CodeScene
Behavioural code analysis with CodeHealth guardrails for AI code
Overview
CodeScene analyses how teams work with a codebase over time to surface technical-debt hotspots, scores code with CodeHealth in pull requests, the IDE and an MCP server for AI coding tools, and offers ACE, an add-on that refactors common code smells with validated LLM output. From €18 per active author a month.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 1 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- CodeHealth analysis and score
- Unlimited private repos
- Automated code review
- CodeHealth MCP Server and PR Refactoring Agent
- 30+ languages
- Everything in Standard
- Software portfolio overview
- Team & delivery insights
- Code coverage
- SSO and RBAC
- Dedicated Customer Success Manager
- Tailored onboarding & workshops
- Refactors 5 code smells
- JavaScript, TypeScript, React, Java, C#
What you can produce with CodeScene
- CodeHealth technical-debt and hotspot analysis
- Automated pull request review and IDE CodeHealth monitor
- CodeHealth MCP Server for AI coding assistants
- ACE AI refactoring for 5 code smell types (paid add-on)
- Knowledge distribution and team insights
- Code coverage (Pro tier)
- Free Community Edition for open-source projects
CodeScene's edge is behavioural: it ranks technical debt by where teams actually change code and who knows it, so you fix the hotspots that cost the most rather than every warning a linter raises. In 2026 that analysis became a guardrail for AI coding — a CodeHealth MCP server lets Copilot, Cursor or Claude Code check generated code against the same score — and ACE refactors five common code smells with validated LLM output, though validation is not a guarantee of identical behaviour and ACE is now a paid add-on. Pricing per active author, from €18 a month billed yearly, is fair for teams that will act on the insights. Reviews are thin and dated, and the interface takes learning. Strong for engineering leaders managing debt in large codebases and reviewing AI-written code; overkill for a small project.
What we found
CodeScene, founded in 2015 by Adam Tornhill in Malmö, Sweden, analyses how teams actually work with a codebase over time — which files change most, who knows which code, where complexity accumulates — to rank technical-debt hotspots by business impact rather than by raw code metrics. Its CodeHealth score rates the maintainability of each file, and the same analysis runs as automated pull-request review, as a monitor inside the IDE, and since 2026 as a CodeHealth MCP server that lets Copilot, Cursor or Claude Code check AI-written code against the score, plus a PR refactoring agent. CodeScene says the MCP approach uses up to 45% fewer tokens and cuts defect risk; those are its own research figures.
ACE, its AI refactoring feature, needs careful description. It rewrites five code smells — complex conditionals, bumpy roads, complex methods, deep nesting and large methods — in JavaScript, TypeScript, React, Java and C#, using models from OpenAI, Anthropic and Google, and validates the output before showing it. CodeScene is explicit that this is not a general proof of equivalence: it withholds suggestions that fail validation outright and shows partly validated ones for closer inspection. ACE was free during its beta; it is now a paid add-on sold through sales.
Pricing is per active author — anyone who committed in the past three months. Standard costs €18 a month billed yearly or €20 monthly ($19 or $21 in the US) and covers CodeHealth analysis, unlimited private repositories, automated code review, the MCP server, the refactoring agent and more than 30 languages. Pro, €27 or €30, adds portfolio, team and delivery insights and code coverage; Enterprise adds SSO, role-based access and a dedicated success manager. There is a free trial and a free Community Edition for open-source projects, but no free plan for private code. CodeScene is ISO 27001 certified and raised €7.5 million led by Neqst in 2023.
Capterra rates it 4.7 out of 5 from 11 reviews, all from 2019 to 2022 and none covering the AI features. Reviewers praise the insights and support; four of the 11 find the interface unintuitive.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
About this listing
This entry was compiled from publicly available data including CodeScene's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with CodeScene unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to CodeScene directly →
Spotted something out of date? Suggest an update →
Alternatives to CodeScene
More in Code & Dev Tools
