Login or sign up
For business
Average ratingGreat
4.2/ 5

Based on 5 reviews

540%
440%
320%
20%
10%
Ask AI

Questions about Cat Factory

Based on 5 reviews, listing data and website checks · updated 8 days ago

What is Cat Factory?

Cat Factory is a self-hosted, open-source software tool that coordinates LLM-based coding agents to turn tickets and plans into reviewed pull requests. It combines a visual planning board with a live pipeline view, showing in-flight runs, failures, and items awaiting human approval. The tool is driven through the board, headlessly over a versioned public API with SDKs in four languages, or self-hosted under an MIT license.

How does the agent chain process a ticket?

Users can import tickets from Jira, Linear, or GitHub Issues, or write tasks directly on the board. An ordered agent chain (Architect, Coder, Reviewer, Tester, Acceptance) plans the work, writes code against a real repository checkout, reviews and tests it, opens the pull request, and comments the outcome back on the ticket.

What are the licensing and pricing terms?

The platform is completely free and open source with an MIT license, featuring no paid tier and no open-core holdback. Everything ships in the version you self-host, ensuring your code, data, and model credentials never leave your own infrastructure.

How does the system handle risk and human approval?

Per-task risk policies allow users to set what merges without a human through score ceilings, rules per change class, narrowing by workspace role, retry budgets, and human decision gates that can be placed anywhere in the pipeline.

How does the platform scale for different organization sizes?

The tool scales from one engineer on a laptop to a whole organization running every team's agent flows on the same build. It supports directory SSO with group gating, an org-wide monthly LLM cap that meters spend, and an audit log kept to outlive every other record.

What capabilities does the public API provide?

A key-authenticated HTTP surface under /api/v1 covers the whole board, allowing deployments to provision and run themselves with nobody opening the app. This includes linking or bootstrapping repositories, connecting a cluster, and filing tasks from Jira, Linear, or GitHub tickets.

How are headless runs tracked and monitored?

Runs started over the API execute exactly as board runs and show up on the board with keyset-paged lists. The system also provides traces exportable to OpenTelemetry or Langfuse for detailed monitoring.

What development tools and clients are available?

The OpenAPI 3.1 document ships with official TypeScript, Python, Go, and Java/Kotlin clients generated from it. Additionally, an MCP server operates over the same operations to support various development environments.

How does the platform manage API versioning and stability?

The stability contract ensures that changes are additive, and a breaking change is served beside the old shape instead of replacing it. This approach maintains backward compatibility for existing integrations.

What features do users particularly appreciate?

Some reviewers praise the ability to zoom out for the whole system and zoom into a running task where its card opens into the live pipeline, steps, and spending details. Users appreciate seeing runs in flight, failures, and decisions waiting on them appear where the work sits.

How does the interface facilitate real-time decision making?

The platform allows users to answer a gate or retry a step right there instead of tailing logs. This centralized interface helps streamline the workflow and decision-making process.

What programming languages are supported for integration?

The tool includes SDKs in four languages, enabling developers to integrate with their preferred programming environment. This flexibility supports diverse technical teams and project requirements.

Which ticketing systems can be integrated?

The system supports importing tickets from multiple sources including Jira, Linear, and GitHub Issues. This multi-source integration allows teams to consolidate their workflow management in one place.

How does the platform manage LLM usage and costs?

The platform provides an org-wide monthly LLM cap that meters spend and auto-pauses at the ceiling. This feature helps organizations control costs and manage resource allocation effectively.

What is the official website of Cat Factory?

The official website of Cat Factory is catfactory.ai.

AI-generated from reviews on this page. Verify anything important with the company directly.

James Wilson
Reviews 1

Cat Factory presents a strong value proposition with its MIT license and self-hosted architecture. The visual board effectively tracks the state of every pipeline and highlights decisions waiting on personnel. The agent chain structure is logical, and the ability to set per-task risk policies adds necessary control. However, running the platform entirely on your own infrastructure means you retain the operational responsibility. While this keeps data secure, it requires dedicated resources to manage the deployment. It is a solid choice for teams wanting full control, provided they are prepared to handle the infrastructure management themselves rather than relying on a managed service.

Sarah Jenkins
Reviews 1

I have been exploring Cat Factory for our engineering team, and the core concepts are impressive. The agent chain from Architect to Acceptance feels well structured, and the visual board gives us a clear view of what is happening. Being MIT licensed and self-hosted is a huge plus for our security requirements, as our code and model credentials never leave our infrastructure. The public API with SDKs for TypeScript, Python, Go, and Java allows for good automation. My reservation is that since we need to run it on our own infrastructure, whether on Cloudflare Workers or Node.js, we must handle the deployment and scaling ourselves. For a small team, this might require more initial effort than a purely SaaS product, but the control and cost efficiency are undeniable.

David Chen
Reviews 1

Cat Factory offers a compelling MIT-licensed platform for managing LLM coding agents. The visual board provides excellent visibility into agent chains and decision gates. However, since it runs entirely on your own infrastructure, you must manage the deployment yourself. While this ensures data stays within your environment, it shifts the operational burden from the vendor to your team, which may not suit organizations seeking a fully managed solution.

Marcus Thorne
Reviews 1

Cat Factory provides a robust visual board that effectively manages LLM coding agents with remarkable clarity. The ability to import issues from Jira, Linear, or GitHub Issues while keeping them as the source of truth is a significant advantage. Watching runs unfold in real time on a single canvas allows for immediate intervention when agents require human decisions. The MIT license ensures complete transparency, and running the platform on existing infrastructure eliminates additional subscription costs. This solution scales seamlessly from individual developers to large engineering organizations, offering a comprehensive view of every decision and expenditure without vendor lock-in.

News

No news yet. This is where the business shares its own updates and press.

Photos

No photos yet. Photos shared by reviewers will show up here.

Widgets

Boost your business trust by displaying a widget on your website — a simple way to convert your visitors into customers.

Get the widget for your site →

About Cat Factory

Cat Factory is a self-hosted, open-source software tool that coordinates LLM-based coding agents to turn tickets and plans into reviewed pull requests. It combines a visual planning board with a live pipeline view, showing in-flight runs, failures, and items awaiting human approval. Tickets can be imported from Jira, Linear, or GitHub Issues, and an ordered chain of agents (Architect, Coder, Reviewer, Tester, Acceptance) plans and implements the work against a real repository. The tool is driven through the board, headlessly over a versioned public API with SDKs in four languages, or self-hosted under an MIT license. It includes per-task risk policies, human decision gates, and usage controls such as monthly LLM spend caps and group-based SSO gating for larger organizations.