Skip to main content

Choose your path

Start with the path that matches what you are trying to do. You can change paths later because the console and coding-agent flow write to the same project state.

I built an app and want to deploy it​

Use the web console.

You will connect the repository, answer product questions, review missing backend and security work, compare providers, and approve deploy only when the evidence is good enough.

Start here:

  1. Open passo.build.
  2. Choose the repository.
  3. Complete product scope before running security and deploy checks.

Read next: Console flow.

I want my coding agent to drive the flow​

Use passo.build inside your coding agent.

Your agent scans the repo locally, asks the missing questions in its own chat, and submits structured findings to passo.build. The agent can then implement the missing tasks while passo.build tracks state, locks, and evidence.

Start here:

  1. Add the passo.build MCP server to the agent.
  2. Authenticate or use the provisioned rollout token for your workspace.
  3. Ask the agent to start the passo.build assessment for the current repo.

Read next: Use passo.build inside your coding agent.

I am reviewing whether the plan is good​

Use the docs as a review checklist.

The most important rule is simple: the visible app, private APIs, database, storage, jobs, auth, secrets, and deploy automation must each have a clear owner and evidence.

Review these pages:

I am integrating MCP or an agent host​

Use the reference layer.

The MCP endpoint and protocol are for agent hosts and platform integrators. Normal users should not need to memorize tool names or hand-edit bearer tokens once local login is available for their host.

Read next:

I am coordinating multiple agents​

Use Agent Mesh.

Agent Mesh is the coordination layer for tasks, locks, evidence, leases, working memory, and handoff. It matters when more than one agent works on the same product or when long tasks must survive context resets.

Read next: Agent Mesh working memory.