--- title: Showing what you mean in the age of AI | daily.dev description: Loom is adding tools to brief AI agents with screen recordings instead of typed prompts. Record for agent turns a screen recording into a structured prompt,... canonical: https://daily.dev/posts/showing-what-you-mean-in-the-age-of-ai-lxv76bj2n twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Showing what you mean in the age of AI | daily.dev og:description: Loom is adding tools to brief AI agents with screen recordings instead of typed prompts. Record for agent turns a screen recording into a structured prompt,... og:url: https://daily.dev/posts/showing-what-you-mean-in-the-age-of-ai-lxv76bj2n og:image: https://api.daily.dev/og/posts/lxv76BJ2n.png og:image:alt: Showing what you mean in the age of AI og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Showing what you mean in the age of AI **[Atlassian](https://daily.dev/sources/atlassian)** · 7 min read · 1 upvotes · 3 comments ## Summary Loom is adding tools to brief AI agents with screen recordings instead of typed prompts. Record for agent turns a screen recording into a structured prompt, capturing transcript, clicked UI elements, and visited links. The new Loom for Mac adds 4K-by-default recording, a Selection tool to pin UI elements, and a Magic pen for annotations. Loom now supports the Model Context Protocol, so agents like Claude, Cursor, VS Code, Codex and Rovo can pull a video's transcript, metadata and action items through Atlassian MCP. Record for agent is in open beta for Loom Business + AI and Enterprise plans. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How does Loom's Record for agent feature turn a screen recording into an AI prompt? It captures the transcript of narrated instructions, screenshots of on-screen elements clicked during recording, and links navigated to while sharing the screen, then turns all of that into a structured set of instructions an agent can act on. The new Loom for Mac also adds a Selection tool to pin a precise UI element and a Magic pen for freehand annotations. _Teams piloting agent handoffs can compare tools like this one on daily.dev before adopting a workflow._ ### Which AI coding agents can access Loom recordings through Atlassian MCP? Claude, Cursor, VS Code, Codex and Rovo can work with Loom directly through Loom tools built into Atlassian MCP, pulling a video's transcript, metadata and action items, adding timestamped comments, and sharing or organizing recordings within existing Loom permissions. Atlassian MCP v2 users get this automatically once their Loom workspace is linked; v1 users must upgrade their connector. _Developers wiring up MCP integrations can track which agents support new tooling via daily.dev._ ## Community discussion Top comments from developers on daily.dev. **@diyweb** · 0 upvotes > Keeping a wrong click followed by a correction is a good point, because that context is hard to write in a typed prompt. I'm curious how Record for agent handles long recordings: does it summarize the steps, or pass the full transcript and click list as is? **@demovexa** · 0 upvotes > Recording for an agent and recording for a person need different habits. For a human viewer you cut the pauses and the wrong clicks. For an agent brief, a wrong click followed by "no, not that one" is useful context, so I'd keep it in. The thing I'd watch is what's on screen. A recording that becomes a prompt with the transcript and clicked elements will carry whatever customer data was visible, so the sample data rule that applies to demo videos probably applies here too. **@bickov** · 0 upvotes > A 4K frame is 3840 px wide. Claude reads images up to 2576 px on the long edge, so any frame an agent pulls from the recording gets scaled to about two thirds anyway, and small UI text suffers first. The transcript and the clicked element names do not have that problem. Does Record for agent keep the text of each clicked element, or only where the click landed? ## Similar posts on daily.dev - [Give AI agents the full picture with Loom video prompts](https://daily.dev/posts/give-ai-agents-the-full-picture-with-loom-video-prompts-xcednkc9s) · Atlassian · 0 upvotes · 0 comments - [Meet the all-new Loom for Mac: faster, sharper, and delightful by default](https://daily.dev/posts/meet-the-all-new-loom-for-mac-faster-sharper-and-delightful-by-default-y2zihafgk) · Atlassian · 0 upvotes · 0 comments - [I replaced Loom with an open-source screen recorder and it's better than Loom ever was](https://daily.dev/posts/i-replaced-loom-with-an-open-source-screen-recorder-and-it-s-better-than-loom-ever-was-vkzv4ar2s) · XDA Developers · 0 upvotes · 0 comments - [How to automate product workflows with AI-powered video: A product manager’s guide](https://daily.dev/posts/how-to-automate-product-workflows-with-ai-powered-video-a-product-manager-s-guide-vfnahovez) · Atlassian · 0 upvotes · 0 comments - [Annotate: Screen recording as prompts](https://daily.dev/posts/annotate-screen-recording-as-prompts-aghrabztu) · Product Hunt · 1 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#mcp](https://daily.dev/tags/mcp), [#atlassian](https://daily.dev/tags/atlassian), [#jira](https://daily.dev/tags/jira) [View this post on daily.dev](https://daily.dev/posts/showing-what-you-mean-in-the-age-of-ai-lxv76bj2n) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Showing what you mean in the age of AI","url":"https://daily.dev/posts/showing-what-you-mean-in-the-age-of-ai-lxv76bj2n","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/showing-what-you-mean-in-the-age-of-ai-lxv76bj2n"},"datePublished":"2026-10-22T15:57:00.374Z","dateModified":"2026-10-22T16:27:58.025Z","description":"Loom is adding tools to brief AI agents with screen recordings instead of typed prompts. Record for agent turns a screen recording into a structured prompt,...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/f5becfe92d44c7524f61665ec88c6346?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/f5becfe92d44c7524f61665ec88c6346?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Atlassian","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Atlassian","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/44750d8e66de4411bb477159b321bb43","url":"https://daily.dev/sources/atlassian"},"commentCount":3,"discussionUrl":"https://daily.dev/posts/showing-what-you-mean-in-the-age-of-ai-lxv76bj2n","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":3}],"keywords":"ai-agents,mcp,atlassian,jira","timeRequired":"PT7M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Atlassian","item":"https://daily.dev/sources/atlassian"},{"@type":"ListItem","position":3,"name":"Showing what you mean in the age of AI"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/showing-what-you-mean-in-the-age-of-ai-lxv76bj2n","comment":[{"@type":"Comment","text":"Keeping a wrong click followed by a correction is a good point, because that context is hard to write in a typed prompt. I’m curious how Record for agent handles long recordings: does it summarize the steps, or pass the full transcript and click list as is?","datePublished":"2026-10-24T19:08:10.789Z","url":"https://daily.dev/posts/lxv76BJ2n#c-rslcr92nY","author":{"@type":"Person","name":"DiyWeb","url":"https://daily.dev/diyweb","image":"https://media.daily.dev/image/upload/s--dgS6jDeN--/f_auto/v1790879420/avatars/avatar_774zDwMaRTNS1fv5OKStX?_a=BAMAMicg0"}},{"@type":"Comment","text":"Recording for an agent and recording for a person need different habits. For a human viewer you cut the pauses and the wrong clicks. For an agent brief, a wrong click followed by “no, not that one” is useful context, so I’d keep it in. The thing I’d watch is what’s on screen. A recording that becomes a prompt with the transcript and clicked elements will carry whatever customer data was visible, so the sample data rule that applies to demo videos probably applies here too.","datePublished":"2026-10-23T03:00:02.594Z","url":"https://daily.dev/posts/lxv76BJ2n#c-juUFuYFIT","author":{"@type":"Person","name":"DemoVexa","url":"https://daily.dev/demovexa","image":"https://media.daily.dev/image/upload/s--I0Ym8soj--/f_auto,pg_1/v1792077531/avatars/avatar_F6wD6jMSPjwEMcztjwnyE?_a=BAMAMicg0"}},{"@type":"Comment","text":"A 4K frame is 3840 px wide. Claude reads images up to 2576 px on the long edge, so any frame an agent pulls from the recording gets scaled to about two thirds anyway, and small UI text suffers first. The transcript and the clicked element names do not have that problem. Does Record for agent keep the text of each clicked element, or only where the click landed?","datePublished":"2026-10-23T06:56:06.400Z","url":"https://daily.dev/posts/lxv76BJ2n#c-iFXB5yxPV","author":{"@type":"Person","name":"ab","url":"https://daily.dev/bickov","image":"https://lh3.googleusercontent.com/a/ACg8ocKwOmJ31UWqM4ZXB5ms-yRdFR4BYzyb7FPBRpg1O6DV0MIAF2EP=s96-c"}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/showing-what-you-mean-in-the-age-of-ai-lxv76bj2n#faq","mainEntity":[{"@type":"Question","name":"How does Loom's Record for agent feature turn a screen recording into an AI prompt?","acceptedAnswer":{"@type":"Answer","text":"It captures the transcript of narrated instructions, screenshots of on-screen elements clicked during recording, and links navigated to while sharing the screen, then turns all of that into a structured set of instructions an agent can act on. The new Loom for Mac also adds a Selection tool to pin a precise UI element and a Magic pen for freehand annotations. Teams piloting agent handoffs can compare tools like this one on daily.dev before adopting a workflow."}},{"@type":"Question","name":"Which AI coding agents can access Loom recordings through Atlassian MCP?","acceptedAnswer":{"@type":"Answer","text":"Claude, Cursor, VS Code, Codex and Rovo can work with Loom directly through Loom tools built into Atlassian MCP, pulling a video's transcript, metadata and action items, adding timestamped comments, and sharing or organizing recordings within existing Loom permissions. Atlassian MCP v2 users get this automatically once their Loom workspace is linked; v1 users must upgrade their connector. Developers wiring up MCP integrations can track which agents support new tooling via daily.dev."}}]} ```