--- title: How to make enterprise APIs agent-ready | daily.dev description: Exposing enterprise APIs to AI agents as one MCP tool per operation breaks down once an API has hundreds of operations. Marco Palladino describes 'code mode'... canonical: https://daily.dev/posts/how-to-make-enterprise-apis-agent-ready-rmm8dqq35 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: How to make enterprise APIs agent-ready | daily.dev og:description: Exposing enterprise APIs to AI agents as one MCP tool per operation breaks down once an API has hundreds of operations. Marco Palladino describes 'code mode'... og:url: https://daily.dev/posts/how-to-make-enterprise-apis-agent-ready-rmm8dqq35 og:image: https://api.daily.dev/og/posts/RmM8DQq35.png og:image:alt: How to make enterprise APIs agent-ready 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 to make enterprise APIs agent-ready **[InfoWorld](https://daily.dev/sources/infoworld)** · 7 min read · 0 upvotes · 0 comments ## Summary Exposing enterprise APIs to AI agents as one MCP tool per operation breaks down once an API has hundreds of operations. Marco Palladino describes 'code mode' as a better pattern: the agent writes short code snippets against an SDK-like interface, and only the final result returns to the model's context. In testing, a workflow using about 120,000 tokens with standard tool calling dropped to roughly 1,200 tokens in code mode. He argues agent traffic should route through the existing API gateway so authentication, rate limiting, and audit logging apply automatically. Kong built this into Kong Context Mesh, now generally available as part of Kong AI Management. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### why does exposing every API operation as a separate MCP tool break down for large enterprise APIs It fails because loading hundreds of tool definitions into the model's context before any work happens increases token cost, degrades tool selection accuracy as the agent confuses similar-looking options, and adds latency from excessive back-and-forth between the model and tools. A typical enterprise CRM or ERP API can have hundreds of operations, overwhelming an agent's reasoning capacity. _Platform teams weighing MCP integration patterns can track real-world approaches like this on daily.dev._ ### what is code mode in the context of AI agents and MCP servers Code mode is a pattern where a server exposes an API as a single SDK-like interface instead of a flat list of MCP tools, so the agent writes short code snippets against that interface that run in a sandbox, with only the final result returned to the model's context. This lets one snippet chain multiple API calls with loops and error handling, cutting round trips and token usage dramatically, for example from about 120,000 tokens to about 1,200 in one tested workflow. _Developers evaluating agent tool-calling architectures can follow emerging patterns like code mode on daily.dev._ ### how should agent traffic be governed when agents call production enterprise APIs Agent calls should route through the same API gateway that governs every other client, so existing authentication, authorization, rate limiting, and audit logging policies apply automatically rather than being recreated per MCP server integration. The agent-facing interface should be generated from the same spec, schema, and access-policy source of truth already used to manage the API, with each agent-ready interface registered in a central catalog for reuse. _Teams building secure agent-to-API integrations can track governance patterns like this on daily.dev._ ## Similar posts on daily.dev - [Cloudflare Launches Code Mode MCP Server to Optimize Token Usage for AI Agents](https://daily.dev/posts/cloudflare-launches-code-mode-mcp-server-to-optimize-token-usage-for-ai-agents-zsfuoyvqc) · InfoQ · 2 upvotes · 0 comments - [Code Mode: give agents an entire API in 1,000 tokens](https://daily.dev/posts/code-mode-give-agents-an-entire-api-in-1-000-tokens-qvyaxboc7) · Cloudflare · 1 upvotes · 0 comments - [How Code Mode Builds on MCP for Agent Tooling](https://daily.dev/posts/how-code-mode-builds-on-mcp-for-agent-tooling-bjty4eiot) · Nordic APIs · 1 upvotes · 0 comments - [MCP on Code Mode with Matt Carey, Cloudflare Agents SDK and MCP \(Changelog Interviews \#681\)](https://daily.dev/posts/mcp-on-code-mode-with-matt-carey-cloudflare-agents-sdk-and-mcp-changelog-interviews-681--xbm64jsql) · Changelog · 0 upvotes · 0 comments - [How we cut AI costs by 80%](https://daily.dev/posts/how-we-cut-ai-costs-by-80--tfyfecyyr) · The New Stack · 1 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#architecture](https://daily.dev/tags/architecture), [#mcp](https://daily.dev/tags/mcp), [#api-gateway](https://daily.dev/tags/api-gateway) [View this post on daily.dev](https://daily.dev/posts/how-to-make-enterprise-apis-agent-ready-rmm8dqq35) ```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 to make enterprise APIs agent-ready","url":"https://daily.dev/posts/how-to-make-enterprise-apis-agent-ready-rmm8dqq35","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/how-to-make-enterprise-apis-agent-ready-rmm8dqq35"},"datePublished":"2026-10-26T09:04:55.784Z","dateModified":"2026-10-26T09:27:00.643Z","description":"Exposing enterprise APIs to AI agents as one MCP tool per operation breaks down once an API has hundreds of operations. Marco Palladino describes 'code mode'...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/7387310338a72be1c48514ae95e7d052?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/7387310338a72be1c48514ae95e7d052?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"InfoWorld","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":"InfoWorld","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/bf6d68a999064029b0bb09aa6268f1f3","url":"https://daily.dev/sources/infoworld"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/how-to-make-enterprise-apis-agent-ready-rmm8dqq35","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,architecture,mcp,api-gateway","timeRequired":"PT7M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"InfoWorld","item":"https://daily.dev/sources/infoworld"},{"@type":"ListItem","position":3,"name":"How to make enterprise APIs agent-ready"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/how-to-make-enterprise-apis-agent-ready-rmm8dqq35#faq","mainEntity":[{"@type":"Question","name":"why does exposing every API operation as a separate MCP tool break down for large enterprise APIs","acceptedAnswer":{"@type":"Answer","text":"It fails because loading hundreds of tool definitions into the model's context before any work happens increases token cost, degrades tool selection accuracy as the agent confuses similar-looking options, and adds latency from excessive back-and-forth between the model and tools. A typical enterprise CRM or ERP API can have hundreds of operations, overwhelming an agent's reasoning capacity. Platform teams weighing MCP integration patterns can track real-world approaches like this on daily.dev."}},{"@type":"Question","name":"what is code mode in the context of AI agents and MCP servers","acceptedAnswer":{"@type":"Answer","text":"Code mode is a pattern where a server exposes an API as a single SDK-like interface instead of a flat list of MCP tools, so the agent writes short code snippets against that interface that run in a sandbox, with only the final result returned to the model's context. This lets one snippet chain multiple API calls with loops and error handling, cutting round trips and token usage dramatically, for example from about 120,000 tokens to about 1,200 in one tested workflow. Developers evaluating agent tool-calling architectures can follow emerging patterns like code mode on daily.dev."}},{"@type":"Question","name":"how should agent traffic be governed when agents call production enterprise APIs","acceptedAnswer":{"@type":"Answer","text":"Agent calls should route through the same API gateway that governs every other client, so existing authentication, authorization, rate limiting, and audit logging policies apply automatically rather than being recreated per MCP server integration. The agent-facing interface should be generated from the same spec, schema, and access-policy source of truth already used to manage the API, with each agent-ready interface registered in a central catalog for reuse. Teams building secure agent-to-API integrations can track governance patterns like this on daily.dev."}}]} ```