# Pomerado Launches Platform to Turn Websites into Agent-Ready MCPs Pomerado converts legacy websites into deterministic APIs to improve AI agent reliability and speed. - **Source:** [Hacker News: Launches](https://news.ycombinator.com) - **Published:** 2026-11-11 - **Topic:** [technology](https://brifing.si/c/technology) - **Original article:** - **On Brifing:** https://brifing.si/s/a6321279-b149-410a-93f9-8ce39fd5f7bc ## Bridging the Agent Gap Pomerado addresses the critical bottleneck where AI agents struggle to interact with legacy websites lacking formal APIs. By transforming standard web interfaces into deterministic, self-healing Model Context Protocol (MCP) endpoints, the platform enables agents to perform tasks reliably. This replaces slow, error-prone browser automation methods currently used for complex web workflows. ## Founding Team Background Founders Alan and Akshay bring experience from McKinsey and Scale AI, where they specialized in deploying agents across healthcare, telecom, and professional services. Their professional history highlighted that the primary failure point in agent deployment was the inability to interface consistently with legacy portals, leading them to build this specific solution. ## Deterministic API Generation The platform allows developers to create APIs for any website in minutes through a self-serve interface. Users provide a natural language description and workflow details, resulting in an MCP in under an hour. Future tool calls utilize this deterministic structure, typically returning results within a five to thirty-second window for efficiency. ## Automated Workflow Recording During the initial setup, an agent explores the target website and performs the requested task once using HTTP requests or Playwright scripts. This example run establishes the necessary navigation and interaction patterns. A separate Guardian model reviews these calls against safety guardrails, while synthetic input tests ensure the generated API remains generalized. ## Handling Authentication Challenges Pomerado manages sites requiring authentication and two-factor authentication without compromising user credentials. The system employs a deterministic sign-in process, ensuring that models and recording tools never directly access sensitive login data. Users can store TOTP seeds or link accounts to a dedicated phone number for automated 2FA code handling. ## Self-Healing Maintenance Jobs If a tool fails during future runs, a maintenance job automatically initiates to repair or improve the integration. This process can fix broken steps, such as renamed buttons, or add new functionality like expanded search parameters. Users can even request performance optimizations through simple natural language prompts to improve execution speed. ## Versatile Industry Applications Early adopters include personal trainers automating client messaging and car dealerships connecting Grok bots to legacy CRM and ERP systems. These users have transitioned away from traditional computer use, which previously required over an hour for tasks. The platform is currently available via an open-source MIT license or a free hosted version. ## Future of Agent Interoperability By prioritizing deterministic runtime performance over the fragility of browser automation, Pomerado aims to scale AI agent utility in high-volume environments. The platform demonstrates that the future of agentic workflows relies on robust, machine-readable interfaces rather than human-centric UI interaction, significantly reducing the cost and time required for complex digital task automation. ## Sources covering this story - https://news.ycombinator.com/item?id=50455811 — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._