AI instruction grading instrument

Grade, rewrite, and test your AI instructions

Grade prompts, agents, skills, commands, and workflows against a clear 10-dimension rubric — then turn the weak ones into testable, version-tracked instructions.

Run an audit See examples

specimen / RBR-082

graded · 0.4s

› You are a helpful assistant. Answer questions well.

/100

Editor mark

Undefined success criteria

No reliable way to know what “good” means.

10-dimension rubric

Every audit scores objective, context, constraints, output, and verification.

Web, CLI & MCP

Run the same audits from the app, the rubrkit npm CLI, or any MCP client.

Compaction

Deterministic, free shortening that keeps every instruction an agent needs.

Free to try

Credit-metered so every run shows why it costs what it does.

How it works

An editorial loop for instructions that need to hold.

Paste an instruction

Add a prompt, command, skill, agent spec, or workflow.

Get a scored critique

Rubrkit grades clarity, context, constraints, output shape, and evaluation criteria.

Rewrite with checks

Get a stronger version plus simple evals that prove whether it works.

What Rubrkit reviews

One grading system for the instructions AI teams actually reuse.

Rubrkit detects the artifact type and applies the 10-dimension rubric that matches how it should behave.

Prompt

Agent

Command

Skill

Workflow

Generating more text is not the point. Rubrkit shows you what is weak, why it matters, how to fix it, and whether the fix can survive a real eval.

Example before / after

From vague request to testable instruction.

Before · 46/100

Write a professional launch email for my new course and make it engaging.

After · 84/100

Write a launch email for [TARGET AUDIENCE] that drives [PRIMARY GOAL]. Use a clear subject line, three short sections, one CTA, and avoid claims that are not supported by [CONTEXT].

Developers

Bring Rubrkit into your toolchain.

Sync graded artifacts into the repos where your agents run, or connect any MCP client to the same audits, bundles, and compaction tools the web app uses.

CLI

rubrkit on npm

Pull approved artifact bundles into local projects and place them where Codex, Claude, or generic agents expect their instructions.

# Pull approved artifacts into your project
npx rubrkit pull

# Place them where your agent expects them
npx rubrkit pull all --agent claude --yes

View on npm API docs

MCP server

Public

Point any MCP client at Rubrkit and call the same artifact bundle, audit, and compaction tools, authenticated with your Rubrkit API key.

{
  "mcpServers": {
    "rubrkit": {
      "url": "https://www.rubrkit.com/api/v1/mcp",
      "headers": {
        "Authorization": "Bearer <your-rubrkit-api-key>"
      }
    }
  }
}

Create an API key API docs

Pricing

Free to try, built for serious review loops.

Pro and Team workflows are in preview while the Rubrkit review loop is being tuned.

Free

0 tokens to start

For quick checks and first rewrites.

0 tokens to start

Grade & improve

Top issues + one rewrite

No card required

Pro

Questions before you grade your first instruction.

Know which instructions are ready to run.

Grade a specimen, read the marks, ship the rewrite that survives an eval.

Run an audit

Newsletter

Follow the review loop as it ships.

Notes on AI artifact testing, prompt compaction, evals, and proof reports.