- Open Source · Cross Platform · Any Tech Stack -
-
- Test beyond
- the happy path.
+
+
+ Your app works fine.
+ Until the API doesn't.
-
- Dev Proxy is an API simulator that helps you test how your app handles errors, throttling, and slow responses — without changing a single line of code.
+
+ Simulate API failures, throttling, and slow responses against your running app. No code changes required.
-
+
-
-
+
+
@@ -49,77 +46,88 @@ import Layout from '../layouts/Layout.astro';
terminal
-
+
❯ devproxy
info Dev Proxy listening on 127.0.0.1:8000...
-
+
req GET https://api.contoso.com/posts
skip GenericRandomErrorPlugin: Pass through
- pass Passed through
-
+ 200 Passed through
+
req GET https://api.contoso.com/posts
skip RetryAfterPlugin: Request not throttled
- oops GenericRandomErrorPlugin: 429 429
+ 429 GenericRandomErrorPlugin: Too Many Requests
+
+ req POST https://api.contoso.com/orders
+ skip LatencyPlugin: Adding 3200ms delay
+ 504 GenericRandomErrorPlugin: Gateway Timeout
-
+
-
+
You test the happy path.
- Your users get the rest.
+ Dev Proxy tests everything else.
- APIs fail. They throttle. They respond slowly. They return unexpected data. And when they do, your app needs to handle it gracefully. But simulating these failures is hard — until now.
+ Errors, throttling, timeouts, unexpected payloads. All the things that break apps in production but are hard to reproduce in development. Dev Proxy simulates them against your running app, locally, without changing your code.
-
-
-
-
-
-
-
- API Mocking
- Mock APIs that don't exist yet.
-
- Stand up a fully-functional CRUD API without writing a single line of server code. Design, prototype, and build your frontend in parallel with your backend.
-
-
- Learn more →
+
+
+
+
+ What do you want to test?
+ Pick a scenario. You'll be running in minutes.
+
+
+ {[
+ { title: 'Test how my app handles API errors', time: '5 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/test-my-app-with-random-errors' },
+ { title: 'Mock an API that doesn\'t exist yet', time: '10 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/simulate-crud-api' },
+ { title: 'Use a local LLM instead of OpenAI', time: '15 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/simulate-openai' },
+ { title: 'Simulate rate limiting locally', time: '10 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/simulate-rate-limit-api-responses' },
+ { title: 'Run API failure tests in CI/CD', time: '15 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/use-dev-proxy-with-github-actions' },
+ { title: 'Find out which APIs my app calls', time: '5 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/discover-urls-to-watch' },
+ ].map((scenario) => (
+
+ {scenario.title}
+
+ {scenario.time}
+ →
+
-
-
- // customers-api.json — CrudApiPlugin
- "baseUrl": "https://api.contoso.com/customers"
- getAll GET /customers
- getOne GET /customers/{id}
- create POST /customers
- merge PATCH /customers/{id}
- delete DELETE /customers/{id}
- // Defined in JSON. Zero server code.
-
+ ))}
+
+
-
+
+
+
-
- Resilience Testing
- See how your app handles API errors.
+
+ Verify your retries, fallbacks, and error states actually work.
- Simulate random errors, timeouts, and failures for any API — without touching your code. Build more robust apps and never lose your customers' data.
+ Inject random errors, timeouts, and 5xx responses into any API your app calls. Test your loading states, retry logic, and error messages before real users depend on them. Zero code changes.
Learn more →
-
+
// errors-contoso-api.json
"url": "https://api.contoso.com/*"
"responses":
@@ -129,130 +137,73 @@ import Layout from '../layouts/Layout.astro';
// Random errors at configured failure rate.
+
+
-
-
-
- AI / LLM Testing
- Build & test AI apps without burning tokens.
-
- Route OpenAI and Azure OpenAI API calls to a local language model during development. Get real responses while building your AI-powered apps — without spending tokens.
-
-
- Learn more →
-
-
-
- // Your app calls OpenAI
- POST /v1/chat/completions
- // Dev Proxy routes to local LLM
- → Ollama (llama3.2)
- 200 { "choices": [{ "message": ... }] }
- // Real responses. Zero API costs.
-
-
+
+
+
-
+
- Rate Limit Simulation
- Verify your app respects API limits.
-
- Simulate rate limiting and throttling responses before they happen in production. Avoid getting blocked and deliver a smoother experience for your users.
+
Build frontend features before the backend is ready.
+
+ Mock realistic CRUD APIs locally so your team can prototype, integrate, and ship frontend work without waiting for backend endpoints. Define it in JSON, get a working API in seconds.
-
+
Learn more →
-
- RateLimit-Limit: 3
-
-
- GET /api/orders
- Remaining: 2
-
-
- GET /api/orders
- Remaining: 1
-
-
- GET /api/orders
- Remaining: 0
-
-
-
- GET /api/orders
- 429 Too Many Requests
-
- Retry-After: 30
-
+
+ // customers-api.json / CrudApiPlugin
+ "baseUrl": "https://api.contoso.com/customers"
+ getAll GET /customers
+ getOne GET /customers/{id}
+ create POST /customers
+ merge PATCH /customers/{id}
+ delete DELETE /customers/{id}
+ // Defined in JSON. Zero server code.
-
+
- CI/CD Integration
- Automate chaos testing in your pipeline.
-
- Run Dev Proxy as part of your CI/CD pipeline to catch API integration issues before they reach production. Works with GitHub Actions, Azure DevOps, and more.
+
Iterate on AI features without burning tokens.
+
+ Route OpenAI and Azure OpenAI calls to a local model while you build. Iterate faster, test AI workflows end-to-end, and keep your API bill at zero during development.
-
+
Learn more →
-
- # .github/workflows/test.yml
- steps:
- - name: Setup Dev Proxy
- uses: dev-proxy-tools/actions/setup@v1
- - name: Run tests
- run: npm test
- - name: Stop Dev Proxy
- uses: dev-proxy-tools/actions/stop@v1
- ✓ 42 passed, 0 failed
-
-
-
-
-
-
- MCP Server
- Configure Dev Proxy with AI assistants.
-
- Use the Dev Proxy MCP server to create configurations using natural language. Get contextual help and discover features — right from your AI assistant.
-
-
- Learn more →
-
-
-
- // .vscode/mcp.json
- {
- "servers": {
- "devproxy": {
- "type": "stdio",
- "command": "npx",
- "args": ["-y", "@devproxy/mcp"]
- }
- }
- }
- // AI-powered configuration. Natural language.
+
+ // Your app calls OpenAI
+ POST /v1/chat/completions
+ // Dev Proxy routes to local LLM
+ → Ollama (llama3.2)
+ 200 { "choices": [{ "message": ... }] }
+ // Real responses. Zero API costs.
-
-
-
-
- Platform Agnostic
- Works with everything you use.
-
- Dev Proxy intercepts network traffic at the system level. It works with any language, framework, and API — no SDKs, no adapters, no code changes.
+
+
+
+
+ Test the app you actually ship.
+
+ Dev Proxy intercepts HTTP traffic at the system level. No SDKs. No adapters. No test-only code paths. It works against the app you already have. The same binary your users run.
-
+
+ Not mocks. Not stubs. Not a test harness.
Real API behavior, simulated locally against your real app.
+
+
{[
{ name: '.NET', icon: `${import.meta.env.BASE_URL}logos/dotnet.svg` },
{ name: 'Node.js', icon: `${import.meta.env.BASE_URL}logos/nodejs.svg` },
@@ -263,22 +214,59 @@ import Layout from '../layouts/Layout.astro';
{ name: 'Browser', icon: `${import.meta.env.BASE_URL}logos/browser.svg` },
{ name: 'Any REST API', icon: `${import.meta.env.BASE_URL}logos/rest-api.svg` },
].map((tech) => (
-
-
- {tech.name}
+
+
+ {tech.name}
))}
-
+
+
+
+
+
+
+ Prevent rate-limit surprises.
+
+ Simulate 429s and throttling locally. Validate your retry logic and backoff strategy before real APIs block your app.
+
+
+ Learn more →
+
+
+
+
+ Catch issues before deploy.
+
+ Run Dev Proxy in CI/CD to test against errors and throttling on every push. Issues surface in the PR, not in production.
+
+
+ Learn more →
+
+
+
+
+ Configure with AI assistants.
+
+ Use the Dev Proxy MCP server to generate configurations in natural language. Describe what you want to simulate and your AI assistant writes the config.
+
+
+ Learn more →
+
+
+
+
+
+
+
- See It In Action
- Get started with Dev Proxy.
-
- Watch how to set up Dev Proxy and start testing your app in minutes.
+
Up and running in 5 minutes.
+
+ Install, point at any URL, and start simulating failures immediately.
@@ -294,66 +282,31 @@ import Layout from '../layouts/Layout.astro';
-
-
+
+
-
- Get Started In Minutes
- What do you want to do?
-
-
- {[
- { title: 'Mock an API that doesn\'t exist yet', time: '10 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/simulate-crud-api' },
- { title: 'Test my app handles API errors', time: '5 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/test-my-app-with-random-errors' },
- { title: 'Simulate OpenAI with a local LLM', time: '15 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/simulate-openai' },
- { title: 'Simulate rate limiting responses', time: '10 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/simulate-rate-limit-api-responses' },
- { title: 'Automate API testing in CI/CD', time: '15 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/use-dev-proxy-with-github-actions' },
- { title: 'Discover what APIs my app calls', time: '5 min', href: 'https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/discover-urls-to-watch' },
- ].map((scenario) => (
-
- {scenario.title}
-
- {scenario.time}
- →
-
-
- ))}
-
-
-
-
-
-
-
-
+
- VS Code Extension
- Dev Proxy Toolkit for Visual Studio Code.
-
- Control Dev Proxy directly from your editor. Manage configurations with IntelliSense, use 100+ code snippets, and leverage the built-in MCP server for AI-assisted development — all without leaving VS Code.
+
Stay in your editor.
+
+ The Dev Proxy Toolkit for VS Code gives you IntelliSense for config files, 100+ code snippets, debugger integration, and a built-in MCP server. All without leaving your editor.
-
+
Install from Marketplace →
-
-
- 
+
+
+
- Dev Proxy Toolkit
+ Dev Proxy Toolkit
garrytrinder
-
+
✓
- Control Dev Proxy from VS Code
+ Start/stop from the command palette
✓
@@ -369,7 +322,7 @@ import Layout from '../layouts/Layout.astro';
✓
- MCP server for AI-assisted development
+ MCP server for AI-assisted config
@@ -377,7 +330,7 @@ import Layout from '../layouts/Layout.astro';
-
+
-
+
-
+
How does your app handle
- API errors?
+ a flaky API?
- Find out in minutes. Dev Proxy is open source and free.
+ Find out in 5 minutes. Free, open source, runs on your machine.
- Get Started →
+ Simulate your first API error →
diff --git a/src/styles/global.css b/src/styles/global.css
index fceecbb..54adc8e 100644
--- a/src/styles/global.css
+++ b/src/styles/global.css
@@ -4,6 +4,7 @@
--color-purple-400: #c084fc;
--color-purple-500: #a855f7;
--color-purple-600: #9333ea;
+ --font-heading: 'Sora', ui-sans-serif, system-ui, sans-serif;
}
:root {
+ Your app works fine.
+ Until the API doesn't.
- - Dev Proxy is an API simulator that helps you test how your app handles errors, throttling, and slow responses — without changing a single line of code. +
+ Simulate API failures, throttling, and slow responses against your running app. No code changes required.
-❯ devproxy
info Dev Proxy listening on 127.0.0.1:8000...
- +req GET https://api.contoso.com/posts
skip GenericRandomErrorPlugin: Pass through
-pass Passed through
- +200 Passed through
+req GET https://api.contoso.com/posts
skip RetryAfterPlugin: Request not throttled
-oops GenericRandomErrorPlugin: 429 429
+429 GenericRandomErrorPlugin: Too Many Requests
+ +req POST https://api.contoso.com/orders
+skip LatencyPlugin: Adding 3200ms delay
+504 GenericRandomErrorPlugin: Gateway Timeout
+
You test the happy path.
- Your users get the rest.
+ Dev Proxy tests everything else.
- Your users get the rest. + Dev Proxy tests everything else.
- APIs fail. They throttle. They respond slowly. They return unexpected data. And when they do, your app needs to handle it gracefully. But simulating these failures is hard — until now. + Errors, throttling, timeouts, unexpected payloads. All the things that break apps in production but are hard to reproduce in development. Dev Proxy simulates them against your running app, locally, without changing your code.
API Mocking
-Mock APIs that don't exist yet.
-- Stand up a fully-functional CRUD API without writing a single line of server code. Design, prototype, and build your frontend in parallel with your backend. -
- - Learn more → + +What do you want to test?
+Pick a scenario. You'll be running in minutes.
+{scenario.title}
+// customers-api.json — CrudApiPlugin
-"baseUrl": "https://api.contoso.com/customers"
-getAll GET /customers
-getOne GET /customers/{id}
-create POST /customers
-merge PATCH /customers/{id}
-delete DELETE /customers/{id}
-// Defined in JSON. Zero server code.
-Resilience Testing
-See how your app handles API errors.
+Verify your retries, fallbacks, and error states actually work.
- Simulate random errors, timeouts, and failures for any API — without touching your code. Build more robust apps and never lose your customers' data. + Inject random errors, timeouts, and 5xx responses into any API your app calls. Test your loading states, retry logic, and error messages before real users depend on them. Zero code changes.
Learn more →// errors-contoso-api.json
"url": "https://api.contoso.com/*"
"responses":
@@ -129,130 +137,73 @@ import Layout from '../layouts/Layout.astro';// Random errors at configured failure rate.
AI / LLM Testing
-Build & test AI apps without burning tokens.
-- Route OpenAI and Azure OpenAI API calls to a local language model during development. Get real responses while building your AI-powered apps — without spending tokens. -
- - Learn more → - -// Your app calls OpenAI
-POST /v1/chat/completions
-// Dev Proxy routes to local LLM
-→ Ollama (llama3.2)
-200 { "choices": [{ "message": ... }] }
-// Real responses. Zero API costs.
-Rate Limit Simulation
-Verify your app respects API limits.
-- Simulate rate limiting and throttling responses before they happen in production. Avoid getting blocked and deliver a smoother experience for your users. +
Build frontend features before the backend is ready.
++ Mock realistic CRUD APIs locally so your team can prototype, integrate, and ship frontend work without waiting for backend endpoints. Define it in JSON, get a working API in seconds.
- + Learn more →RateLimit-Limit: 3
-Retry-After: 30
-// customers-api.json / CrudApiPlugin
+"baseUrl": "https://api.contoso.com/customers"
+getAll GET /customers
+getOne GET /customers/{id}
+create POST /customers
+merge PATCH /customers/{id}
+delete DELETE /customers/{id}
+// Defined in JSON. Zero server code.
CI/CD Integration
-Automate chaos testing in your pipeline.
-- Run Dev Proxy as part of your CI/CD pipeline to catch API integration issues before they reach production. Works with GitHub Actions, Azure DevOps, and more. +
Iterate on AI features without burning tokens.
++ Route OpenAI and Azure OpenAI calls to a local model while you build. Iterate faster, test AI workflows end-to-end, and keep your API bill at zero during development.
- + Learn more →# .github/workflows/test.yml
-steps:
-- name: Setup Dev Proxy
-uses: dev-proxy-tools/actions/setup@v1
-- name: Run tests
-run: npm test
-- name: Stop Dev Proxy
-uses: dev-proxy-tools/actions/stop@v1
-✓ 42 passed, 0 failed
-MCP Server
-Configure Dev Proxy with AI assistants.
-- Use the Dev Proxy MCP server to create configurations using natural language. Get contextual help and discover features — right from your AI assistant. -
- - Learn more → - -// .vscode/mcp.json
-{
-"servers": {
-"devproxy": {
-"type": "stdio",
-"command": "npx",
-"args": ["-y", "@devproxy/mcp"]
-}
-}
-}
-// AI-powered configuration. Natural language.
+// Your app calls OpenAI
+POST /v1/chat/completions
+// Dev Proxy routes to local LLM
+→ Ollama (llama3.2)
+200 { "choices": [{ "message": ... }] }
+// Real responses. Zero API costs.
Platform Agnostic
-Works with everything you use.
-
- Dev Proxy intercepts network traffic at the system level. It works with any language, framework, and API — no SDKs, no adapters, no code changes.
+
+
+ Dev Proxy intercepts HTTP traffic at the system level. No SDKs. No adapters. No test-only code paths. It works against the app you already have. The same binary your users run.
+ Not mocks. Not stubs. Not a test harness. {tech.name} {tech.name}
+ Simulate 429s and throttling locally. Validate your retry logic and backoff strategy before real APIs block your app.
+
+ Run Dev Proxy in CI/CD to test against errors and throttling on every push. Issues surface in the PR, not in production.
+
+ Use the Dev Proxy MCP server to generate configurations in natural language. Describe what you want to simulate and your AI assistant writes the config.
+ See It In Action
- Watch how to set up Dev Proxy and start testing your app in minutes.
+
+ Install, point at any URL, and start simulating failures immediately.
Get Started In Minutes VS Code Extension
- Control Dev Proxy directly from your editor. Manage configurations with IntelliSense, use 100+ code snippets, and leverage the built-in MCP server for AI-assisted development — all without leaving VS Code.
+
+ The Dev Proxy Toolkit for VS Code gives you IntelliSense for config files, 100+ code snippets, debugger integration, and a built-in MCP server. All without leaving your editor.
Dev Proxy Toolkit Dev Proxy Toolkit garrytrinder
- Find out in minutes. Dev Proxy is open source and free.
+ Find out in 5 minutes. Free, open source, runs on your machine.
Test the app you actually ship.
+
Real API behavior, simulated locally against your real app.
+ Prevent rate-limit surprises.
+ Catch issues before deploy.
+ Configure with AI assistants.
+ Get started with Dev Proxy.
- Up and running in 5 minutes.
+ What do you want to do?
- {scenario.title}
- Dev Proxy Toolkit for Visual Studio Code.
- Stay in your editor.
+
+
+
How does your app handle
- API errors?
+ a flaky API?