Avalyz
Run acceptance tests on a web app from your assistant, with a GO or NO-GO verdict
- Pricing
- Free trial
- Platforms
- Web, API, CLI
- Runs
- Remotely
- Version
- 1.3.1
- Listing updated
- 1 day ago
About Avalyz
Avalyz is an MCP server for developer tools. It runs as a hosted remote server, so there is nothing to install locally.
Avalyz tests a web app from the outside and reports a verdict: GO, GO under conditions, NO-GO or inconclusive. Findings are graded S1 to S4 and linked to a report, so an assistant can fix the serious ones, redeploy and test again. Nothing is installed on the site under test.
By default it submits no form except signing in with your test account. In write mode, after you state that you own the app or are authorised to test it (recorded) or provide proof, it plays full journeys such as sign-up, forms, test-mode payments and admin areas. Sensitive domains, including government, banking and payments, health, energy and justice, and load tests require proof of ownership.
It connects through an online MCP server with OAuth or an API key, a downloadable local MCP script, a rules block for Claude Code or Cursor, a GitHub Actions workflow that fails on NO-GO, or a link and badge for Lovable, Bolt, v0 and Replit. A 14-day trial needs no card.
Install
Run in a terminal
claude mcp add --transport http avalyz https://avalyz.com/mcpAdd to Settings → Connectors → Add custom connector
https://avalyz.com/mcpAdd to ~/.cursor/mcp.json
{
"mcpServers": {
"avalyz": {
"url": "https://avalyz.com/mcp"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"avalyz": {
"type": "http",
"url": "https://avalyz.com/mcp"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"avalyz": {
"serverUrl": "https://avalyz.com/mcp"
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"avalyz": {
"type": "http",
"url": "https://avalyz.com/mcp"
}
}
}What it does
- Verdict reports
- Returns GO, GO under conditions, NO-GO or inconclusive with findings.
- Severity grading
- Grades findings from S1 to S4.
- Write mode
- Plays full journeys after an ownership statement or proof.
- CI integration
- A GitHub Actions workflow fails the build on NO-GO.
- One-link tests
- A URL or badge starts a test without a key.
- Safeguards
- Sensitive domains and load tests need proof of ownership.
Good for
- Test an AI-built app after every deployment
- Fail a pipeline when acceptance tests regress
- Check sign-up and payment flows in test mode
- Add a one-click test badge to a project README
Pricing
Free trial
14-day trial with no card; plan prices not listed
Prices change. Check avalyz.com before you buy.
Avalyz FAQ
Is there a free trial?
The site mentions a 14-day trial with no card for API keys, and a daily limit of free tests via the link.
Does it submit forms on my app?
Not by default. Only sign-in with a test account, unless you enable write mode with an ownership statement.
Which sites need proof of ownership?
Government, banking and payments, health, energy and justice domains, and any load tests.
How do I connect Lovable?
Add https://avalyz.com/mcp as an MCP server connector and authorise it.
Is Avalyz free?
Avalyz is a paid product with a free trial.
How do I add Avalyz to Claude Code?
Run this command in your terminal: claude mcp add --transport http avalyz https://avalyz.com/mcp
What are the best alternatives to Avalyz?
The closest alternatives to Avalyz are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.
Alternatives to Avalyz

REA
MCP serverCLI and MCP server that lets coding agents reverse engineer apps and binaries
Runs locally- Open source
- 464K GitHub stars
- Developer tools

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

GitHub
MCP serverOfficial MCP server connecting AI assistants to GitHub repos, issues and workflows
Not checked yet- Remote
- Open source
- 34.5K GitHub stars
- Developer tools

Serena MCP: the IDE for your agent
MCP serverMCP toolkit giving coding agents symbol-level code retrieval and editing
Runs locally- Open source
- 31.5K GitHub stars
- Developer tools

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Ouroboros
MCP serverSpec-first layer that pins requirements and verifies results of AI coding agent work
Runs locally- Open source
- 6.4K GitHub stars
- Developer tools