Ironwood AI reviews code using AI providers selected by the team and engineering context such as coding standards, architecture documents, requirements and technical guidance. Engineers can inspect review findings and their evidence. Teams can process several repositories in one batch, including cross-repository analysis when service boundaries matter. Listed source-control integrations are GitHub, Azure DevOps and Bitbucket; delivery integrations include Jira, Azure Boards and Confluence. Ironwood runs on infrastructure controlled by the customer, with local Admin, Reviewer and Viewer roles. The customer manages source-control and AI-provider credentials. With a hosted AI provider, prompts and selected code or context are sent to that provider; Ollama can keep the AI request within the customer's infrastructure. Ironwood says it does not need to receive or store code, prompts or findings to run a review. Professional includes PDF reports, role-based access control, platform health and diagnostics. The Evaluation plan is free for 30 days for two developers and one active installation. Professional costs $15.00 USD per month, and Team costs $200.00 USD per month for up to 20 developers. AI-provider usage is billed separately.
Who it is for
It suits engineering teams that want AI-assisted code reviews while keeping control of infrastructure, providers and review decisions. Teams using multiple repositories or documented engineering standards may find its review context and batch analysis useful.
What is good
- Supports batch reviews across multiple repositories
- Accepts coding standards and architecture documents as context
- Runs on customer-controlled infrastructure
- Professional includes PDF reports and diagnostics
What to know first
- Hosted providers receive prompts and selected code or context
- AI-provider usage is billed separately
- Evaluation is limited to two developers and one installation
Verdict
Ironwood AI pairs configurable review context with customer-controlled deployment and several provider choices. Teams should account for the data sent to hosted AI providers and their separate usage charges.
Get started with Ironwood AI
- Open the Ironwood website.
- Choose the 30-day Evaluation or a paid plan.
- Run Ironwood on infrastructure you control.
- Connect a supported source-control service and manage its credentials.
- Select an AI provider and manage its credentials.
- Provide engineering guidance and start a manual review.
Limits to know first
The Evaluation plan lasts 30 days and covers two developers and one active installation. Professional is $15.00 USD per month; Team is $200.00 USD per month and supports up to 20 developers. AI-provider usage is billed separately.
Questions about Ironwood AI
Does Ironwood AI have a free plan or trial?
There is no ongoing free plan. Evaluation is free for 30 days, includes full Professional features, and requires no credit card.
How much do the plans cost?
Professional is $15.00 USD per month. Team is $200.00 USD per month. Enterprise is billed by agreement, with no price listed.
Which platforms and source-control services does it support?
The listed platforms are macOS, self-hosted, web, and Windows. Source-control integrations include GitHub, Azure DevOps, and Bitbucket.
Which AI providers and delivery integrations are listed?
AI providers are OpenAI, Anthropic Claude, Google Gemini, and Ollama. Delivery integrations include Jira, Azure Boards, and Confluence.
Where does Ironwood run, and what data goes to an AI provider?
Ironwood runs on customer-controlled infrastructure. With a hosted AI provider, review prompts and selected code or context are sent to that provider; Ollama can keep the AI request inside the customer's infrastructure.
Are there usage charges beyond the subscription?
AI-provider usage is billed separately by the providers. Professional and Team have unlimited repositories and reviews, no codebase-size limit, and no per-review charge.
Ironwood AI plans and pricing
All plansCompared on AI code review tools
- Free plan
- No
- Custom review rules
- Yes
- Deployment model
- self_hosted
- Review trigger
- manual
- Code hosts
- GitHub, Azure DevOps, Bitbucket
