Agent-native E2E runtime with verifiable safety. 13 MCP tools including alethia_propose_tests (agent generates tests from a URL), alethia_assert_safety (proves destructive actions are blocked), and the expect block: NLP primitive unique to Alethia. Zero-IPC, ~45x faster than Playwright, signed evidence packs. Works with Claude Code, Cursor, Cline.
95
94%
Does it follow best practices?
Impact
97%
2.77xAverage score across 5 eval scenarios
Advisory
Suggest reviewing before use
The Orion team has just started using Alethia for browser-based E2E testing. A new developer needs to configure the environment from scratch and get a sample test running against a local HTML fixture. The developer is using Claude Code as their AI agent and needs to configure the MCP integration correctly so the agent can call Alethia tools.
The developer has no existing Alethia setup. Produce a complete setup guide and the required configuration files so the developer can follow the steps and have a working Alethia test environment.
Produce the following files:
setup_guide.md — Step-by-step instructions for installing Alethia and configuring the MCP bridge, including the exact install command and what to add to the agent's MCP configuration.mcp_config.json — The complete MCP configuration snippet to add to the agent config, ready to copy and paste.sample_test.txt — A sample Alethia test instruction set (one step per line) that:
file:///workspace/orion/index.html