Skip to content

Skill1 step

Prompt Polish

Rewrite a rough idea into the prompt each model responds to best.

How it runs

One call, every step handled.

Models read prompts differently. Write your idea once and the skill rewrites it for each model you plan to use, with the detail, structure and vocabulary that model rewards.

Every intermediate result is kept with the run, so you can inspect a step, reuse its output or rerun it with different settings.

  1. Input
    • Idea
    • Target models
  2. 01
    Rewrite for each modelPrompt writer
  3. OutputA tuned prompt for each model

Run it

Call the skill by name.

Send the inputs, get back a run you can wait on or follow with a webhook. The same skill is available in the studio and to AI agents through the MCP server.

You get back

A tuned prompt for each model

curl https://api.monogram.dev/v1/skills/prompt-polish/runs \
  -H "Authorization: Bearer $MONOGRAM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "input": {
      "idea": "cozy bookshop on a rainy night",
      "models": [
        "bfl/flux-2-pro",
        "google/veo-3.1"
      ]
    }
  }'