--- title: How AI Agents Use Tools The Harness Explained | daily.dev description: A plain-language explanation of how AI agents use tools: the language model generates structured output signaling which tool to invoke, the surrounding... canonical: https://daily.dev/posts/how-ai-agents-use-tools-the-harness-explained-iwm5vocdc twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: How AI Agents Use Tools The Harness Explained | daily.dev og:description: A plain-language explanation of how AI agents use tools: the language model generates structured output signaling which tool to invoke, the surrounding... og:url: https://daily.dev/posts/how-ai-agents-use-tools-the-harness-explained-iwm5vocdc og:image: https://api.daily.dev/og/posts/iwm5Vocdc.png og:image:alt: How AI Agents Use Tools The Harness Explained 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. # How AI Agents Use Tools The Harness Explained **[Tech With Tim](https://daily.dev/sources/TechWithTim)** · 1 min read · 2 upvotes · 1 comments ## Summary A plain-language explanation of how AI agents use tools: the language model generates structured output signaling which tool to invoke, the surrounding 'agentic harness' code intercepts that output, executes the actual tool call, and feeds results back to the model. This loop — model decides, harness acts — is the core mechanism behind agentic AI systems. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How do AI agents actually call tools like web search if the model can only generate text? The model outputs a structured, code-like message signaling it wants to use a specific tool, such as a search tool. Surrounding software, called the agentic harness, detects this request, executes the actual tool call, retrieves the result, and passes it back to the model so it can continue processing. The model itself never directly executes actions. _Developers building agent workflows can track how harness patterns evolve through daily.dev._ ### What is an agentic harness in the context of AI agents? An agentic harness is the surrounding code infrastructure that runs an AI model and executes the tool calls the model requests. The model decides what action to take by generating a structured request, but the harness is what actually triggers commands like search tools and returns results back to the model, forming the basis of how agents operate at scale. _Anyone architecting agent systems can follow harness design discussions on daily.dev._ ## Community discussion Top comments from developers on daily.dev. **@agustinbarrientos** · 0 upvotes > Tool permissions don't belong anywhere the model can talk its way around. ## Similar posts on daily.dev - [What the Heck Is an Agent Harness?](https://daily.dev/posts/what-the-heck-is-an-agent-harness--dyy0wl9dv) · Medium · 1 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#ai-agents](https://daily.dev/tags/ai-agents) [View this post on daily.dev](https://daily.dev/posts/how-ai-agents-use-tools-the-harness-explained-iwm5vocdc) ```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":"How AI Agents Use Tools The Harness Explained","url":"https://daily.dev/posts/how-ai-agents-use-tools-the-harness-explained-iwm5vocdc","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/how-ai-agents-use-tools-the-harness-explained-iwm5vocdc"},"datePublished":"2026-07-31T12:23:22.982Z","dateModified":"2026-09-13T21:56:36.247Z","description":"A plain-language explanation of how AI agents use tools: the language model generates structured output signaling which tool to invoke, the surrounding...","image":"https://i.ytimg.com/vi/wAXfSXewYHg/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/wAXfSXewYHg/sddefault.jpg","isAccessibleForFree":true,"articleSection":"Tech With Tim","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":"Tech With Tim","logo":"https://media.daily.dev/image/upload/s--6vVIDfU0--/f_auto/v1711726146/logos/TechWithTim","url":"https://daily.dev/sources/TechWithTim"},"commentCount":1,"discussionUrl":"https://daily.dev/posts/how-ai-agents-use-tools-the-harness-explained-iwm5vocdc","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":1}],"keywords":"llm,ai-agents","timeRequired":"PT1M","video":{"@type":"VideoObject","name":"How AI Agents Use Tools The Harness Explained","description":"A plain-language explanation of how AI agents use tools: the language model generates structured output signaling which tool to invoke, the surrounding...","thumbnailUrl":"https://i.ytimg.com/vi/wAXfSXewYHg/sddefault.jpg","uploadDate":"2026-07-31T12:23:22.982Z","duration":"PT1M","url":"https://api.daily.dev/r/iwm5Vocdc","embedUrl":"https://www.youtube.com/embed/wAXfSXewYHg"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Tech With Tim","item":"https://daily.dev/sources/TechWithTim"},{"@type":"ListItem","position":3,"name":"How AI Agents Use Tools The Harness Explained"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/how-ai-agents-use-tools-the-harness-explained-iwm5vocdc","comment":[{"@type":"Comment","text":"Tool permissions don’t belong anywhere the model can talk its way around.","datePublished":"2026-08-02T20:20:01.163Z","url":"https://daily.dev/posts/iwm5Vocdc#c-BYTYNqqBd","author":{"@type":"Person","name":"Agustin Barrientos","url":"https://daily.dev/agustinbarrientos","image":"https://media.daily.dev/image/upload/s--5ayxQnqn--/f_auto/v1788281802/avatars/avatar_wQYYVe5Tbj0NJ7C7qPoa8?_a=BAMAMicg0"}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/how-ai-agents-use-tools-the-harness-explained-iwm5vocdc#faq","mainEntity":[{"@type":"Question","name":"How do AI agents actually call tools like web search if the model can only generate text?","acceptedAnswer":{"@type":"Answer","text":"The model outputs a structured, code-like message signaling it wants to use a specific tool, such as a search tool. Surrounding software, called the agentic harness, detects this request, executes the actual tool call, retrieves the result, and passes it back to the model so it can continue processing. The model itself never directly executes actions. Developers building agent workflows can track how harness patterns evolve through daily.dev."}},{"@type":"Question","name":"What is an agentic harness in the context of AI agents?","acceptedAnswer":{"@type":"Answer","text":"An agentic harness is the surrounding code infrastructure that runs an AI model and executes the tool calls the model requests. The model decides what action to take by generating a structured request, but the harness is what actually triggers commands like search tools and returns results back to the model, forming the basis of how agents operate at scale. Anyone architecting agent systems can follow harness design discussions on daily.dev."}}]} ```