Agentic Product Delivery Platform

Turn AI-generated code into shippable products

Agent-Dev manages the full lifecycle from requirement to production. You own the product decisions; Agent-Dev turns them into a shippable, verifiable, maintainable engineering system.

Golden Path

The complete pipeline from requirement to delivery report

  1. Requirement

    Describe goal and scope in natural language

  2. Clarification

    Ask only about real uncertain decisions

  3. Spec

    Acceptance criteria and implementation tasks

  4. Codex implements

    Execute in isolated worktree

  5. Local verify

    lint / typecheck / unit / build

  6. PR

    Create PR with quality gates

  7. Vercel API Preview

    Deploy API first

  8. Cloudflare Pages Preview

    Deploy frontend after getting API URL

  9. Joint smoke test

    CORS / Auth / env var validation

  10. Manual acceptance

    Preview Gate passed

  11. Delivery report

    Evidence chain and residual risks

Why Agent-Dev

Not a faster code generator, but accountability for the full delivery

User-owned

Code, repos, data, domains, cloud resources, and Blueprint belong to the user. After leaving Agent-Dev, the product still builds, deploys, and maintains independently.

Neutral and portable

Models and providers are replaceable. No lock-in to a specific Agent Runtime or cloud platform. Switching databases or deploy providers is a controlled migration, not a setting change.

Real gates

An LLM may propose, but cannot obtain production permission via prompt. Deterministic Policy, GitHub Rulesets, and Environment Approval are the final constraints.

Evidence-based delivery

Delivery status is grounded in GitHub Checks, Deployment Records, Preview URLs, database state, smoke tests, and manual acceptance, not the Agent's natural-language summary.

Compared to adjacent products

Agent-Dev doesn't compete on who generates pages faster

Agent-Dev vs Replit Agent, Lovable, GitHub Spec Kit
DimensionAgent-DevReplit AgentLovableGitHub Spec Kit
Unit managedProduct lifecycleSingle sessionSingle generationSpecs and tasks
User-owned resourcesYes, own GitHub/cloudNo, Replit-hostedPartialYes
Replaceable AgentYesNoNoPartial
Portable providersYes, controlled migrationNoNoN/A
Production permissionPolicy + manual gatePlatform-managedPlatform-managedManual gate
Delivery evidenceExternal fact chainPlatform logsPreviewTask status
Long-term maintenanceDrift detection + fix policyLimitedLimitedN/A

Responsibility Boundary

Transparent split between user and platform builds trust

User is responsible for

  • Who the product serves and what problem it solves
  • Feature scope, experience, aesthetics, and business priority
  • Cost, privacy, data, and external-impact decisions
  • Preview experience acceptance
  • Production release and final Approval

Agent-Dev is responsible for

  • Provide verified product Blueprints and default standards
  • Turn requirements into specs, acceptance criteria, and tasks
  • Call replaceable Agent Runtimes to implement, test, and fix code
  • Orchestrate GitHub, Supabase, Cloudflare, Vercel, and more
  • Enforce deterministic Policy to constrain Agent permissions
  • Aggregate real CI, deploy, test, and manual-acceptance evidence
  • Detect drift, expired config, and maintenance tasks
  • Report clear delivery status and residual risks

Agent-Dev guarantees process and evidence integrity, not market success.

Product Constitution Excerpts

Core principles that constrain platform behavior

01

Defaults over choices

Beginner mode decides ~90% of engineering questions for the user, asking only about product, cost, privacy, or risk. Options are escape hatches, not the default experience.

02

AI reasons, Policy authorizes

An LLM may propose, but cannot obtain production permission via prompt. Deterministic Policy and provider permissions are the final constraints.

03

External facts over Agent self-report

Delivery status is grounded in GitHub Checks, Deployment Records, Preview URLs, and smoke tests, not the Agent's natural-language summary.

04

User owns the product

Code, repos, data, domains, cloud resources, and Blueprint belong to the user. Models and providers are replaceable.

Frequently Asked Questions

Key questions about ownership, security, and automation

What is Agent-Dev?

Agent-Dev is an Agentic Product Delivery Platform. It is not a replacement for Codex or Claude Code, nor a source-code-only app builder. It manages the full product lifecycle from requirement to production.

Who owns my code and cloud resources?

Code, repos, data, domains, cloud resources, and Blueprint all belong to the user. After leaving Agent-Dev, the product still builds, deploys, and maintains independently. Models and providers are replaceable.

How does Agent-Dev control AI production permissions?

An LLM may propose, but cannot obtain production permission via prompt. Deterministic Policy, GitHub Rulesets, Environment Approval, and provider permissions are the final constraints. Autonomy scales per project and per action; there is no single risk-blind "full auto" switch.

How is delivery actually verified as complete?

Delivery status is grounded in GitHub Checks, Deployment Records, Preview URLs, database state, smoke tests, and manual acceptance, not the Agent's natural-language summary. Unexecuted verifications cannot be marked as passed.

How is Agent-Dev different from Replit Agent or Lovable?

Replit Agent and Lovable focus on conversational generation and one-click hosting on platform-owned resources. Agent-Dev's minimum unit is the product lifecycle: it uses user-owned GitHub/cloud resources, with replaceable Agents and providers, and is accountable for full delivery and long-term maintenance.

What tech stack does the first version support?

The first version fixes a Web SaaS Golden Path: React + Vite + TypeScript frontend, Hono API, Supabase DB and auth, Cloudflare Pages for frontend hosting, Vercel Functions for API hosting, GitHub Actions for CI, and local Codex as the Agent Runtime.

Start your first product delivery

Get a Web product baseline you own in 60 minutes