Code Rule and Code Query
Hundreds of default code rules to check against best practices. Support for Code Query over C# LINQ (CQLinq) to easily customize rules and query code.
NewNDepend.AI: Fix .NET Issues with Your Coding Agent
Our harness, your agent. NDepend.AI hands the issues found by NDepend to Claude Code, Codex, GitHub Copilot CLI or Junie, and verifies every fix with a build, the impacted tests and a new analysis. On eShopOnWeb: 496 of 511 issues fixed, 531 tests added, coverage from 45% to 98%.
Discover NDepend.AI See the 6 run reports
Easily Manage Large Projects
Most companies have large systems that work together, often in mysterious ways. NDepend's in-depth reporting and toolset can help your company enforce its own coding standards and maintain complex projects.
Powerful Dependency Graph and Matrix
Explore how the code is actually structured and shed light on architectural flaws thanks to the most powerful code dependency graph and dependency matrix available in the tool industry.
NDepend.MCP.Server
Your coding agent asks NDepend: the open-source MCP server gives it the issues, rules, quality gates, dependencies and metrics of your code, to fix NDepend rules, Roslyn Analyzers and R# Code Inspections issues while you steer.
Continuous .NET Code Quality and Security
A unique approach to continuously keep the technical debt under control and reimburse it with time.
Interactive Web Report
Integrate NDepend analysis into your CI / CD pipeline and obtain interactive and detailed web reports to explore progression and prevent code quality degradation.
Quality Gates
Fail the build pipeline upon customizable PASS / WARN / FAIL code quality and security criteria.
Import Roslyn and ReSharper Issues
Gain a 360 view of your .NET code quality by importing your preferred Roslyn Analyzers' issues and ReSharper Code Inspections' issues into the NDepend web report and UI.
Monitor the Health of Your Application
NDepend allows your team to accurately gauge the overall health of your application, no matter how complex it is.
Code Quality and Code Metrics
Lines of Code, Cyclomatic Complexity, Coupling, Nesting Depth, Rank, NDepend supports them all (and much more).
Trend Monitoring
Get Trend Charts about pretty much any code metrics to master the evolution of your application.
Harness Test Coverage Data
Import test coverage data, and make the most of it in the NDepend feature-rich context.
Reduce Friction Between Your Developer Teams
NDepend's reporting and analysis helps architects and developers to communicate and reduce confusion.
Complexity and Diagrams
Spot too complex code at a glance thanks to unique diagramming capabilities.
Code Diff since Baseline
Compare two versions of a code base and its issues set, and browse diff and changes in any way you can think of.
Warnings on Build Process Health
Be warned of potential build process problems before they end up causing friction and pain.
Working with Legacy Code is Now Easier
By mapping out the structure of the project, it is much easier to see how components and dependencies are connected.
Detect Dependency Cycles
Get rid of dependency cycles between components to achieve higher code maintainability.
Enforce Immutability and Purity
NDepend helps maintaining clean side-effects free code to write highly maintainable products.
Real-World Proof
Integrate with Visual Studio, Azure DevOps, GitHub and any other CI/CD pipeline, super fast and lightweight analysis, NDepend is conceived for real-world programmers.
NDepend.API and Power Tools
Write your own static analyzer based on NDepend.API, or tweak existing open-sources Power Tools.
In-Depth Issues Management
Explore the issues-set in a few clicks and focus your effort on what is urgent and important.
Smart Technical Debt Estimation
For each issue, the cost to fix and the severity are estimated through smart and customizable C# formulas. Thus estimations proposed are realistic.
NDepend.AI: Fix .NET Issues with Your Coding Agent New
A coding agent like Claude Code, OpenAI Codex, GitHub Copilot CLI or JetBrains Junie can fix real issues in your code. NDepend.AI steers your agent with NDepend: it hands the agent the issues found by NDepend rules (vulnerabilities, code smells, new issues since the baseline, or any rule or query you pick), and checks every fix with a build, the impacted tests and a new analysis. Hundreds of issues get fixed unattended, and you review one report at the end.
A Real Run on eShopOnWeb
- 468 / 493issues fixed
- 506tests added
- 45% → 97.5%code coverage
- 1h21 · $44duration and cost
Every number comes from the run report, an AI Assistant tab of the NDepend web report: see this report live, and the reports of the 5 other runs, including Jellyfin at large scale: 1,521 issues fixed in one run.
How It Works: Our Harness, Your Agent
Issue set selection
through a C# LINQ query:
where i.Severity >= Severity.High
NDepend.AI harness
- Analyze
- Targetthe next batch of issues
- Agent sessionsyour agent: Claude Code, Codex, Copilot or Junie, with NDepend.MCP.Server
- Build · Tests · Coveragetests red → repair
- Checkpoint
next round
AI review, then the full report
AI Assistant tab: before → after, change review, code diff, architecture
Why Not Just Ask Your Agent?
- The agent stays on track: the run goes in fix rounds. Each round targets a batch of issues, then the solution is built, the impacted tests run and the code is analyzed again.
- Fewer tokens: NDepend already knows your code. The agent gets each issue with its location and queries the code model through the integrated NDepend.MCP.Server, instead of exploring the code base.
- Every issue is accounted for: each issue targeted ends up fixed, suppressed or left, and the issues introduced by the fixes are caught. An issue counts as fixed only when the new analysis doesn't report it anymore.
- Tests where coverage is missing: the agent writes fast unit tests for the uncovered code, and reports a potential bug instead of writing a test that would lock in a wrong behavior.
- Predictable: every run follows the same steps. Two runs on the same code, with the same agent and settings, give results less than 1% apart.
Start a Run from VisualNDepend, Visual Studio, the Console or CI/CD
The AI Assistant menu offers one-click commands: Fix Vulnerabilities, Fix New Issues and Fix Code Smells, and an AI Fix Issues button shows up whenever a rule or a code query matches issues. The same run is available as a command line and as a settings file, ready for your CI/CD pipeline, on Windows, Linux and macOS.
NDepend.AI also writes, refines and fixes CQLinq code queries and rules from a request in natural language.
Commands, AI providers, arguments and CI/CD setup: NDepend.AI documentation.
Your agent can also query NDepend directly while you steer the session: NDepend.MCP.Server.
Your agent can also query NDepend directly while you steer the session: NDepend.MCP.Server.