Magenie33 Quality Dimension Generator
Turn vague task descriptions into explicit quality dimensions and scoring guides so you can judge agent output before release.
Overview
Quality Dimension Generator is a MCP server for the Ship phase that builds task-specific quality criteria and scoring guides from your descriptions.
What is this MCP server?
- Generates tailored quality criteria from natural-language task descriptions.
- Produces scoring guides you can reuse for human or LLM-as-judge checks.
- Supports refining objectives so rubrics match what ‘good’ means for each task.
- Hosted Smithery MCP remote—no local rubric DSL to learn first.
- Bearer auth required (smithery_api_key) on streamable-http transport.
- Server schema version 2025-09-29; package version 1.0.0
- 1 Smithery streamable-http remote
- Repository: github.com/magenie33/quality-dimension-generator
What problem does it solve?
Agent and LLM deliverables feel subjective until you have written rubrics, so shipping feels like guessing whether quality is acceptable.
Who is it for?
Builders running custom agent tasks who want lightweight eval design without building a full benchmark suite first.
Skip if: Teams that only need unit tests on deterministic code paths with no generative output to score.
What do I get? / Deliverables
After connecting the MCP remote, you get reusable dimensions and scoring guides that make pass/fail and iteration decisions explicit before launch.
- Task-specific quality dimension lists
- Scoring guides aligned to stated objectives
- Sharper acceptance language for Ship-phase reviews
Recommended MCP Servers
Journey fit
How it compares
MCP rubric and scoring-guide generator, not a test runner like pytest or a security audit skill.
Common Questions / FAQ
Who is Quality Dimension Generator for?
Solo builders and small teams shipping AI-assisted workflows who need clear quality bars for generative or multi-step agent tasks.
When should I use Quality Dimension Generator?
Use it in Ship when you are defining or tightening acceptance criteria before release, prompt changes, or structured review of agent output.
How do I add Quality Dimension Generator to my agent?
Add the Smithery remote https://server.smithery.ai/@magenie33/quality-dimension-generator/mcp with Authorization Bearer {smithery_api_key} in your MCP client.