#ai agent
65 tutorials, projects and news items tagged ai agent.
Claude Code's experimental Code Mode lets Claude write a short JavaScript snippet that runs several tool calls together. It suits bulk file edits, filtered searches, and parallel commands. The article covers which versions, accounts, and models qualify, how to turn the feature on and off, its limits and permission checks, and a prompt that has Claude Code check whether you can use it.
Beginner tutorial: install and sign in to the ChatGPT desktop app, switch to Codex from the top-left corner, create a practice folder and copy your files into it, set the permission below the input box to Ask for approval (it works inside the folder and asks before going outside), hand it a small spreadsheet cleanup, then request changes in the same chat. Once you're comfortable, paste a tutorial prompt from this site and let it follow along. Includes 2 copy-ready prompts.
Intermediate tutorial: Give your AI agent the free, open-source rea-video tool. Drop in a video, and it splits the video into shots on your own computer and judges each one as live footage, AI-generated, or code-rendered. It also identifies where the voiceover and music come from, then gives a recreation plan for each shot: a shot list for footage to film, prompts and batch plans for AI-generated shots, and an approach for shots that need code. The output is a web report for you and a VIDEO.md for your AI agent. The tutorial covers why an AI agent should review frames for videos downloaded from platforms, how to connect a video-watching model for more accurate judgments, and a copyright reminder to learn the method without reusing the footage. Includes 2 copy-paste prompts.
Beginner tutorial: install SmartSub, a free and open-source desktop app, download a local speech model, turn your video into timed subtitles with bilingual translation, proofread them line by line (with the built-in AI assistant if you like), add dubbing with a local voice or your own cloned voice when you need it, then burn the subtitles into the picture and export the final video. The whole workflow runs free on local models, and your videos are never uploaded. It also covers how your AI agent can process a whole folder at once through MCP. Includes 3 copy-ready prompts.
In about a month, validate a business for a small cost, then decide whether to double down or switch projects based on the ledger.
Beginner tutorial: Give your AI agent the free, open-source lanshu-create-ai-presenter-video Skill, run a free dry test to confirm your setup, choose the explainer route with no on-screen presenter, make silent drafts of all nine visual styles to compare, lock the style and script before voice-over, then have the agent animate each sentence, review the storyboard and key frames before rendering, and check its report on the final video. Also covers the authorization and cost issues of the digital presenter route. Includes 3 copy-ready prompts.
Beginner tutorial: Install Pagekin (formerly AI Novel Writer), a free open-source writing app, and connect it to your own cloud model or a local model through Ollama. Set up the story premise, characters, world and plot outline, then plan each chapter's blueprint and write, review, revise and finalize one chapter at a time, backing up along the way. The app sends the AI only what each chapter needs, which reduces the forgetting that happens in long books, and the guide includes 3 prompts you can copy and paste.
A repeatable AI product video production line where you can swap products and shots, and spend follows the ad data.
Intermediate tutorial: Add the open-source, free claude-animated-shorts skill to Claude Code and use three.js and Remotion to render a roughly one-minute vertical cartoon story short on your own computer. Flat 2D characters move through a colorful 3D town, with ink outlines, depth of field, camera moves, music and sound effects, and no AI video model is involved. The guide includes three copy-ready prompts that take you from setup to the final render.
A work folder that lets you resume at any time, a source library, a first article in your drafts, and a posting assistant you only need to check each day.
OpenCode has released browse 0.1.0, a full Rust rewrite that drops Node.js and Playwright and, according to OpenCode, uses 25 times less memory. Your AI agent can work in your own Chrome, reading pages, clicking, filling forms and taking screenshots in background tabs, and it hands control back to you at logins and payments.
Beginner tutorial: Download Booth, a free open-source app, and make AI voiceovers on your own computer with the ZONOS3-Mini model. Start with preset voices to tune emotion and speed, generate several takes per line and keep the best one, then clone your own voice and let your AI agent voice a whole script in batches. Your scripts and recordings stay on your computer, with no sign-up or subscription. Includes 2 copy-ready prompts.
Each common office task keeps one skill that has been tested on real material, and your agent picks the right one next time.
Beginner tutorial: Install the free, open-source ffmpeg-skill so your AI agent can edit videos on your own computer, with no API key and no uploads. It can trim silent gaps, reframe landscape footage to vertical, add subtitles and adjust volume. Before you review the result, it checks size, loudness and duration against platform standards, and you confirm the crops and subtitles from screenshots.
Intermediate tutorial: Have your AI agent check your VRAM and disk space read-only to see which tier your computer can run, then pick one of three paths based on your footage (MiniMax H3 photo talking, LatentSync lip-sync, or LongCat-Video-Avatar long narration) and have it install the chosen one and run a clip under 10 seconds. Rewrite the prompt for a fixed camera and low hands, then run several seeds and keep the best take. Includes a creator's measured timing comparison, a settings cheat sheet, and 3 copy-ready prompts.
The OpenDesign team open-sourced OpenMotion, an Apache-2.0 alternative to CapCut and Claude Motion. Your AI agent breaks a reference video into shots, rebuilds it in HTML with your brand, copy, and assets, and you fine-tune and export an MP4 in the built-in editor, with support for more than 20 agents including Codex and Claude Code.
Ready-made transitions, title effects and talking characters for code-built videos, with licenses checked for commercial use and attribution. Includes a prompt that lets your AI agent pick and wire them in.
Advanced tutorial: Use cwyz.com as a live example to set up IndexNow with an AI agent as your guide. Generate a key file in the site root, test one URL to learn what the status codes mean, push only the sitemap pages that changed in batches, and hook pushes into publishing so each new post goes out automatically. Then confirm the submissions in Bing Webmaster Tools. One push reaches Bing, Yandex, Naver and other participating engines, but Google is not supported. Includes 3 copy-ready prompts.
A cheap model has a low unit price, but reworking back and forth can cost more in total. Give a big feature with clear boundaries to the strong model in one go, and split out small fixed-step jobs for the cheap one. User feedback, not verified yet.
Google's official Chrome DevTools MCP lets an AI agent record page performance, check LCP, INP and CLS, run Lighthouse and point to problem elements, take memory snapshots to find leaks, and save screen recordings as mp4. It also ships a command-line tool for scheduled jobs. Includes a prompt that has your AI agent check your site and compare before and after.
