Crucible — Project-Agnostic Test Harness & Visual Regression MCP
Crucible — Project-Agnostic Test Harness & Visual Regression MCP
📦 Crucible has been extracted into its own public repo: github.com/danhannah94/crucible. The canonical design lives there now at
docs/design.md. This Foundry doc remains for historical context — session decisions, refinement threads, and the original framing — but ongoing design updates happen in the repo.
One-line pitch: An agentic QA framework that gives AI agents eyes. Crucible provides visual verification, baseline management, and repeatable test environments — while agents use your app's own MCP tools (or generic browser tools as a fallback) for interaction. The agent is the test runner.
Status: Extracted to public repo on 2026-05-26 — v0.1 "Eyes Only" validated end-to-end.
Created: 2026-04-11
Updated: 2026-04-13 — major reframe from "Playwright wrapper + spec runner" to "visual verification layer for agentic QA"
Third package in the @claymore-dev suite, alongside Foundry (doc review) and Anvil (semantic search).
Overview
System Architecture
Data Model
Agentic QA Pipeline
This is the end-to-end flow that Crucible enables. A separate QA agent verifies the implementation agent's work — independent verification, not self-review.
AI Interface Architecture
This section is the whole point of the project.