Omniscio

Recipe Drafter (describe a workflow, get a validated recipe)

The Recipe Drafter turns a natural-language description of an automation into a draft recipe, returned with its reasoning, warnings and any remaining validation errors. The draft is not saved to disk, so the caller decides whether to save it, edit it, or retry. It reads a bundled pattern library, validates the candidate, and can refine its own draft for up to three turns.

On this page

What it is

Audience: AI agents (with or without repo access), and Omniscio users who want to know what the "Create from description" drafter does before wiring it up themselves.

Where to find it

This is the drafter behind the "Create from description" recipe flow, and it has no screen of its own — and no in-app caller. An agent, or a script using the CLI, sends a description (POST /recipe/draft, mirroring the RECIPE_CREATE_FROM_NL IPC channel) and gets a draft recipe back, along with its reasoning, warnings and any remaining validation errors.

The Recipes UI does not call it: the New-recipe button opens the static template gallery, and there is no "Create from description" control anywhere in the app. Nothing in src/renderer, src/preload or src/plugins invokes RECIPE_CREATE_FROM_NL, so treat this as an agent-facing surface, not a described-but-missing button.

How it behaves

What it does

You give Omniscio a natural-language description of an automation — "every Monday morning, check three repos for failing CI and DM me a digest" or "audit each prompt in audit-reports/refactor-prompts/ with a 4-stage pipeline" — and Omniscio returns a draft RecipeConfig plus its reasoning, warnings, and any remaining validation errors. The draft is NOT saved to disk; the caller decides whether to save it (via RECIPE_SAVE_CONFIG), edit it, or retry.

When the drafter refuses

  • Chitchat / non-recipe requests ("what's the weather?", "1+1?")
  • Requests that need capabilities Omniscio doesn't have (live web scraping, arbitrary code execution)

The output is { recipe: null, source: "refused", reasoning: "..." } — caller surfaces the reasoning.

Cost & rate behavior

  • Each turn is one Sonnet Messages API call. Typical one-shot ≈ $0.05–0.15. Refinement loop ≈ $0.20–0.50.
  • maxDrafterCostUsd halts the loop when crossed.
  • The drafter does NOT cache; identical NL inputs produce a fresh call each time. (Caching is a future feature.)
  • Auth: Messages API needs an API key. OAuth-only accounts get auto-fallback to the first API key account. No API key in the vault → handler returns an error envelope.
  • Pattern library: use-recipes.md, RECIPE_AUTHORING_GUIDE.md
  • Roadmap: 2026-04-22-recipe-robustness-roadmap.md Phase 3 item #12
  • Design doc: 2026-04-22-recipe-nl-create-ipc-design.md

Last verified 2026-09-26