How to Generate a Specification with the Guided Intake Form

By Paul Flanders · · Beginner

Learn how to use the Guided Intake Form to create a detailed software specification, ensuring all necessary decisions are made upfront. This beginner-friendly guide outlines the steps and options available for generating a structured brief.

Estimated time: 4 minutes
Difficulty: Beginner

Why you'd use this

Typing a paragraph often leaves out the decisions a planner needs: language, framework, storage, authentication, target platform, packaging and testing. The specification form asks for each of those up front, with "Let the planner choose" as the default, and turns your answers into a structured brief. Plan Mode then starts from that brief, so the first planning turn is already grounded and fewer clarifying questions come back.

Before you start

Permissions required:

  • Same as Plan Mode: write access to the workspace.

You'll need:

  • A list of the key features the software must have.
  • Any hard constraints, for example offline only, no external services, or a required language.

Steps

  1. Press / in the composer and choose Generate specification (guided intake), or run eLLM: Generate Specification.
  2. A form opens in the chat panel. Choose the Type: web app, API or backend service, CLI tool, desktop app, mobile app, library, game, or other.
  3. Choose the Language and type a Framework if you have one in mind, or leave the planner to decide.
  4. Set UI, Data storage, Authentication, Target platform, Packaging and Testing. Each has a "Let the planner choose" option.
  5. In Key features / requirements, list the main things it must do, one per line.
  6. In Constraints / non-goals, note anything it must not do.
  7. Pick the Build mode: Guided, Autopilot or Assist.
  8. Click Start planning. The brief is sent and Plan Mode begins at the chosen level.

What you should see

The brief appears as your first message, written as clean markdown with blank and default fields left out. The planner may still ask questions for anything the form did not cover, using the same inline question form as Plan Mode.

Troubleshooting

  • You clicked Cancel by mistake: reopen the form from the / menu; the fields are not saved between openings, so keep your feature list in a scratch file if it is long.
  • The planner chose a stack you did not want: exit Plan Mode, reopen the form and set the Language and Framework fields explicitly.

Related guides

  • How to Plan and Build a Feature with Plan Mode
  • How to Use the Standalone Planning Commands

person people found this useful.

Related

Knowledge base 11 Sep 2026

How to Send Identity and Intent Headers to the Gateway

Learn how to configure identity and intent headers for a shared gateway using eLLM Code. This guide explains how to adjust settings in VS Code, ensuring requests are routed correctly and usage is attributed accurately.

Knowledge base 11 Sep 2026

How to Tune Generation Settings for Your Model

Learn how to fine-tune generation settings for local and self-hosted models to improve performance and accuracy. Adjust parameters like repetition, prompt size, and command timeouts to better match your model's behaviour.

Knowledge base 11 Sep 2026

How to Run the Benchmark Suite

This article provides a detailed guide on running a benchmark suite to test model configurations, including setup requirements, steps for execution, and troubleshooting tips. It's aimed at advanced users looking to evaluate model capabilities before deployment.

Expand 29 Jul 2026

How AI can support pupils with SEND

Explore how AI tools are aiding pupils with special educational needs and disabilities by simplifying text, breaking down tasks, and enhancing accessibility. Learn about the considerations schools should make before implementation.

Expand 29 Jul 2026

AI policy and practice in further education

Further education colleges need distinct AI policies due to their unique mix of vocational courses, diverse age groups, and specific funding and inspection requirements. This guidance highlights the importance of tailored AI approaches in FE settings.