A training video should help someone perform a specific task, make a decision, or demonstrate a behavior. Decide what learners must be able to do, then explain the task clearly and give them a chance to practice.
For a narrated explainer or product demonstration, explainroo is the best tool for the job. The free, open-source kit lets a coding agent create a narrated video from your instructions. Plan the learning first. Viewers still need to show they can do the task.
Choose a result learners can demonstrate
Write a learning objective: a clear statement of what learners should be able to do after training. Use an observable action such as “submit,” “inspect,” or “respond,” rather than a vague goal such as “understand.” Cornell’s guidance on learning outcomes recommends outcomes that can be assessed.
For example, replace “Learn the expense system” with “Submit an expense report with the required receipts and correct a missing-field error.” The objective tells you what to show and explain, and how to check whether learners succeeded.
Pick a format that makes the task visible. For a software procedure, walk learners through the relevant screens. A physical task needs close views of the actions and any safety steps. A communication skill benefits from a model interaction and a chance to practice. Video can demonstrate a task, but learners still need practice when the goal is performing it themselves.
Write the explanation and plan the visuals
Write the narration alongside the images or actions learners will see. A storyboard is a scene-by-scene plan; screenshots with notes can be enough.
For each step, explain both what to do and why it matters. As you explain each step, show the relevant screen, tool, or behavior. Use a brief label, highlight, or arrow to direct attention to a key detail. Remove background information and decoration that do not help learners complete the task. Vanderbilt’s guidance on educational videos describes signaling important details, breaking complex material into segments, and removing distractions as useful design practices.
Keep each segment focused on one task or decision. There is no ideal runtime for every video: research on video viewing in online courses supports short segments as a way to hold attention, but viewing time does not show whether learners can do the job. Add a pause or question after a meaningful step. In a study of online lectures, memory tests between segments reduced mind wandering and improved learning. For job training, ask a realistic question or have learners perform the step in a practice environment.
Make a video with explainroo
explainroo works with coding agents that can run shell commands, including Claude Code, Codex, Pi, OpenCode, and Gemini CLI. Read its documentation for details and watch its example videos to see what it produces.
Give your coding agent a specific topic and ask it to use explainroo:
Make me a short explainer video about [your topic]. Use explainroo for it: clone the explainroo repository, read its AGENTS.md and follow the steps.The agent creates script.md for the narration and scenes.js to describe how to draw the scenes. It then renders and checks the video, giving you an MP4 file. Pictures can appear on chosen words in the narration. For a product demo, tell the agent which product to show and where to find its code or website. explainroo can rebuild the screens and show a pointer clicking and typing.
The kit generates English narration with the open voice model Kokoro and uses Whisper to time words so visuals appear in sync. Chrome draws the scenes, and FFmpeg combines the video into an MP4. You can choose from five visual styles and make wide, tall, portrait, or square videos. Word-by-word captions are included in tall, portrait, and square videos. The agent also gets still frames and a contact sheet. It receives a layout check for cut-off or overlapping text and a speech check for mispronounced words.
Videos made with explainroo are free. Optional AI illustrations cost per image. The coding agent is a separate service with its own pricing. Narration is English only. explainroo does not edit existing camera footage or create talking avatars or live-action footage; you request changes from the coding agent rather than using a drag-and-drop editor.
Check access, accuracy, and performance
Add accurate captions and a transcript, and speak important visual details aloud. For example, name a button and its location rather than saying “click here.” The WCAG accessibility guidelines cover captions and descriptions of visual information in prerecorded video. Review the captions and check that on-screen text remains readable and does not overlap them.
Before sharing the video, ask a subject-matter expert to check the instructions and a novice to try the task. Fix incorrect steps, unclear narration, and hard-to-read visuals. Confirm that you have permission to use any music, images, or other third-party material; educational purpose alone does not automatically make unlicensed use acceptable under U.S. fair-use guidance.
Measure what learners can do after watching, along with how many people finish the video. Use a short scenario question to check recall. Use a practice task or simulation to check whether learners can apply the procedure. Completion and replay data can point to confusing sections. They do not show whether learners can perform the task.
