drafthouse
Provides a verification and correction harness for Hermes Agent, adding reference-based generation, P0 checklist, five-dimension evaluation, linting, vision, and human review gates to agent outputs.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@drafthouselint my landing page against the default design system"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Drafthouse
v1.0.0 — Open-source Claude Design–class verify harness for Hermes Agent.
Hermes already has the agent core. Drafthouse ships the hidden check-and-correct loop:
refs → bind tokens → generate → P0 checklist → 5-dim → lint → vision → humanQuick start (safe)
export PYTHONPATH="$PWD/src" DRAFTHOUSE_ROOT="$PWD"
python3 -m drafthouse.doctor
bash bin/drafthouse lint tests/fixtures/clean.html --system design-systems/default
python3 scripts/eval_plate.pyDocker lab (does not touch live ~/.hermes):
sudo docker compose -f docker/docker-compose.yml run --rm drafthouse-labOptional live install: python3 -m drafthouse.install_hermes --dry-run first.
Related MCP server: Design Constraint Validator MCP Server
What’s in the box
Area | Contents |
Skills | verify · systems · references · vision |
Gates | lint P0 · tokens · 5-dim · vision (max 3 rounds) |
MCP | 10 stdio tools (lint/bind/refs/vision/…) |
Systems | default · editorial-field · saas-minimal · developer-docs · dark-product |
Refs | 57+ galleries + 10 playbooks |
Evals | golden plates + honesty stub |
Ops | doctor · sandbox installer · Docker CI |
Docs
Monorepo finish plan:
../docs/DRAFTHOUSE_FINISH_PLAN.mdHandbook: https://k1ng0mar.github.io/drafthouse-hermes-handbook/
License
MIT
This server cannot be deployed
Maintenance
Related MCP Connectors
Statically audits MCP tool surfaces for token cost, schema quality, and design issues.
Hosted MCP for creating, checking, deploying, and hosting static sites for AI agents.
MCP server for secureFlows: token-free URL builders and integration-linting tools for AI agents.
OCR, transcription, file extraction, and image generation for AI agents via MCP.
Related MCP Servers
- FlicenseNot gradedqualityBmaintenancePolicy-as-code gate for AI-SDLC, providing MCP tools to review prompts, diff tool manifests, vet MCP servers, and run evaluation suites for LLM agent repos.1-
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to validate design system constraints, including WCAG accessibility, token relationships, and consistency, through MCP tools like validate, explain, and suggest-fix.53 npmMIT
- AlicenseNot gradedqualityAmaintenanceEnables AI agents to score live URLs against a 40-check design contract, validate DTCG tokens and Lottie animations, audit accessibility, and retrieve design-system contracts, catalogs, and review rubrics.30 npmMIT
- FlicenseNot gradedqualityCmaintenanceMCP server that exposes a design system and UX rulebook as tools, enabling AI to generate UI consistent with your design tokens, components, layout rules, accessibility guidelines, and templates. It also provides a review tool to score UI proposals against the design system.-