Skip to content
English
Sign in Book a meeting
Platform Overview

Platform overview

Kadmo is an AI coding agent platform: a fleet of agents in the roles you define, working from your tickets, in your tools, by your rules. This page shows how the parts fit together. Each has a page of its own, and the docs hold the setup steps.

On this page 5 sections

How the parts fit together

You drop a ticket from your task board, or name a goal and let an agent plan the tickets. A playbook gives each step to an AI coding agent in its role, each on its own machine, with a gate at every handoff.

Playbook · Feature implementation

One of the playbooks that come with Kadmo. At Review the run waits for you, and a failed review sends the work back to Build.

Agents work with your Claude Code skills, your MCP servers and your repositories. Every step reports in the ticket, what it did and why, and the run keeps its full transcript.

Run the work

What gets done, in which order, and by which role.

  • Goals

    Name an outcome. Kadmo writes the epic, an agent plans it into tickets, and they run wave by wave.

  • Playbooks

    The steps for each kind of ticket, each with a role, and a gate at every handoff.

  • AI coding agents

    Agents in the roles you define, each on its own machine, on Kadmo or on yours.

  • Ask Kadmo

    Answers from your account, with sources. A request for work comes back as a plan.

A goal named Checkout in one page on its Build step: its owner, Decide skipped, the plan at four tasks in three waves, one of four tasks done and three to start.
A goal on Build: planned into four tasks in three waves, one done and three to start.

Equip the agents

What agents know about your product, and what they can reach in your tools.

  • Claude Code skills

    Your account’s own marketplace: agents install it and developers connect to it.

  • Skill packs

    Knowledge the runtime serves to every run, and what one run learned for the next.

  • MCP servers

    Kadmo builds and runs your MCP server from its repository, behind your account’s sign-in.

  • Integrations

    Jira, Linear, Notion, GitHub, GitLab, Slack and Teams, with results back in the ticket.

A hosted MCP server named orders: its address on mcp.kadmo.ai, the state Live serving one commit, the branch that deploys it, and who may use it.
One of your own MCP servers: its address, the commit it serves, and who may use it.

Keep control

What an admin sets once: where agents run, what they may spend, and what comes back.

  • Self-hosted agents

    Inside your perimeter: one outbound connection and no inbound port, on Ubuntu, Docker or Kubernetes.

  • Guardrails

    Budgets with a ceiling, the regions and models you allow, and rules for what chats may send.

  • Models and keys

    A model per step, on your Kadmo credit or your own keys, with no lock in.

  • Reporting

    The report on the ticket, the notice in your chat, and a video of the job when you ask.

The Guardrails page: data residency for the whole account and for a workspace, by EU, US and Global, and the models switched on for every lane.
Where your account’s models may run, and which of them are switched on.

Where to go next