--- title: Build Applications on NVIDIA BlueField Faster with... description: NVIDIA has released DOCA AI agent skills on GitHub, a structured, machine-readable format (SKILL.md files) that gives general-purpose AI coding agents verified... canonical: https://daily.dev/posts/build-applications-on-nvidia-bluefield-faster-with-nvidia-doca-agent-skills-fuusf1nft twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Build Applications on NVIDIA BlueField Faster with NVIDIA DOCA Agent Skills | daily.dev og:description: NVIDIA has released DOCA AI agent skills on GitHub, a structured, machine-readable format (SKILL.md files) that gives general-purpose AI coding agents verified... og:url: https://daily.dev/posts/build-applications-on-nvidia-bluefield-faster-with-nvidia-doca-agent-skills-fuusf1nft og:image: https://api.daily.dev/og/posts/FuUSF1nFT.png og:image:alt: Build Applications on NVIDIA BlueField Faster with NVIDIA DOCA Agent Skills 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. # Build Applications on NVIDIA BlueField Faster with NVIDIA DOCA Agent Skills **[NVIDIA Developer](https://daily.dev/sources/nvidiadev)** · 9 min read · 0 upvotes · 0 comments ## Summary NVIDIA has released DOCA AI agent skills on GitHub, a structured, machine-readable format (SKILL.md files) that gives general-purpose AI coding agents verified API signatures, hardware capability requirements, and build constraints for developing on NVIDIA BlueField DPUs. In a 65-prompt evaluation, agents without skills satisfied only 19% of graded checklist items, commonly misusing APIs (59/65 prompts), failing to verify hardware capabilities (46/65), and routing to wrong tools (39/65). With skills loaded, agents satisfied 100% of checklist items across all prompts. A side-by-side demo building an RDMA traffic program showed the skills-equipped agent used 73% less handwritten code and roughly half the hardware commands. Skills cover DOCA Flow, GPUNetIO, PCC, RDMA, and other library components, and are available now in the NVIDIA/skills GitHub repository. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What are NVIDIA DOCA AI agent skills and how do they improve AI coding agents working with BlueField DPUs? DOCA AI agent skills are a standardized, open SKILL.md format that gives AI coding agents verified API signatures, hardware capability requirements, and build constraints for NVIDIA DOCA development on BlueField DPUs. They cover components like DOCA Flow, GPUNetIO, PCC, and RDMA, letting agents reason against real API contracts instead of guessing from general training data. They are available on the NVIDIA/skills GitHub repo. _daily.dev surfaces releases like this for engineers wiring AI agents into specialized hardware workflows._ ### How much do agent skills improve AI coding accuracy on NVIDIA DOCA tasks compared to a general-purpose agent? In a 65-prompt evaluation of real DOCA developer tasks, agents without skills satisfied only 19% of graded checklist items, while agents with DOCA skills satisfied 100% across all prompts. Common failures without skills included API misuse (59/65 prompts), unverified hardware capability (46/65), wrong tool routing (39/65), and skipped smoke tests (34/65). _track measurable gains like this when evaluating which agent skills are worth adopting on daily.dev._ ### How much less code does an AI agent need to write a DOCA RDMA program when using DOCA agent skills versus a general-purpose agent? An agent equipped with DOCA skills used 73% less handwritten code (189 lines versus 695) and about half the hardware commands (20 versus 37, a 46% reduction) to build a program sending real RDMA traffic on BlueField-3, compared to an agent without the skills, while both ultimately succeeded at the task. _developers benchmarking agent efficiency on infrastructure code can follow results like these on daily.dev._ ## Similar posts on daily.dev - [Scaling Agentic AI Factories Through Extreme Co-Design with NVIDIA BlueField](https://daily.dev/posts/scaling-agentic-ai-factories-through-extreme-co-design-with-nvidia-bluefield-keqdn8r6u) · NVIDIA Developer · 1 upvotes · 0 comments - [Advancing AI Infrastructure for Agentic AI with NVIDIA DOCA In-Silicon Security](https://daily.dev/posts/advancing-ai-infrastructure-for-agentic-ai-with-nvidia-doca-in-silicon-security-kzllhlca9) · NVIDIA Developer · 0 upvotes · 0 comments - [NVIDIA BlueField-Powered Cybersecurity and Acceleration Arrive on NVIDIA Enterprise AI Factory Validated Design](https://daily.dev/posts/nvidia-bluefield-powered-cybersecurity-and-acceleration-arrive-on-nvidia-enterprise-ai-factory-valid-scewskfys) · NVIDIA · 0 upvotes · 0 comments - [NVIDIA-Verified Agent Skills Provide Capability Governance for AI Agents](https://daily.dev/posts/nvidia-verified-agent-skills-provide-capability-governance-for-ai-agents-ynkbw6vjp) · NVIDIA Developer · 0 upvotes · 0 comments --- [View this post on daily.dev](https://daily.dev/posts/build-applications-on-nvidia-bluefield-faster-with-nvidia-doca-agent-skills-fuusf1nft) ```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":"Build Applications on NVIDIA BlueField Faster with NVIDIA DOCA Agent Skills","url":"https://daily.dev/posts/build-applications-on-nvidia-bluefield-faster-with-nvidia-doca-agent-skills-fuusf1nft","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/build-applications-on-nvidia-bluefield-faster-with-nvidia-doca-agent-skills-fuusf1nft"},"datePublished":"2026-10-01T18:16:52.420Z","dateModified":"2026-10-01T18:31:06.950Z","description":"NVIDIA has released DOCA AI agent skills on GitHub, a structured, machine-readable format (SKILL.md files) that gives general-purpose AI coding agents verified...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/467528a65c6b11dd0a92db8694605975?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/467528a65c6b11dd0a92db8694605975?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"NVIDIA Developer","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":"NVIDIA Developer","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/86e45aab42ba48ce83103d01b1119910","url":"https://daily.dev/sources/nvidiadev"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/build-applications-on-nvidia-bluefield-faster-with-nvidia-doca-agent-skills-fuusf1nft","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"","timeRequired":"PT9M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"NVIDIA Developer","item":"https://daily.dev/sources/nvidiadev"},{"@type":"ListItem","position":3,"name":"Build Applications on NVIDIA BlueField Faster with NVIDIA DOCA Agent Skills"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/build-applications-on-nvidia-bluefield-faster-with-nvidia-doca-agent-skills-fuusf1nft#faq","mainEntity":[{"@type":"Question","name":"What are NVIDIA DOCA AI agent skills and how do they improve AI coding agents working with BlueField DPUs?","acceptedAnswer":{"@type":"Answer","text":"DOCA AI agent skills are a standardized, open SKILL.md format that gives AI coding agents verified API signatures, hardware capability requirements, and build constraints for NVIDIA DOCA development on BlueField DPUs. They cover components like DOCA Flow, GPUNetIO, PCC, and RDMA, letting agents reason against real API contracts instead of guessing from general training data. They are available on the NVIDIA/skills GitHub repo. daily.dev surfaces releases like this for engineers wiring AI agents into specialized hardware workflows."}},{"@type":"Question","name":"How much do agent skills improve AI coding accuracy on NVIDIA DOCA tasks compared to a general-purpose agent?","acceptedAnswer":{"@type":"Answer","text":"In a 65-prompt evaluation of real DOCA developer tasks, agents without skills satisfied only 19% of graded checklist items, while agents with DOCA skills satisfied 100% across all prompts. Common failures without skills included API misuse (59/65 prompts), unverified hardware capability (46/65), wrong tool routing (39/65), and skipped smoke tests (34/65). track measurable gains like this when evaluating which agent skills are worth adopting on daily.dev."}},{"@type":"Question","name":"How much less code does an AI agent need to write a DOCA RDMA program when using DOCA agent skills versus a general-purpose agent?","acceptedAnswer":{"@type":"Answer","text":"An agent equipped with DOCA skills used 73% less handwritten code (189 lines versus 695) and about half the hardware commands (20 versus 37, a 46% reduction) to build a program sending real RDMA traffic on BlueField-3, compared to an agent without the skills, while both ultimately succeeded at the task. developers benchmarking agent efficiency on infrastructure code can follow results like these on daily.dev."}}]} ```