
Kadmo
kadmo.aiAgentic engineering platform with role-based fleets
Run Claude Code and Codex agents with roles, playbooks, and gates on your own cloud.

Overview
Kadmo is a sovereign agentic engineering platform that runs AI coding agents like Claude Code and Codex with defined roles, playbooks, and gates on your own infrastructure. It orchestrates a fleet of agents—Software Engineer, Product Manager, QA, Security, Documentation, and more—to handle tickets end-to-end, from planning to delivery, with human approval at key steps. Built for engineering teams that want to scale output while keeping control over code, models, and data, Kadmo integrates with Jira, Linear, GitHub, GitLab, Slack, and Teams.
How it works
You drop a ticket into your task board, and the fleet takes it from there. A Product Manager agent turns the goal into an epic and splits it into tickets sized for one agent each. A Software Engineer works on its own machine and branch, opens a pull request, and fixes what review sends back. A QA Engineer runs your test suite on the live environment after every change. Every step leaves a comment in the ticket explaining what it did and why, and gates at every handoff ensure nothing moves forward until it passes.
Key capabilities
Kadmo includes ready-made agents for Software Engineering, Product Management, QA, Security, Documentation, Skill Discovery, and Website Building. Playbooks define fixed processes with approval gates, and agents run on your cloud with scoped credentials and allowlisted egress. Governance features include roles with scoped tools, budgets with spend limits, data residency enforcement, and full run transcripts. The platform supports Claude Code and Codex on models from Anthropic, OpenAI, and Google, with no lock-in.
Who it is for
Kadmo is for engineering teams that want to increase output without sacrificing quality or control. It is designed for teams already using Jira or Linear and GitHub or GitLab, and it works inside those tools with nothing to replace or migrate. The platform is especially suited for organizations with strict data residency or security requirements, as agents run on your own cloud and model calls go through your own gateway. It is also for teams that want to scale with AI while keeping human oversight through approval gates.
Pricing and limits
Kadmo offers a free 15-day trial with €5 of model credit on economy models. The Business plan is flat monthly per workspace with unlimited seats, up to 100 agents on your cloud, 100K runs a month, every model at Kadmo prices, and governance included. The Enterprise plan offers custom terms with your own keys or gateway, more than 100 agents by agreement, priority support, and SLA. There are no run packages; you pay for the tokens the models use and the compute of your own cloud.
Key features
Role-based agents
Agents take on roles like Software Engineer, Product Manager, QA Engineer, Security, Documentation, Skill Discovery, and Website Builder, each with scoped tools and credentials.
Playbooks with gates
Fixed processes define which role takes which step, with approval gates at every handoff to ensure quality.
Runs on your infrastructure
Agents run on your cloud (AWS, Google Cloud, Hetzner) with your own model gateway, keeping code within your perimeter.
Model flexibility
Supports Claude Code and Codex on models from Anthropic, OpenAI, and Google, with no lock-in.
Governance and audit
Includes roles with scoped tools, approval policies, budgets, data residency enforcement, and full run transcripts.
Integrations
Works inside Jira, Linear, GitHub, GitLab, Slack, and Teams.
Use cases
Feature implementation
Hand a ticket to the fleet; a Product Manager plans, a Software Engineer builds, a reviewer checks, and QA tests.
Bug fixes
Drop a bug ticket and let the fleet handle the fix, with regression testing on the live site.
Documentation updates
The Documentation agent reads shipped code, updates docs, and drafts customer updates.
Security scans
The Security agent runs nightly scans over dependencies, secrets, and configuration, filing tickets for fixes.
Website building
The Website Builder creates pages from a brief, with layout, copy, and components, published after approval.
Pricing & limits
From the official site · checked 10/30/2026Frequently asked questions about Kadmo
- What is Kadmo?
- Kadmo is a sovereign agentic engineering platform that runs AI coding agents like Claude Code and Codex with defined roles, playbooks, and gates on your own infrastructure.
- Who is Kadmo for?
- It is for engineering teams that want to scale output with AI while keeping control over code, models, and data.
- Is Kadmo free?
- Kadmo offers a free 15-day trial with €5 of model credit. After the trial, you choose a paid Business or Enterprise plan.
- Which platforms does Kadmo support?
- Kadmo integrates with Jira, Linear, GitHub, GitLab, Slack, and Teams, and runs on AWS, Google Cloud, or Hetzner.
- How does Kadmo ensure security?
- Agents run on your cloud with scoped credentials, allowlisted egress, and full audit trails. Governance features include budgets, data residency policies, and SAML SSO.
Related tools

Launchdog.dev
Launch your product on a weekly board ranked by human votes, with your AI agent handling the listing.

GPTZZZ
One account to access Claude, GPT, Grok, DeepSeek, and GLM models through a single API.

SightRadar
Rekognition-compatible face recognition API: index, search, compare and detect faces, pay per photo.

PineGen AI
Generate, debug, and backtest TradingView-ready Pine Script v6 strategies with AI.

AICost
Independent LLM API price desk: compare model pricing across OpenAI, Anthropic, Google and more.

Code Guardian
Enforce quality gates and safety latches on AI-generated code in Claude Code and Codex.

Wallarm
Discover, observe, enforce, and govern AI workloads on AWS with kernel-level visibility and no code changes.

Kintsu.ai
Edit your WordPress site by describing changes in plain English, with safe previews and one-click publishing.