Atom
English

Writing Effective Prompts

A great prompt sounds like clear direction you’d give a teammate. Keep it simple, specific, and conversational.

Start with the essentials

  • Say what you want — “Add a one-second cross-dissolve between the selected clips.”
  • Name the target — Mention the sequence, track, clip, bin, or marker you have in mind.
  • Give numbers when they matter — frame counts, positions, colors, durations.

Paint the picture

  • Add the “why” when it helps: “Create a vertical version so this interview is ready for social.”
  • Reference existing elements: “Match the title style already used in the opening sequence.” Atom reads the project and follows the breadcrumbs.

Provide visual context

Attach files to give Atom visual reference when words alone aren’t enough:

  • Editorial references — Attach a script, paper edit, or client notes and ask Atom to shape the sequence around them
  • Visual references — Drag in a style frame and ask Atom to match its titles, colors, or layout
  • Technical specs — Attach a delivery brief with exact durations, aspect ratios, captions, or export requirements
  • Audio context — Attach voice audio for transcription, timing, or marker suggestions

Example prompts with references:

Use this interview transcript to build a selects sequence around the answers marked in yellow
Use this delivery brief to create the required sequence versions and exports

Files you attach are sent to the AI model through your selected provider. Atom also sends sequence frame previews to give the AI visual context. Full-resolution source footage and final exports stay local.

Build momentum

  • Break bigger goals into short prompts and confirm each step.
  • Stay in the same conversation when you’re iterating. Atom remembers what just happened.
  • Need a different approach? Ask for options (“Suggest three variations on this animation”).

Try voice prompting

LLMs thrive on details, and it’s much easier to provide context by speaking than typing. Voice input lets you naturally describe what’s in your head without condensing your thoughts. Give handy.computer a try. It’s a free, open-source transcriber that runs locally on your machine.

Match the model to the task

Use Models & Reasoning beside the attachment button. Lower reasoning works well for simple batch changes; raise it when the request needs more planning or judgment.

Set persistent instructions

Open Settings → AI → Custom instructions to add guidance that Atom follows in every conversation. Use this for preferences you don't want to repeat, like naming conventions, preferred tools, easing styles, or workflow patterns.

When results miss the mark

  1. Add more detail to your prompt.
  2. Tighten the scope (“only the selected clips”, “just the active sequence”).
  3. Rephrase with the outcome up front.

Quick reminders

  • Use clear nouns, numbers, and intent.
  • Reference actual item names or patterns Atom can search for.
  • Skip vague phrases (“fix it”, “make it better”).