May 2026 Summaries

Merge's Cookie Policy and its integration platform are highlighted alongside its Quartr connector, which facilitates AI access to extensive finance-related data from over 15,000 listed companies. This connector allows AI systems to efficiently query a range of investor relations documents, such as earnings calls and financial reports, with features like AI-generated summaries and timestamped transcript navigation. The Quartr connector is integrated with Merge's Agent Handler to ensure secure access control, authentication, and governance, allowing users to either connect Quartr tools for internal workflows or embed them into AI products. This integration aims to streamline processes for asset managers, wealth management platforms, and investment banks, enabling them to synthesize and analyze critical financial information swiftly. The Quartr connector is available for free, offering a way for users to test its capabilities and explore its security features through Merge's platform.

Merge's platform provides solutions to streamline the use of employee agents, which are tools designed to automate tasks and workflows across existing systems used by teams. These agents require careful management of permissions to ensure they match the roles of employees, mitigating risks such as data exposure. The platform emphasizes the importance of implementing strong security measures, such as SCIM-based provisioning and identity-bound authentication, to maintain compliance and safeguard sensitive information. Merge helps organizations integrate these agents efficiently with existing applications like Salesforce, Slack, and Expensify, offering features such as user-level audit logs and centralized policy enforcement to ensure secure and effective operation. By using the Merge Agent Handler, companies can avoid the complexities and potential security pitfalls of building these integrations in-house, allowing IT and security teams to maintain visibility and control over agent activities.

Merge's integration platform facilitates the connection between Codex, an AI tool, and Jira, a project management system, allowing Codex to access real-time ticket data for more accurate task execution. By using the Merge Agent Handler, developers can link Codex directly to Jira through the Merge CLI, providing Codex with comprehensive ticket information rather than relying on potentially incomplete developer summaries. This setup enhances the accuracy of Codex's output by allowing it to read full ticket descriptions, acceptance criteria, and related issues directly from Jira, thus ensuring that the generated code aligns closely with the actual project requirements. Additionally, the platform offers features such as credential storage, token refresh, and user-level audit logging, ensuring secure and efficient management of Codex's access to Jira, which is particularly beneficial for engineering teams working with shared projects.

Merge has introduced "Build Your Own Router" in its Gateway platform, providing customizable control over AI model routing for teams utilizing large language models (LLMs). This feature allows users to define routing logic based on specific benchmarks and priorities, such as cost, latency, and accuracy, enabling personalized model selection without manual routing code. Merge maintains a library of standard, industry, and custom benchmarks to aid in this process, offering presets for ease of use. The platform aims to enhance transparency, auditability, and security in model selection, making it especially useful for regulated industries. Users can create configurations via the Gateway dashboard or API, with the option to request additional benchmarks. For a limited time, Merge offers a promotion with zero fees, encouraging early adoption of this flexible routing infrastructure.

The text discusses the concept of multi-model routing in the context of large language models (LLMs), emphasizing its importance in selecting the best model for specific tasks based on performance criteria such as quality, speed, and cost. It highlights several strategies for implementing multi-model routing, including ranking preferred models, minimizing time to first token, and assigning weights to benchmarks. The text also examines tools like OpenRouter, LiteLLM, TrueFoundry, and Merge Gateway, each with unique features and challenges, such as ease of integration, control over routing, and security considerations. Merge Gateway stands out as a unified control plane offering centralized multi-model access, routing, cost governance, and security, allowing users to define custom routing policies that align with their specific needs.

Merge has established itself as a critical infrastructure provider for AI-driven businesses, initially through its Unified API, which offers seamless integration across the SaaS ecosystem by simplifying access to hundreds of software integrations. As the AI market evolved, Merge expanded its offerings to address emerging challenges in AI production, such as the need for synchronized business context, flexible live access to software systems, and a robust AI gateway and control plane. The company's suite of products—including Unified, Agent Handler, and Gateway—addresses these needs by ensuring reliable data integrations, secure interactions with third-party applications, and comprehensive model management for scalable AI deployment. This strategic positioning has made Merge a foundational element for leading AI companies, enabling them to focus on innovation while leveraging Merge's infrastructure for data access, action execution, and model control.

Merge leverages the Notion connector from Agent Handler to integrate institutional knowledge into its operational workflows, enhancing efficiency without disrupting focus. By using this connector, team members can access relevant information from Notion directly within existing applications, such as Slack, to make informed decisions promptly. The connector supports various operations, including searching pages and databases and updating content, allowing employees to retrieve policy-aware answers and runbook context efficiently during active incidents or tasks. It also facilitates the automatic updating of Notion pages as a byproduct of everyday work, ensuring that the knowledge base remains current. This integration is designed to streamline processes, reduce time spent on manual searches, and ensure that team members can access up-to-date information swiftly, thus improving overall productivity and decision-making accuracy.

Merge's integration platform offers a method to connect Codex directly to Google Sheets, enhancing the accuracy of code generated by Codex by accessing real-time data from spreadsheets rather than relying on developers' interpretations. By using the Merge Agent Handler, developers can link Codex to Google Sheets through a straightforward setup involving the Merge CLI, which handles OAuth token storage and refresh, eliminating the need for manual configuration and enhancing security. This connection allows Codex to read live data and field definitions directly from spreadsheets, enabling it to generate code that accurately reflects the source data, which is particularly beneficial for tasks involving field mappings, feature flags, and test datasets. Additionally, the Merge Agent Handler offers audit logging and access control, allowing IT teams to manage and review access to spreadsheets, thus ensuring secure and governed usage of AI tools like Codex within organizations.

Merge's platform emphasizes the use of SCIM (System for Cross-domain Identity Management) to streamline AI integration within organizations by managing employee access to AI tools and systems. SCIM's functionality extends beyond traditional applications, allowing it to determine not only the access levels employees receive but also which AI applications they can use. This involves leveraging an identity and access management (IAM) platform like Okta, along with an agentic integration solution such as Merge Agent Handler, to handle provisioning and deprovisioning. SCIM offers significant advantages, including automated offboarding, adherence to the principle of least privilege, improved auditability, and expedited enterprise AI rollout. The platform showcases various departmental use cases, including sales, engineering, customer support, and finance, highlighting how SCIM can enhance workflow automation and security while aligning with existing identity management processes.

Merge employs its Ahrefs connector to efficiently manage SEO and AEO tasks by automating keyword research, competitive analysis, and content refresh strategies. The connector, part of Merge's integration platform, facilitates tasks typically performed in Ahrefs by offering over 30 tools, which help prioritize impactful keywords, track competitor content, and identify content refresh opportunities. The integration with Claude Code simplifies the process by using remote agents to analyze keyword gaps and monitor content performance, allowing Merge to enhance visibility and traffic. Setting up the Ahrefs connector is straightforward, involving the installation of the Merge CLI and authentication through Claude Code, making it accessible for users to leverage Ahrefs data within their workflow.

Merge's platform facilitates the integration of Google Slides with Codex, allowing developers to directly access presentation data for more accurate and contextually relevant coding tasks. This integration enables Codex to pull the original architecture decks, technical specs, and system design slides, ensuring that generated code closely aligns with the original designs and specifications rather than a developer's interpretation. By using Merge Agent Handler, organizations can manage access controls, audit logging, and security protocols, ensuring that sensitive data within Google Slides is securely handled. This setup mitigates the potential for information loss during manual transcription from slides to task descriptions, thereby bridging the gap between designed intent and implemented code. The platform is designed to streamline operations at scale, allowing for centralized management of credentials and task permissions, which is particularly beneficial for team-based environments where architecture decisions are critical.

Merge offers an integration platform that simplifies connecting various tools like Google Drive to development environments through its Merge Agent Handler and CLI, easing the process for developers who need to access and manipulate data without leaving their coding interface. By connecting Google Drive to Cursor, developers can query Drive files and folders directly within the editor, eliminating the need for disruptive context switches to a browser for information like folder paths, file IDs, and API response shapes. This setup involves installing the Merge CLI, authenticating with Merge and Google Drive, and configuring Cursor to use Merge tools, which streamlines the integration process while providing enhanced security and logging features. Self-hosting alternatives are available but come with complexities like managing OAuth credentials and ensuring secure access, whereas Merge provides a centralized, auditable solution that allows for granular control over Drive operations. This approach is particularly beneficial for teams, as it reduces overhead and ensures consistent access management across multiple developers.

Merge's blog post explores the differences between two AI integration platforms, Nango and Composio, as well as their respective advantages and drawbacks. Nango is recognized for its unified API platform that supports both product and agentic integrations, offering free access to MCP connectors despite being a smaller startup with limited documentation and customer proof. Composio, on the other hand, provides a broader connector catalog and has raised more funding, but it recently faced a significant security incident and offers limited product support. The article suggests that Merge Agent Handler is a superior alternative, providing managed MCP connectors with advanced governance, observability, and authentication features, which allow AI agents to interact securely with third-party systems. Leading AI companies prefer Merge for its robust enterprise-grade solutions, which help streamline integration processes while ensuring security and compliance.

Merge offers a streamlined solution for integrating BambooHR with its Merge Agent Handler, allowing developers to efficiently manage HR data and automate workflows without the need for one-off integrations. By leveraging the Merge CLI and a single authentication process, users can connect Claude Code to the BambooHR API, enabling access to employee records, PTO balances, and organizational data directly from the terminal. This integration not only simplifies API key management but also offers data loss prevention controls to safeguard sensitive information. The Merge platform supports a wide range of use cases by providing direct access to live HR data, reducing the need for exporting CSVs or building custom integrations for each new workflow requirement. Through its 150+ connectors, Merge enhances the ability of developers to create secure, reliable, and robust agents that can efficiently handle HR data tasks, thus minimizing engineering overhead and ensuring data privacy.

Merge provides a platform that facilitates the integration of tools like Sentry with development environments such as Cursor, streamlining the process of building and managing customer integrations. By using Merge Agent Handler and its CLI, developers can connect to the Sentry API directly from their project environments, eliminating the need to leave Cursor for data queries and authentication tasks. This setup allows developers to efficiently inspect Sentry's data model, fetch project-specific data, and manage authentication centrally, ensuring secure and controlled access. Merge's approach addresses the challenges of scaling API integrations across multiple developers or agents by offering centralized authentication and detailed logging of API interactions, which provides both security and an audit trail for actions taken by different agents.

Merge's platform offers a streamlined approach to connecting and managing integrations with tools like HubSpot through its Merge Agent Handler, which connects Cursor to HubSpot's CRM API. By installing the Merge CLI and authenticating through the platform, developers can seamlessly query HubSpot data without leaving Cursor, facilitating the inspection and manipulation of real objects to ensure code accuracy. The centralized management of HubSpot OAuth tokens by Merge enhances security and ease of use, eliminating the need for local credential storage. This setup not only aids individual developers but also supports teams by providing access control, audit logging, and policy enforcement, all of which are critical for managing sensitive data effectively. Merge's approach allows developers to integrate AI capabilities into systems like HubSpot while maintaining centralized control over access and operations, ensuring functional and defensible production deployments.

Merge is an integration platform that facilitates secure and efficient connections between products and AI agents, like Claude Code, with services such as Expensify. By using the Merge Agent Handler, developers can streamline expense workflows for finance and engineering teams by directly interacting with the Expensify API without managing partner credentials. This setup allows for the automation of tasks such as exporting reports, creating expenses, and managing spend policies, all from a terminal interface. Merge provides a controlled environment that logs tool usage and restricts access to only necessary functions, which is particularly advantageous over self-hosted solutions that require extensive credential management and lack specific access controls. With the integration of Expensify MCP, Claude Code can easily handle expense data, ensuring compliance and efficient data retrieval for further organizational use, such as policy violation identification or cost-center summaries.

Merge's platform offers a streamlined method for developers to connect WHOOP's API to Cursor, facilitating access to physiological data without leaving the development environment. By utilizing the Merge Agent Handler, developers can manage OAuth credentials and token refreshes centrally, eliminating the need for individual authorization flows for each user, which can become cumbersome as applications scale. This integration allows developers to directly query WHOOP's API data within Cursor, enhancing accuracy and efficiency in code development by providing real-time data rather than relying on potentially outdated documentation. The Merge platform supports a broader range of integrations beyond WHOOP, helping developers manage customer integrations with a robust set of connectors and ensuring secure and efficient data handling.

Merge's platform facilitates the integration of AI agents with third-party tools through Model Context Protocol (MCP) gateways, which streamline the connection process by providing a single endpoint for multiple MCP servers. This approach addresses security risks, such as credential leaks, and reduces the complexity of direct MCP server connections, which many developers find resource-intensive and time-consuming. MCP gateways like Merge Agent Handler offer centralized governance, authentication, and credential management, enhancing security and reducing engineering overhead. They also provide observability and troubleshooting features, allowing for comprehensive logging and auditing of tool interactions. Platforms like Arcade.dev and Composio offer alternatives with varying strengths, such as developer-centric features and extensive connector catalogs, although they may present challenges like pricing complexity and limited emphasis on governance.