Squidler.io
@squidlerio
Model Context Protocol (MCP) server that connects AI assistants like Claude to Squidler's accessibility and quality checking platform
Overview
What is Squidler.io?
Squidler.io validates web apps as a human would, using natural language use cases without brittle DOM-dependent tests. It enables AI coding agents to analyze their own work from a user perspective and close the autonomous Build → Verify → Fix loop.
How to use Squidler.io?
Add the MCP server to your client configuration with transport http, URL https://mcp.squidler.io, and an Authorization: Bearer YOUR_SQUIDLER_API_TOKEN header. Get your API token from squidler.io/integrations/api-keys. The server includes built-in agent guidance that teaches a Test-After-Feature workflow, so the agent automatically knows how to use Squidler.
Key features of Squidler.io
- Natural language test case creation via guided conversation
- Site-bound and standalone (repo-stored) test case workflows
- Persona UX reviews with actionable findings
- Problem tracking and resolution for site issues
- Label-based test catalog organization and filtering
- Long-poll test execution status and event retrieval
Use cases of Squidler.io
- AI coding agent builds a feature, then describes the use case in natural language for Squidler to test it.
- Agent reviews test run outcomes and events to fix bugs or update obsolete tests.
- Build a growing catalog of tests organized by feature to prevent regressions.
- Run standalone test cases against a local dev server using a Chrome proxy.
- Manage test case lifecycle from creation to promotion to site-bound storage.
FAQ from Squidler.io
How do I get an API token for Squidler.io?
Generate your API token from squidler.io/integrations/api-keys.
What transport does the MCP server use?
The server uses HTTP transport with the URL https://mcp.squidler.io and requires an Authorization header with a Bearer token.
What is the built-in Test-After-Feature Workflow?
The server teaches agents to run existing tests, investigate failures, label the feature, cover the new feature with test cases, and confirm all tests pass before moving on.
What is the difference between site-bound and standalone test cases?
Site-bound test cases are stored on Squidler and created with site context. Standalone test cases are produced as markdown files for the agent to manage in its repo, useful for local dev iteration.
How do I run a test case against a local development server?
Use the test_case_run_standalone tool with a custom baseUrl (e.g., http://localhost:<port>) and pair it with the @squidlerio/squidler-mcp npx proxy to route runs through a local Chrome.
