floot✓fullprobed8/10

   Floot's core product is an AI chat assistant that builds, debugs, queries databases, generates images, and manages checkpoints for the app directly from the chat interface, with tool actions like get_logs, query_database, generate_image documented. missing for 10: independent/hands-on corroboration beyond vendor docs, and clearer detail on task complexity/limits the assistant can handle autonomously.

   • [claimed-docs] “build and ship full-stack apps without leaving your AI chat”
   • [claimed-docs] “get_logs — the first step for any 500, blank page, or failed request.”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “create_checkpoint labels a point you can revert to from the Checkpoint panel in your project view.”
   • [claimed-docs] “point at an element or annotate a screenshot for the assistant to pick up”
   • [claimed-docs] “generate_image writes AI images into the project.”
   • [probe] “PROBE llms.txt: HTTP 200 at https://floot.com/llms.txt # Floot > Floot turns a plain-language description into a real, deployed full-stack …”

   proof ↗⚑ flag

4. ai-native userOperate the product with natural-language commands

   weight 2 · round drawn
   v0✓fullprobed8/10

   v0's core interaction model is natural-language chat: docs and SDK show building apps from prompts like 'Build me a personal website', plus NL-driven fixes, migrations, and mockup-to-UI conversion, and community reports confirm fast prompt-to-deploy workflows in practice. Missing for 10: independent evidence on breadth/reliability of more complex multi-step NL commands and any documented failure modes for ambiguous instructions.

   • [claimed-docs] “Create custom hooks and functionality”
   • [claimed-docs] “Turn mockups into real, high-fidelity user interfaces that reflect user flows and constraints.”
   • [claimed-docs] “v0 reads the relevant logs, diagnoses the failure, and applies a fix.”
   • [claimed-docs] “Quickly scaffold full-stack apps or components following best practices and modern standards.”
   • [claimed-docs] “Migrate to the latest Next.js features”
   • [claimed-docs] “Automatically fix errors in your code with intelligent diagnostics.”
   • [claimed-docs] “v0 plans, creates tasks, and connects to databases as it builds.”
   • [github] “const response = await v0.chats.create({ message: 'Build me a personal website', })”
   • [community] “Built a feature-rich internal app (time tracker w/ auth, roles, export) and had a working MVP in users' hands by Monday. 'The speed was incr…”
   • [probe] “PROBE docs-md: HTTP 200 at https://v0.app/docs.md --- title: What is v0? description: v0 is an AI agent that helps anyone create real code a…”

   proof ↗⚑ flag

   Floot✓fullprobed8/10

   Floot's core product is natural-language chat-based app building ('build and ship full-stack apps without leaving your AI chat', 'turns a plain-language description into a real, deployed full-stack app'), with rich agentic tool support (get_logs, query_database, create_checkpoint, generate_image) operable via NL commands. Missing for 10: independent/hands-on corroboration beyond vendor docs, and no detail on limits of NL command coverage for advanced operations.

   • [claimed-docs] “build and ship full-stack apps without leaving your AI chat”
   • [probe] “PROBE llms.txt: HTTP 200 at https://floot.com/llms.txt # Floot > Floot turns a plain-language description into a real, deployed full-stack …”
   • [claimed-docs] “get_logs — the first step for any 500, blank page, or failed request.”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “create_checkpoint labels a point you can revert to from the Checkpoint panel in your project view.”
   • [claimed-docs] “generate_image writes AI images into the project.”

   proof ↗⚑ flag

Api quality

1. ai-native userExplore an interactive API reference with runnable examples

   weight 2 · round drawn
   v0—none0/10

   v0 does publish API docs and an SDK with code snippets (v0-docs-4, v0-gh-1, v0-gh-2), but there is no evidence of an interactive API reference (e.g., a browsable console with runnable/try-it examples) — only static documentation pages and GitHub code samples. missing for 10: an interactive API console, in-browser runnable examples, evidence of a Swagger/OpenAPI-style explorer.

   • [claimed-docs] “The v0 API gives you programmatic access to all of v0's capabilities, letting you integrate v0's AI-powered code generation, chat interface,…”
   • [claimed-docs] “Custom chat interfaces - Build your own UI for v0's AI-powered code generation”
   • [github] “const response = await v0.chats.create({ message: 'Build me a personal website', })”
   • [github] “const result = readV0Stream(serverResult.toResponse())”
   • [github] “Set V0_API_KEY, or deploy server-side code on Vercel with OIDC enabled, then use the default client.”

   proof ↗⚑ flag

   Floot—none0/10

   There is no evidence of an interactive API reference or runnable examples; probes explicitly show no OpenAPI/Swagger docs (404s) and no docs.md, and the MCP tool docs are plain text descriptions without runnable playgrounds.

   • [probe] “PROBE docs-md: HTTP 404 at https://floot.com/docs.md”
   • [probe] “PROBE openapi: all candidate paths 404 (https://floot.com/openapi.json, https://floot.com/swagger.json, https://floot.com/api/openapi.json, …”
   • [claimed-docs] “get_logs — the first step for any 500, blank page, or failed request.”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”

   proof ↗⚑ flag

2. ai-native userDownload a machine-readable API spec (OpenAPI or equivalent)

   weight 2 · round drawn
   v0—none0/10

   v0 documents a REST-like API platform and SDK (v0-docs-4, v0-gh-1) but no evidence anywhere shows a downloadable OpenAPI/Swagger spec or other machine-readable API description; missing for 10: OpenAPI/JSON schema file, published spec endpoint, or SDK auto-generation from spec.

   • [claimed-docs] “The v0 API gives you programmatic access to all of v0's capabilities, letting you integrate v0's AI-powered code generation, chat interface,…”
   • [github] “const response = await v0.chats.create({ message: 'Build me a personal website', })”
   • [github] “Set V0_API_KEY, or deploy server-side code on Vercel with OIDC enabled, then use the default client.”

   proof ↗⚑ flag

   Floot—none0/10

   Direct probes for OpenAPI/Swagger specs all returned 404 at every candidate path, and no documentation mentions a machine-readable API spec for Floot's own platform API.

   • [probe] “PROBE openapi: all candidate paths 404 (https://floot.com/openapi.json, https://floot.com/swagger.json, https://floot.com/api/openapi.json, …”
   • [probe] “PROBE docs-md: HTTP 404 at https://floot.com/docs.md”

   proof ↗⚑ flag

3. ai-native userTest against a sandbox environment without touching production data

   weight 1 · round to v0
   v0~partialclaimed6/10

   v0 supports isolated branches and preview deployments that are separate from the production URL, letting changes be reviewed/tested before they affect production (v0-docs-38, v0-docs-39, v0-docs-29, v0-docs-11). However, there is no explicit sandbox/staging environment for data (e.g., a separate database or mock data layer) — integrations like Supabase/Neon/Upstash appear to connect directly to real backend services rather than a sandboxed copy. Missing for 10: explicit data-isolation/sandbox environment separate from production databases, first-party documentation of a test/staging data layer, and independent confirmation that production data is never touched during preview/testing.

   • [claimed-docs] “Preview deployments are built from working branches.”
   • [claimed-docs] “work on isolated branches, preview changes, run CI, and publish through pull requests”
   • [claimed-docs] “When multiple chats share a Project, deploying from any of them updates the same production URL.”
   • [claimed-docs] “One Project shares deployment, hosting, domains, and environment variables for the chats connected to it.”
   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase, and AI mod…”

   proof ↗⚑ flag

   Floot—none0/10

   Evidence shows database tools (query_database, execute_sql) and checkpoint/revert features, but nothing describes a separate sandbox/staging environment isolated from production data — checkpoints revert code state, not provide a parallel test database. Missing for 10: any mention of a staging/sandbox database, environment cloning, or explicit separation between test and production data.

   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “create_checkpoint labels a point you can revert to from the Checkpoint panel in your project view.”

   proof ↗⚑ flag

4. ai-native userRely on versioned APIs with a documented deprecation policy

   weight 2 · round drawn
   v0—none0/10

   While v0 documents an API platform (v0-docs-4, v0-docs-19) and an SDK (v0-gh-1..3), there is no evidence of API versioning scheme or a documented deprecation policy anywhere in the evidence pack.

   ⚑ flag

   Floot—none0/10

   No evidence of any versioned API or documented deprecation policy; OpenAPI probes returned 404 and no changelog/docs mention API versioning or deprecation practices. missing for 10: versioned API documentation, deprecation policy statement, API stability guarantees.

   • [probe] “PROBE openapi: all candidate paths 404 (https://floot.com/openapi.json, https://floot.com/swagger.json, https://floot.com/api/openapi.json, …”
   • [probe] “PROBE docs-md: HTTP 404 at https://floot.com/docs.md”

   proof ↗⚑ flag

1. ai-native userPerform bulk operations across many items at once

   weight 2 · round drawn
   v0—none0/10

   The evidence shows v0's API can programmatically create individual chats/projects, but there is no documentation or example of performing bulk/batch operations across many items at once (e.g., batch generation, bulk edits, mass deployment). Nothing in the docs or SDK references batching, looping over collections, or multi-item transactions.

   ⚑ flag

   Floot—none0/10

   Floot's evidence covers single-app database queries/writes (query_database, execute_sql) and other AI-assisted dev tools, but nothing describes bulk operations across many items/records/projects at once as a distinct capability. missing for 10: any documentation of batch/bulk editing, mass updates across records, or multi-item automation workflows.

   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”

   proof ↗⚑ flag

2. ai-native userDefine rules that trigger actions automatically on events

   weight 3 · round drawn
   v0—none0/10

   The evidence describes v0's code generation, GitHub integration, deployment, and API capabilities, but nothing shows a rules/trigger engine where users define 'on event X, do action Y' automations. Missing for 10: any documented rule engine, event triggers, or workflow automation configuration beyond routine CI/deployment fixes.

   ⚑ flag

   Floot—none0/10

   Evidence covers Floot's AI-assisted app-building tools (logs, DB queries, checkpoints, image generation, MCP integration) but nothing describes user-defined rules/triggers that automatically fire actions on events (e.g., webhooks, event-condition-action automation). No workflow-automation or trigger/rule engine is documented.

   ⚑ flag

3. ai-native userSchedule recurring jobs or workflows

   weight 2 · round drawn
   v0—none0/10

   No evidence anywhere in the pack of scheduling, cron-like triggers, or recurring workflow automation in v0; the product is oriented around chat-driven code generation, deployment, and API-triggered chats, not scheduled/recurring jobs.

   ⚑ flag

   Floot—none0/10

   No evidence anywhere in the pack of scheduled/recurring jobs, cron-like automation, or workflow scheduling features; Floot's documented tools cover logs, database queries, checkpoints, image generation, mobile app conversion, and self-hosting but nothing about recurring or scheduled execution.

   ⚑ flag

4. ai-native userVersion, review, and roll back my automations

   weight 1 · round to v0
   v0~partialclaimed6/10

   v0's GitHub integration gives git-based version control (isolated branches, commits, PRs, CI checks) which supports versioning and review of generated projects, and branch/PR mechanics implicitly allow reverting to prior states. However, there is no explicit 'rollback' feature or chat/version history UI documented specifically for automations, and 'automations' as a distinct concept (vs. generated code) isn't addressed. Missing for 10: dedicated rollback/undo capability, explicit versioning of automations (not just code), and independent evidence of review/rollback workflows in practice.

   • [claimed-docs] “Connect a v0 project to GitHub to keep its code in a repository, work on isolated branches, preview changes, run CI, and publish through pul…”
   • [claimed-docs] “The Duplicate Branch dialog creates a new chat and branch containing the current branch's commits.”
   • [claimed-docs] “open the branch menu and select Fix Conflicts, Fix CI, or Fix CI + Conflicts.”
   • [claimed-docs] “CI checks: See whether checks are running, passing, skipped, or failing.”
   • [claimed-docs] “work on isolated branches, preview changes, run CI, and publish through pull requests”

   proof ↗⚑ flag

   Floot~partialclaimed5/10

   Floot documents a checkpoint system where create_checkpoint labels a revertible point accessible from the Checkpoint panel, covering versioning and rollback of AI-driven changes. However, there is no evidence of a review/diff mechanism to inspect what changed before reverting, nor of a broader automation version history beyond simple checkpoints. Missing for 10: explicit review/diff tooling, multi-version history browsing, and independent confirmation of rollback reliability.

   • [claimed-docs] “create_checkpoint labels a point you can revert to from the Checkpoint panel in your project view.”

   proof ↗⚑ flag

Ai features

1. ai-native userAdd AI features like LLM calls and agents to my generated app without managing provider keys myself

   weight 2 · round to v0
   v0~partialclaimed5/10

   v0's project integrations explicitly include 'AI model providers' alongside databases like Upstash and Neon, and v0 can securely store API keys/tokens for connected services, suggesting some ability to wire in LLM/agent providers without manual key wrangling in code — but v0-docs-44 implies the user still supplies and stores the provider key themselves rather than v0 fully abstracting key management, and there's no dedicated doc on built-in LLM/agent calling with zero-config provider access. missing for 10: explicit documentation of a first-party managed LLM/agent provider (no user-supplied key required), hands-on confirmation that AI features work without configuring provider credentials.

   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase, and AI mod…”
   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase”
   • [claimed-docs] “Securely store sensitive information like API keys, tokens, and other credentials.”
   • [claimed-docs] “v0 plans, creates tasks, and connects to databases as it builds.”

   proof ↗⚑ flag

   Floot—none0/10

   The evidence shows Floot's builder assistant can call tools like generate_image during app creation, but there is no documentation of end-user apps being able to make LLM calls or run agents with Floot managing the provider API keys on the app's behalf.

   ⚑ flag

Auth

1. developerAdd user sign-up and login (email, social providers) scaffolded by the builder

   weight 3 · round to v0
   v0~partialcommunity5/10

   v0 integrates with backend services like Supabase (which typically provide auth) and one community report describes building an app 'with auth, roles' using v0, suggesting the builder can scaffold authentication flows. However, there is no first-party documentation specifically describing sign-up/login scaffolding, social provider integration, or auth templates. Missing for 10: explicit docs on email/social auth scaffolding, named auth provider integrations (e.g., Auth.js, Clerk), and independent verification of social login flows.

   • [community] “Built a feature-rich internal app (time tracker w/ auth, roles, export) and had a working MVP in users' hands by Monday. 'The speed was incr…”
   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase, and AI mod…”
   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase”
   • [claimed-docs] “Connect to backend to build rich, data driven applications.”

   proof ↗⚑ flag

   Floot—none0/10

   The evidence pack covers Floot's database, logging, image generation, mobile app conversion, and self-hosting features, but contains no mention of authentication, user sign-up/login scaffolding, or social provider integration. This is a reasonable capability to expect from a full-stack app builder, so its absence is a gap rather than an inapplicable axis.

   ⚑ flag

Database

1. developerGet a database provisioned and wired into my app without leaving the builder

   weight 3 · round drawn
   v0✓fullclaimed7/10

   v0's Project docs explicitly describe connecting databases (Neon Postgres, Upstash Redis) and backend services (Supabase) directly within a Project, with v0 stated to 'plan, create tasks, and connect to databases as it builds,' and secrets/env vars are managed within the same Project without leaving the tool. This matches the story of provisioning and wiring a database inline in the builder. Missing for 10: no independent/hands-on report confirming the actual provisioning flow works smoothly, and no detail on whether v0 auto-creates the database instance or just connects an existing one.

   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase, and AI mod…”
   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase”
   • [claimed-docs] “Connect your project to third-party services to give your Project access to external data and functionality.”
   • [claimed-docs] “Connect to backend to build rich, data driven applications.”
   • [claimed-docs] “v0 plans, creates tasks, and connects to databases as it builds.”
   • [claimed-docs] “Securely store sensitive information like API keys, tokens, and other credentials.”

   proof ↗⚑ flag

   Floot✓fullprobed7/10

   Floot documents built-in database tooling (query_database, execute_sql, pull_database_schema) accessible directly through the AI assistant inside the builder, indicating a database is provisioned and wired automatically without leaving the chat/builder. Missing for 10: independent/hands-on confirmation of automatic provisioning flow and details on database type/scaling, and no explicit doc showing the provisioning UX step-by-step.

   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “build and ship full-stack apps without leaving your AI chat”
   • [probe] “PROBE llms.txt: HTTP 200 at https://floot.com/llms.txt # Floot > Floot turns a plain-language description into a real, deployed full-stack …”

   proof ↗⚑ flag

Security

1. developerHave the platform review my generated app for security issues before it goes live

   weight 2 · round drawn
   v0—none0/10

   No evidence pack item mentions any security review, vulnerability scanning, or pre-launch security audit feature; deployment features focus on CI/build fixes and hosting, not security review. Missing for 10: any documented security scanning/review capability, evidence of vulnerability detection, or pre-deploy security checklist.

   ⚑ flag

   Floot—none0/10

   No evidence pack item mentions security review, vulnerability scanning, or pre-launch security audits of generated apps; the docs cover logging, database tools, checkpoints, mobile export, and self-hosting but nothing on security review workflows. Missing for 10: any mention of automated security scanning, vulnerability checks, or a review gate before deployment.

   ⚑ flag

Server logic

1. developerAdd server-side functions and third-party integrations like payments and email from the builder

   weight 2 · round to Floot
   v0~partialclaimed5/10

   v0 docs show clear support for connecting projects to backend services and third-party integrations (Supabase, Upstash Redis, Neon Postgres, AI providers) and securely storing API keys/credentials, plus a general 'third-party services for external data and functionality' hook, which covers server-side function-like backend connections. However, no evidence specifically confirms out-of-the-box payments (e.g., Stripe) or email (e.g., Resend/SendGrid) integrations, nor detailed docs on writing custom server-side functions beyond generic 'connect to backend' language. Missing for 10: explicit payment-provider integration examples, explicit email-service integration examples, and documentation of custom server-side function authoring within the builder.

   • [claimed-docs] “Connect to backend to build rich, data driven applications.”
   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase, and AI mod…”
   • [claimed-docs] “Connect your project to third-party services to give your Project access to external data and functionality.”
   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase”
   • [claimed-docs] “Securely store sensitive information like API keys, tokens, and other credentials.”

   proof ↗⚑ flag

   Floot~partialclaimed6/10

   Floot's docs show database read/write tools (query_database, execute_sql), built-in email sending/receiving, and other backend-like capabilities exposed to the AI builder, indicating server-side functionality can be added conversationally. However, there is no explicit mention of general-purpose custom server-side functions or third-party payment integrations, which are core parts of this story. Missing for 10: explicit payments integration support, generic server-side function/webhook creation beyond DB and email, and independent verification of these integrations working in practice.

   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “Your apps can now send AND receive email — no third-party email service needed.”
   • [claimed-docs] “get_logs — the first step for any 500, blank page, or failed request.”

   proof ↗⚑ flag

Export

1. developerExport the full source code and run or host the app outside the platform

   weight 3 · round to Floot
   v0✓fullclaimed7/10

   v0 lets users connect a project to GitHub, push code directly to a repo, import existing GitHub code, and the API platform docs explicitly list 'Export capabilities for backup and migration,' meaning the generated code is not locked into the platform and can be run/hosted elsewhere. This is first-party documented but lacks independent hands-on verification of exporting and self-hosting outside Vercel's infrastructure. missing for 10: independent/community confirmation of exporting and successfully self-hosting a v0 app outside Vercel, and details on the exact export format/process.

   • [claimed-docs] “Connect to GitHub and push code directly to your repository.”
   • [claimed-docs] “Connect a v0 project to GitHub to keep its code in a repository, work on isolated branches, preview changes, run CI, and publish through pul…”
   • [claimed-docs] “To start from code already on GitHub, use Import from GitHub.”
   • [claimed-docs] “Export capabilities for backup and migration”
   • [claimed-docs] “GitHub integration for repository imports”

   proof ↗⚑ flag

   Floot✓fullclaimed8/10

   Floot's docs explicitly state you can export the code and database and deploy them wherever you like, directly matching the story of exporting source and self-hosting outside the platform. Missing for 10: independent/hands-on verification of the export process and details on completeness/portability of exported code beyond the doc claim.

   • [claimed-docs] “you can export the code and the database and deploy them wherever you like.”

   proof ↗⚑ flag

2. developerWork on the app locally with my own editor and tooling and push changes back to the platform

   weight 2 · round to v0
   v0✓fullclaimed7/10

   v0's GitHub integration lets developers connect a project to a repo, work on isolated branches, push code directly, and open PRs — enabling a workflow where code can be edited locally and pushed back to sync with the platform. Missing for 10: explicit mention of a local CLI/dev-server workflow and independent confirmation that locally-pushed commits are correctly re-synced into the v0 chat/build pipeline.

   • [claimed-docs] “Connect a v0 project to GitHub to keep its code in a repository, work on isolated branches, preview changes, run CI, and publish through pul…”
   • [claimed-docs] “To start from code already on GitHub, use Import from GitHub.”
   • [claimed-docs] “To start from code already on GitHub, use Import from GitHub. Importing creates a chat at the selected base branch”
   • [claimed-docs] “Connect to GitHub and push code directly to your repository.”
   • [claimed-docs] “work on isolated branches, preview changes, run CI, and publish through pull requests”
   • [claimed-docs] “CI checks: See whether checks are running, passing, skipped, or failing.”
   • [claimed-docs] “open the branch menu and select Fix Conflicts, Fix CI, or Fix CI + Conflicts.”

   proof ↗⚑ flag

   Floot~partialclaimed5/10

   Floot lets you connect external editors (Cursor, VS Code, Zed) via its remote MCP server, giving access to tools like execute_sql, query_database and create_checkpoint that can modify the live project, and separately lets you export code/database for self-hosting — but there's no documented local file-sync or git-style push workflow; the platform's own framing emphasizes staying in the AI chat rather than local file editing pushed back. missing for 10: explicit local-editor file sync/push mechanism, evidence of two-way git-based workflow, independent confirmation this MCP path is used for real code editing rather than just DB/log operations.

   • [claimed-docs] “Add Floot to Cursor, VS Code, Zed, or any client that supports remote MCP servers”
   • [claimed-docs] “get_logs — the first step for any 500, blank page, or failed request.”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “create_checkpoint labels a point you can revert to from the Checkpoint panel in your project view.”
   • [claimed-docs] “you can export the code and the database and deploy them wherever you like.”
   • [claimed-docs] “build and ship full-stack apps without leaving your AI chat”

   proof ↗⚑ flag

Git sync

1. developerSync my project to my own GitHub repository and keep changes flowing both ways

   weight 3 · round to v0
   v0✓fullclaimed8/10

   v0's docs describe robust two-way GitHub sync: connecting a project to a repo, working on isolated branches, previewing, running CI, publishing via PRs, importing existing GitHub code, and resolving conflicts/CI failures directly from v0 (v0-docs-2,3,26,27,30,39). This covers both directions of the sync story (v0→GitHub via branches/PRs, GitHub→v0 via import) with first-party specificity. Missing for 10: independent/hands-on community verification that the bidirectional sync works reliably in practice beyond vendor docs.

   • [claimed-docs] “Connect a v0 project to GitHub to keep its code in a repository, work on isolated branches, preview changes, run CI, and publish through pul…”
   • [claimed-docs] “To start from code already on GitHub, use Import from GitHub.”
   • [claimed-docs] “open the branch menu and select Fix Conflicts, Fix CI, or Fix CI + Conflicts.”
   • [claimed-docs] “To start from code already on GitHub, use Import from GitHub. Importing creates a chat at the selected base branch”
   • [claimed-docs] “CI checks: See whether checks are running, passing, skipped, or failing.”
   • [claimed-docs] “work on isolated branches, preview changes, run CI, and publish through pull requests”
   • [claimed-docs] “The Duplicate Branch dialog creates a new chat and branch containing the current branch's commits.”

   proof ↗⚑ flag

   Floot—none0/10

   The evidence pack shows Floot supports exporting code/database for self-hosting, but there is no mention of a GitHub integration or bidirectional repo sync anywhere in the docs, changelog, or probes.

   • [claimed-docs] “you can export the code and the database and deploy them wherever you like.”

   proof ↗⚑ flag

Collaboration

1. founderMy team can work on the same project together in a shared workspace

   weight 2 · round to v0
   v0~partialclaimed5/10

   v0 supports Projects with shared deployments, environment variables, and multiple chats connected to one project, plus a Shared Credit Pool for team plans, suggesting basic team collaboration on a shared project. However there's no evidence of real-time co-editing, granular roles/permissions, comments, or explicit multi-user workspace management beyond credit pooling. missing for 10: evidence of real-time multi-user collaboration, role-based permissions, in-app commenting/review, and explicit documentation of team workspace setup beyond credit sharing.

   • [claimed-docs] “One Project shares deployment, hosting, domains, and environment variables for the chats connected to it.”
   • [claimed-docs] “You can connect multiple chats to the same Project. This is useful when working different parts of one Project”
   • [claimed-docs] “When multiple chats share a Project, deploying from any of them updates the same production URL.”
   • [claimed-docs] “Shared Credit Pool credits are shared among all team members”

   proof ↗⚑ flag

   Floot—none0/10

   No evidence in the pack mentions multi-user workspaces, team member invites, roles/permissions, or shared project collaboration features. Evidence focuses on AI chat tooling, MCP integration, mobile app conversion, self-hosting, and checkpoints, none of which speak to teams working together in a shared workspace.

   ⚑ flag

Domains

1. founderConnect a custom domain to my published app

   weight 2 · round to v0
   v0~partialclaimed3/10

   Docs mention that Projects share 'deployment, hosting, domains, and environment variables' implying custom domain support exists at the project level, but there is no explicit guidance, UI flow, or confirmation of connecting a custom domain to a published app. missing for 10: dedicated custom-domain setup docs, DNS/verification steps, hands-on confirmation that a founder can attach their own domain.

   • [claimed-docs] “One Project shares deployment, hosting, domains, and environment variables for the chats connected to it.”
   • [claimed-docs] “Deploy with one click to secure, scalable infrastructure powered by Vercel.”
   • [claimed-docs] “Go live instantly with one-click deployment to production in seconds.”

   proof ↗⚑ flag

   Floot—none0/10

   No evidence pack item mentions custom domain connection, DNS setup, or domain management for published apps; only deployment-adjacent features like self-hosting export and mobile app conversion are documented.

   ⚑ flag

Publish

1. developerDeploy the generated app to an external host or my own infrastructure

   weight 1 · round to Floot
   v0~partialclaimed5/10

   v0 emphasizes one-click deploy to Vercel's own infrastructure and GitHub push/export code, which allows deploying elsewhere via exported code, but there is no first-party support or documentation for direct deployment to non-Vercel hosts (e.g., AWS, Netlify, self-hosted servers) beyond pushing code to a repo. missing for 10: explicit support/instructions for deploying to external hosts or self-managed infrastructure beyond GitHub push, independent confirmation that exported/GitHub code deploys successfully outside Vercel.

   • [claimed-docs] “Deploy with one click to secure, scalable infrastructure powered by Vercel.”
   • [claimed-docs] “Connect to GitHub and push code directly to your repository.”
   • [claimed-docs] “Connect a v0 project to GitHub to keep its code in a repository, work on isolated branches, preview changes, run CI, and publish through pul…”
   • [claimed-docs] “Export capabilities for backup and migration”
   • [claimed-docs] “Go live instantly with one-click deployment to production in seconds.”

   proof ↗⚑ flag

   Floot~partialclaimed6/10

   Floot explicitly documents exporting code and database to self-host or deploy elsewhere, directly supporting this story, but details are thin—no evidence of specific export formats, supported hosts, CI/CD, or independent confirmation of a successful external deployment. missing for 10: detailed export/deploy workflow docs, supported infrastructure targets, independent hands-on verification of external hosting.

   • [claimed-docs] “you can export the code and the database and deploy them wherever you like.”

   proof ↗⚑ flag

2. founderPublish my app to a live URL in one step

   weight 3 · round to v0
   v0✓fullcommunity8/10

   v0 docs explicitly describe one-click deploy to production via Vercel infrastructure, with project-level shared deployment URLs and instant go-live claims, corroborated by a community account of a full dev-to-deploy cycle to real users. Missing for 10: independent third-party verification of the exact one-step publish flow and no hands-on confirmation of custom domain setup ease.

   • [claimed-docs] “Deploy with one click to secure, scalable infrastructure powered by Vercel.”
   • [claimed-docs] “Go live instantly with one-click deployment to production in seconds.”
   • [claimed-docs] “When multiple chats share a Project, deploying from any of them updates the same production URL.”
   • [claimed-docs] “One Project shares deployment, hosting, domains, and environment variables for the chats connected to it.”
   • [community] “Built a feature-rich internal app (time tracker w/ auth, roles, export) and had a working MVP in users' hands by Monday. 'The speed was incr…”

   proof ↗⚑ flag

   Floot~partialprobed6/10

   The llms.txt probe confirms Floot 'turns a plain-language description into a real, deployed full-stack app,' implying deployment is automatic as part of building, and self-hosting docs show apps can be exported/deployed elsewhere. However, there is no explicit documentation of a one-click 'publish' action, custom domain setup, or live URL management workflow. missing for 10: explicit publish/deploy button documentation, custom domain configuration details, confirmation of single-step publishing UX.

   • [probe] “PROBE llms.txt: HTTP 200 at https://floot.com/llms.txt # Floot > Floot turns a plain-language description into a real, deployed full-stack …”
   • [claimed-docs] “you can export the code and the database and deploy them wherever you like.”

   proof ↗⚑ flag

Agent integration

1. ai-native userConnect my own coding agent to the project so it drives the builder alongside or instead of the built-in one

   weight 2 · round to Floot
   v0~partialclaimed3/10

   The v0 API/SDK exposes programmatic access to v0's generation capabilities (e.g. v0.chats.create()) and GitHub integration for code sync, which could let an external tool or script drive v0 programmatically, but there is no evidence of a documented way to plug a third-party coding agent (e.g. Claude Code, Cursor) into a v0 project to replace or run alongside the built-in agent inside the builder UI. Missing for 10: explicit 'bring your own agent' support, agent-swap or multi-agent orchestration inside a v0 project, and any first-party or community confirmation of this working end-to-end.

   • [claimed-docs] “The v0 API gives you programmatic access to all of v0's capabilities, letting you integrate v0's AI-powered code generation, chat interface,…”
   • [claimed-docs] “Custom chat interfaces - Build your own UI for v0's AI-powered code generation”
   • [github] “const response = await v0.chats.create({ message: 'Build me a personal website', })”
   • [github] “Set V0_API_KEY, or deploy server-side code on Vercel with OIDC enabled, then use the default client.”
   • [claimed-docs] “Connect a v0 project to GitHub to keep its code in a repository, work on isolated branches, preview changes, run CI, and publish through pul…”

   proof ↗⚑ flag

   Floot~partialprobed6/10

   Floot documents an official remote MCP server that can be added to Cursor, VS Code, Zed, or any MCP-capable client, letting an external agent drive builder actions like get_logs, query_database, create_checkpoint, and generate_image — this directly supports connecting an outside coding agent to the project. However, evidence doesn't confirm this fully replaces or runs 'alongside' the built-in agent in real workflows, nor is there independent/hands-on confirmation of external agents successfully driving full builds. missing for 10: hands-on/independent verification of external agent parity with built-in agent, clarity on simultaneous or exclusive control, and any limitations when using third-party clients.

   • [claimed-docs] “Add Floot to Cursor, VS Code, Zed, or any client that supports remote MCP servers”
   • [claimed-docs] “get_logs — the first step for any 500, blank page, or failed request.”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “create_checkpoint labels a point you can revert to from the Checkpoint panel in your project view.”
   • [claimed-docs] “generate_image writes AI images into the project.”
   • [probe] “official MCP server documented at https://floot.com/docs/integrations/other-mcp-clients”

   proof ↗⚑ flag

Code editing

1. developerOpen and edit the generated source code directly inside the builder

   weight 2 · round drawn
   v0—none0/10

   The evidence pack describes visual controls, live preview, GitHub push/pull, and export capabilities, but never explicitly confirms an in-builder code editor where developers can directly view and edit the generated source files. Missing for 10: explicit documentation or screenshot of a code/editor pane inside the v0 builder, hands-on confirmation of direct code editing (not just prompting or visual controls).

   • [claimed-docs] “Fine-tune every detail with visual controls and live preview.”
   • [claimed-docs] “Connect to GitHub and push code directly to your repository.”
   • [claimed-docs] “Export capabilities for backup and migration”

   proof ↗⚑ flag

   Floot—none0/10

   Floot's evidence describes chat/AI-driven building, MCP tool integrations, exporting code for self-hosting, and mobile/app features, but there is no mention of an in-builder code editor for directly viewing/editing generated source code within the product itself.

   ⚑ flag

Preview

1. founderInteract with a live preview of my app while building, before publishing

   weight 2 · round to v0
   v0✓fullcommunity8/10

   v0's docs explicitly describe live preview with visual controls while iterating ('Fine-tune every detail with visual controls and live preview') and preview deployments built from working branches before publishing, matching the founder's need to see a live app preview pre-publish. Community evidence corroborates rapid build-to-preview workflows, though it's not preview-specific. Missing for 10: independent hands-on account specifically describing the live-preview interaction (vs generation speed) and detail on preview fidelity/limitations.

   • [claimed-docs] “Fine-tune every detail with visual controls and live preview.”
   • [claimed-docs] “Preview deployments are built from working branches.”
   • [community] “Built a feature-rich internal app (time tracker w/ auth, roles, export) and had a working MVP in users' hands by Monday. 'The speed was incr…”

   proof ↗⚑ flag

   Floot~partialclaimed5/10

   Docs imply an interactive project view where users can 'point at an element or annotate a screenshot' for the AI, and a 'Checkpoint panel in your project view,' suggesting a live, before-publish preview exists — but no evidence explicitly describes a dedicated live-preview pane, its update speed, or founder-facing walkthrough of iterating on it before shipping. missing for 10: explicit live-preview UI documentation, screenshots/demo of pre-publish preview, independent hands-on confirmation.

   • [claimed-docs] “point at an element or annotate a screenshot for the assistant to pick up”
   • [claimed-docs] “create_checkpoint labels a point you can revert to from the Checkpoint panel in your project view.”
   • [claimed-docs] “build and ship full-stack apps without leaving your AI chat”

   proof ↗⚑ flag

Prompt iteration

1. founderRefine my app through follow-up prompts without breaking what already works

   weight 3 · round drawn
   v0~partialcommunity5/10

   v0 supports iterative refinement (chat-based prompts, automatic error fixing, isolated branches, project structure preserving deployment/config) and community reports confirm fast, incremental build-to-deploy cycles, but no evidence explicitly addresses preserving existing functionality across follow-up prompts or preventing regressions. missing for 10: explicit regression-safety guarantees, hands-on evidence that follow-up prompts don't break prior working features, independent testing of iterative stability.

   • [claimed-docs] “Automatically fix errors in your code with intelligent diagnostics.”
   • [claimed-docs] “v0 reads the relevant logs, diagnoses the failure, and applies a fix.”
   • [claimed-docs] “You can connect multiple chats to the same Project. This is useful when working different parts of one Project”
   • [claimed-docs] “One Project shares deployment, hosting, domains, and environment variables for the chats connected to it.”
   • [community] “Built a feature-rich internal app (time tracker w/ auth, roles, export) and had a working MVP in users' hands by Monday. 'The speed was incr…”

   proof ↗⚑ flag

   Floot~partialclaimed5/10

   Floot supports checkpoint/revert (floot-docs-5) and chat-based iterative building (floot-docs-1), which are relevant to safely refining an app via follow-up prompts, but there's no direct evidence describing regression prevention, test coverage, or how the assistant avoids breaking existing functionality during edits. missing for 10: evidence of automated regression/testing safeguards, independent hands-on account of iterative prompting not breaking prior features, and detail on how conflicts between old and new instructions are resolved.

   • [claimed-docs] “build and ship full-stack apps without leaving your AI chat”
   • [claimed-docs] “create_checkpoint labels a point you can revert to from the Checkpoint panel in your project view.”

   proof ↗⚑ flag

Visual editing

1. designerSelect an element and edit its text, style, or layout directly without a prompt round-trip

   weight 2 · round to v0
   v0~partialclaimed3/10

   v0.app claims 'Fine-tune every detail with visual controls and live preview' (v0-docs-7), suggesting some direct-manipulation UI, but nearly all other documentation describes chat/prompt-driven generation and iteration cycles rather than click-to-select element editing. Missing for 10: explicit documentation or demo of selecting an element on canvas and editing its text/style/layout inline, independent corroboration that this bypasses the prompt loop, and detail on what 'visual controls' actually cover.

   • [claimed-docs] “Fine-tune every detail with visual controls and live preview.”

   proof ↗⚑ flag

   Floot—none0/10

   Floot's documented workflow is prompt/chat-driven — you 'point at an element or annotate a screenshot for the assistant to pick up' (floot-docs-6), which still requires an AI round-trip rather than direct in-canvas editing of text, style, or layout. No evidence describes a direct-manipulation editor mode for designers to bypass the AI prompt loop.

   • [claimed-docs] “point at an element or annotate a screenshot for the assistant to pick up”
   • [claimed-docs] “build and ship full-stack apps without leaving your AI chat”

   proof ↗⚑ flag

1. ai-native userDo everything through the API that I can do in the UI

   weight 2 · round drawn
   v0~partialclaimed5/10

   v0 explicitly claims the API gives 'programmatic access to all of v0's capabilities' including chat, project management, and deployment (v0-docs-4), backed by SDK examples (v0-gh-1/2/3) and API docs on GitHub import and export (v0-docs-37, v0-docs-43). However, many UI-specific workflows (Figma/screenshot cloning, CI/conflict-fix buttons, command palette search, iOS app, visual style controls) are only documented as UI features with no API equivalent shown, and no independent test confirms true 1:1 parity. missing for 10: API-exposed equivalents for design import (Figma/screenshots), GitHub PR/CI-fix actions, and independent verification that API surface matches UI feature set.

   • [claimed-docs] “The v0 API gives you programmatic access to all of v0's capabilities, letting you integrate v0's AI-powered code generation, chat interface,…”
   • [claimed-docs] “GitHub integration for repository imports”
   • [claimed-docs] “Export capabilities for backup and migration”
   • [github] “const response = await v0.chats.create({ message: 'Build me a personal website', })”
   • [github] “Set V0_API_KEY, or deploy server-side code on Vercel with OIDC enabled, then use the default client.”
   • [claimed-docs] “Custom chat interfaces - Build your own UI for v0's AI-powered code generation”
   • [claimed-docs] “Clone pages with screenshots or Figma files”
   • [claimed-docs] “open the branch menu and select Fix Conflicts, Fix CI, or Fix CI + Conflicts.”

   proof ↗⚑ flag

   Floot~partialprobed5/10

   Floot exposes a documented MCP server with tools (get_logs, query_database/execute_sql, create_checkpoint, generate_image) that let an AI agent perform many core dev actions, mirroring parts of the UI workflow, but there's no public REST/OpenAPI API (probes return 404 for openapi.json/swagger.json and docs.md), so broader UI features like mobile app packaging, push notifications, email, analytics, and self-hosting export are not confirmed to be reachable via API. missing for 10: a general-purpose REST/OpenAPI API, evidence that all UI-exposed features (mobile builds, notifications, analytics, hosting/export) are also API-accessible, independent confirmation of API parity.

   • [claimed-docs] “Add Floot to Cursor, VS Code, Zed, or any client that supports remote MCP servers”
   • [claimed-docs] “get_logs — the first step for any 500, blank page, or failed request.”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “create_checkpoint labels a point you can revert to from the Checkpoint panel in your project view.”
   • [claimed-docs] “generate_image writes AI images into the project.”
   • [probe] “PROBE docs-md: HTTP 404 at https://floot.com/docs.md”
   • [probe] “PROBE openapi: all candidate paths 404 (https://floot.com/openapi.json, https://floot.com/swagger.json, https://floot.com/api/openapi.json, …”
   • [probe] “official MCP server documented at https://floot.com/docs/integrations/other-mcp-clients”

   proof ↗⚑ flag

2. ai-native userExport all of my data in open formats and leave

   weight 3 · round to Floot
   v0~partialclaimed5/10

   v0 lets users push/export generated code to GitHub in a standard, open format (docs-2, docs-31, docs-39), and the API docs mention generic 'Export capabilities for backup and migration' (docs-43), supporting a degree of data portability. However there's no concrete evidence of exporting full account data (chat history, project settings, credentials) in open formats, and no independent confirmation the export path is complete or lock-in-free. Missing for 10: evidence of exporting non-code data (chats, project metadata, integrations config), a documented full-account export/download feature, and independent verification that migrated projects run outside v0/Vercel without further dependency.

   • [claimed-docs] “Connect a v0 project to GitHub to keep its code in a repository, work on isolated branches, preview changes, run CI, and publish through pul…”
   • [claimed-docs] “Connect to GitHub and push code directly to your repository.”
   • [claimed-docs] “work on isolated branches, preview changes, run CI, and publish through pull requests”
   • [claimed-docs] “Export capabilities for backup and migration”

   proof ↗⚑ flag

   Floot~partialclaimed6/10

   Floot documents exporting both code and database for self-hosting elsewhere, directly supporting data portability and leaving the platform, but there's no evidence of a broader open-format export (e.g., structured data exports, account data, media assets) beyond code/DB, nor independent confirmation the export is complete or in fully open standard formats. missing for 10: evidence of full data export scope (assets, user data, configs), independent/hands-on verification of export completeness, and confirmation of open standard formats used.

   • [claimed-docs] “you can export the code and the database and deploy them wherever you like.”

   proof ↗⚑ flag

3. ai-native userRead the product's source under an open license

   weight 2 · round drawn
   v0—none0/10

   v0 is a closed, proprietary hosted product (with an SDK client library on GitHub, not the core product source); there is no evidence the v0 application/model itself is released under an open license. missing for 10: any statement of an open-source license for v0's core product/source code.

   ⚑ flag

   Floot—none0/10

   Floot is a closed proprietary SaaS platform for generating apps; there is no evidence of Floot's own source code being available under an open license. Evidence only shows users can export generated app code (floot-docs-11), which is not the same as the platform's source being open.

   ⚑ flag

Pricing

1. founderUnderstand what credits or tokens each action consumes and what my plan includes

   weight 2 · round to v0
   v0!disputedcontradicted5/10

   Docs describe a credits/plan system (Premium $20/month, shared credit pools, ability to buy more credits) and even flag one action as token-free (v0-docs-10, v0-docs-23, v0-docs-42, v0-docs-25), but no documentation breaks down per-action credit costs. A hands-on community report contradicts the idea that users can predict consumption, describing unexpectedly high usage-based costs (~$50 in a weekend, '>$20/day') versus flat subscription alternatives (v0-comm-3), showing the credit model is opaque in practice. missing for 10: a clear per-action credit/token cost table, first-party UI showing real-time credit usage, and corroborating community confirmation that costs are predictable.

   • [claimed-docs] “You can buy more credits any time (Premium, Plus, and Business plans).”
   • [claimed-docs] “Premium: $20/month. For higher limits and power users.”
   • [claimed-docs] “Shared Credit Pool credits are shared among all team members”
   • [claimed-docs] “Start with existing files (fast, no tokens)”
   • [community] “Cost complaint: 'V0 is pay by usage... we talking > +$20/day vs $XX/month' for a subscription-based alternative like Copilot; spent about $5…”

   proof ↗⚑ flag

   Floot—none0/10

   Only floot-docs-8 mentions a credit system (you can 'add credits whenever you need them'), but there is no documentation of what specific actions consume credits/tokens, how many, or what different plans include — the core of the founder's ask is unaddressed.

   • [claimed-docs] “you no longer have to upgrade or downgrade your plan — just add credits whenever you need them.”

   proof ↗⚑ flag

1. ai-native userChoose where my data is stored (region/residency)

   weight 2 · round drawn
   v0—none0/10

   No evidence in the pack mentions data residency, regional storage options, or any data-location controls for v0; the docs focus on code generation, deployment, GitHub integration, and pricing, none of which address where data is stored.

   ⚑ flag

   Floot—none0/10

   No evidence anywhere in the pack mentions data residency, region selection, or hosting location choices; only self-hosting/export is mentioned which does not equate to choosing a storage region within Floot's platform.

   ⚑ flag

2. ai-native userPrevent my data from being used to train AI models

   weight 3 · round drawn
   v0—none0/10

   No evidence in the pack addresses data-training opt-out, privacy controls, or any statement about whether user data/code is used to train AI models; nothing mentions a privacy policy toggle or enterprise no-training guarantee.

   ⚑ flag

   Floot—none0/10

   No evidence pack item mentions AI training data usage, opt-out controls, or data-privacy commitments regarding model training; nothing addresses this axis at all.

   ⚑ flag

3. ai-native userControl data retention and deletion

   weight 2 · round drawn
   v0—none0/10

   No evidence in the pack addresses data retention policies, deletion controls, or privacy settings for user data/chats/projects in v0; only export capabilities for backup/migration are mentioned, which is unrelated to retention/deletion control. Missing for 10: documented data retention policy, user-facing deletion/export-of-data controls, opt-out of training data usage, any privacy/compliance documentation.

   ⚑ flag

   Floot—none0/10

   No evidence pack items describe data retention controls, deletion policies, or privacy/export-of-personal-data mechanisms; export/self-hosting docs cover code/database portability, not retention or deletion settings. missing for 10: retention policy documentation, data deletion controls/API, privacy policy references, account/data purge mechanisms.

   ⚑ flag

4. ai-native userOpt out of telemetry and usage tracking

   weight 2 · round drawn
   v0—none0/10

   No evidence pack item mentions telemetry, analytics opt-out, or usage tracking controls for v0; nothing addresses privacy settings of this kind.

   ⚑ flag

   Floot—none0/10

   The evidence pack contains no mention of telemetry opt-out or privacy controls for AI-native usage; it only shows built-in visitor tracking for end-user apps (floot-docs-10), which is unrelated to Floot's own telemetry practices toward its users.

   ⚑ flag

Design import

1. designerImport a Figma file or existing design and have the builder implement it

   weight 2 · round to v0
   v0~partialclaimed6/10

   Docs explicitly claim ability to 'Clone pages with screenshots or Figma files' and 'Create high-fidelity UIs from your wireframes or mockups,' directly supporting the story, but this is first-party doc text only with no workflow detail, screenshots, or independent hands-on confirmation of Figma import fidelity. Missing for 10: detailed Figma import workflow documentation, independent/community verification that Figma import works as claimed, and coverage of fidelity/limitations when translating Figma designs to code.

   • [claimed-docs] “Clone pages with screenshots or Figma files”
   • [claimed-docs] “Create high-fidelity UIs from your wireframes or mockups.”
   • [claimed-docs] “Turn mockups into real, high-fidelity user interfaces that reflect user flows and constraints.”

   proof ↗⚑ flag

   Floot—none0/10

   No evidence pack item mentions Figma import, design file import, or converting existing designs into a Floot app; the docs focus on chat-based prompting, MCP tools, mobile export, and self-hosting.

   ⚑ flag

Generation

1. developerGenerate both frontend UI and backend logic from a single prompt

   weight 3 · round drawn
   v0✓fullcommunity8/10

   v0 docs and community evidence show single-prompt generation of both frontend UI and connected backend logic (databases, auth, Supabase/Neon/Upstash integrations) plus full-stack scaffolding and deployment, corroborated by a hands-on HN report of building a full app with auth/roles/export from prompt to deployed MVP. missing for 10: no controlled benchmark of backend logic correctness/robustness, and no independent review specifically validating backend code quality (community feedback focuses mostly on UI generation and cost).

   • [claimed-docs] “Quickly scaffold full-stack apps or components following best practices and modern standards.”
   • [claimed-docs] “Connect to backend to build rich, data driven applications.”
   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase, and AI mod…”
   • [claimed-docs] “v0 plans, creates tasks, and connects to databases as it builds.”
   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase”
   • [community] “Built a feature-rich internal app (time tracker w/ auth, roles, export) and had a working MVP in users' hands by Monday. 'The speed was incr…”

   proof ↗⚑ flag

   Floot✓fullprobed8/10

   Floot's own docs describe it as turning a plain-language description into a full-stack app with React+TypeScript frontend and backed by database tools (query_database, execute_sql, pull_database_schema), indicating both UI and backend logic are generated from a single prompt. This is corroborated by the llms.txt probe confirming the core positioning. Missing for 10: independent/hands-on developer confirmation of single-prompt generation quality and no explicit example walkthrough of a prompt producing both layers together.

   • [probe] “PROBE llms.txt: HTTP 200 at https://floot.com/llms.txt # Floot > Floot turns a plain-language description into a real, deployed full-stack …”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “build and ship full-stack apps without leaving your AI chat”

   proof ↗⚑ flag

2. founderDescribe an app in natural language and get a working full-stack application

   weight 3 · round drawn
   v0✓fullprobed8/10

   Docs and API show natural-language-to-app generation, full-stack scaffolding, backend/database integrations, and one-click deployment (v0-docs-15, v0-docs-24, v0-docs-28, v0-docs-32, v0-gh-1). A first-hand community report confirms a working MVP built and deployed quickly from a prompt (v0-comm-1). Missing for 10: broader independent benchmarking on complex, non-trivial apps and evidence addressing the noted design-sameness limitation (v0-comm-4).

   • [claimed-docs] “Quickly scaffold full-stack apps or components following best practices and modern standards.”
   • [claimed-docs] “Connect to backend to build rich, data driven applications.”
   • [claimed-docs] “Available integrations include databases like Upstash for Redis and Neon for Serverless Postgres, backend services like Supabase, and AI mod…”
   • [claimed-docs] “Go live instantly with one-click deployment to production in seconds.”
   • [github] “const response = await v0.chats.create({ message: 'Build me a personal website', })”
   • [community] “Built a feature-rich internal app (time tracker w/ auth, roles, export) and had a working MVP in users' hands by Monday. 'The speed was incr…”
   • [probe] “PROBE docs-md: HTTP 200 at https://v0.app/docs.md --- title: What is v0? description: v0 is an AI agent that helps anyone create real code a…”

   proof ↗⚑ flag

   Floot✓fullprobed8/10

   Floot's own docs describe turning plain-language descriptions into a deployed full-stack app (React+TypeScript frontend, database, backend tooling like query_database/execute_sql, image generation, mobile app conversion), directly matching the prompt-to-app story, and this is corroborated by a live probe of llms.txt confirming the same positioning. Missing for 10: independent/third-party hands-on reviews validating real-world output quality beyond vendor docs and probes.

   • [probe] “PROBE llms.txt: HTTP 200 at https://floot.com/llms.txt # Floot > Floot turns a plain-language description into a real, deployed full-stack …”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “Floot converts your project into native iOS and Android apps”
   • [claimed-docs] “get_logs — the first step for any 500, blank page, or failed request.”

   proof ↗⚑ flag

Programmatic generation

1. ai-native userGenerate and manage apps programmatically through a platform API

   weight 3 · round to v0
   v0✓fullclaimed8/10

   v0 documents a platform API that gives programmatic access to chat/code generation, project management, and deployment (v0-docs-4), and the official v0-sdk on GitHub shows concrete code for creating chats and streaming results with API-key auth (v0-gh-1, v0-gh-2, v0-gh-3), directly matching the 'generate and manage apps programmatically' story. Missing for 10: independent/hands-on corroboration of the API in production use and more detail on full app lifecycle management (list/delete/update) via the API beyond chat creation.

   • [claimed-docs] “The v0 API gives you programmatic access to all of v0's capabilities, letting you integrate v0's AI-powered code generation, chat interface,…”
   • [claimed-docs] “Custom chat interfaces - Build your own UI for v0's AI-powered code generation”
   • [claimed-docs] “Start with existing files (fast, no tokens)”
   • [claimed-docs] “GitHub integration for repository imports”
   • [claimed-docs] “Export capabilities for backup and migration”
   • [github] “const response = await v0.chats.create({ message: 'Build me a personal website', })”
   • [github] “const result = readV0Stream(serverResult.toResponse())”
   • [github] “Set V0_API_KEY, or deploy server-side code on Vercel with OIDC enabled, then use the default client.”

   proof ↗⚑ flag

   Floot~partialprobed6/10

   Floot exposes a remote MCP server with tools (get_logs, query_database, execute_sql, generate_image, create_checkpoint) that let an AI client programmatically manage an app's data, logs, and images, which is a form of programmatic control aligned with ai-native/prompt-to-app usage. However, probes show no REST/OpenAPI API (all openapi.json/docs.md paths 404), and there's no evidence of programmatically creating new apps from scratch via API—only managing an existing project through MCP tools. Missing for 10: a documented REST/OpenAPI platform API, evidence of app creation (not just management) via API/MCP, and independent confirmation of these tool calls working end-to-end.

   • [claimed-docs] “get_logs — the first step for any 500, blank page, or failed request.”
   • [claimed-docs] “query_database for reads, execute_sql for writes and migrations, pull_database_schema to regenerate typed helpers.”
   • [claimed-docs] “create_checkpoint labels a point you can revert to from the Checkpoint panel in your project view.”
   • [claimed-docs] “generate_image writes AI images into the project.”
   • [probe] “PROBE openapi: all candidate paths 404 (https://floot.com/openapi.json, https://floot.com/swagger.json, https://floot.com/api/openapi.json, …”
   • [probe] “official MCP server documented at https://floot.com/docs/integrations/other-mcp-clients”

   proof ↗⚑ flag

Templates

1. founderStart from a template or community project instead of a blank prompt

   weight 2 · round to v0
   v0~partialcommunity4/10

   v0 supports starting from existing code via 'Import from GitHub' and 'Start with existing files (no tokens)', which lets founders begin from a pre-existing project rather than a blank prompt, and community testimony confirms it solves the 'blank canvas problem'. However there is no evidence of a curated template gallery or a community-project showcase/marketplace to browse and remix, which is the core of the story. Missing for 10: a dedicated template/starter gallery, community project discovery or remix feature, and evidence of founders actually starting from such templates.

   • [claimed-docs] “To start from code already on GitHub, use Import from GitHub.”
   • [claimed-docs] “To start from code already on GitHub, use Import from GitHub. Importing creates a chat at the selected base branch”
   • [claimed-docs] “Start with existing files (fast, no tokens)”
   • [community] “Built a feature-rich internal app (time tracker w/ auth, roles, export) and had a working MVP in users' hands by Monday. 'The speed was incr…”

   proof ↗⚑ flag

   Floot—none0/10

   No evidence in the pack mentions templates, starter kits, or community project galleries as a starting point for new Floot projects; all docs focus on chat-based generation, MCP integrations, and infrastructure features.

   ⚑ flag