Browser Agent
Product · Developer tool by telli
Automate tasks in systems without APIs through direct browser interaction.
A specialized web automation tool for telli AI agents to interact with browser-based software that lacks APIs. It enables businesses to automate tasks in legacy or custom web systems as part of their broader conversational workflows.
Category: Workflow Automation
- Platforms
- Web-based software
- Integrations
- CRMs, calendars, custom APIs, and browser-based software
- Deployment
- Cloud
- Target Systems
- Browser-based software without APIs
- Integration Type
- Direct browser interaction
- Platform Foundation
- Model-agnostic AI infrastructure
- Completes tasks in browser-based software without APIs
- Connects legacy web systems to automated workflows
- Verifies sequences of actions through simulations
- Integrates with CRMs, calendars, and custom APIs
Overview
telli Browser Agent is a specialized automation tool within the telli platform designed for businesses that need to automate tasks in web-based systems lacking traditional APIs. It allows AI agents to interact directly with browser-based software to look up information, update records, or trigger processes that would otherwise require manual intervention. By bridging the gap between conversational AI and legacy web applications, it ensures that end-to-end workflows can be completed across a company's entire software stack. The tool is managed through a central platform that provides context, business knowledge, and process rules. It also supports rigorous testing through simulations, allowing developers to verify the sequence of actions an agent takes before deployment.
Web Automation Capabilities
- No-API Automation: The Browser Agent completes tasks directly in browser-based software, allowing automation to reach systems that do not have a native API.
- Legacy System Connectivity: It enables AI agents to interact with custom or legacy web applications to check statuses, submit requests, or update records.
- Workflow Extension: The agent works alongside native integrations for CRMs, calendars, and other business tools to provide a unified workflow.
- Process Integration: Developers can connect more of their existing processes to the telli platform by using the browser as an interface for automation.
Simulation and Testing
- Scenario Testing: The platform allows for testing common requests, difficult questions, and exceptions that matter to the business.
- Task Verification: Users can verify expected outcomes, appointment bookings, and the specific sequence of actions an agent takes in web systems.
- Regression Testing: Scenarios can be re-run against updated agent versions to review passes or fails before publishing changes.
Insights and Analytics
- KPI Tracking: The platform tracks bookings, qualified enquiries, and success criteria defined for each specific use case.
- Continuous Improvement: Findings can be turned into improvements by investigating recurring issues and preparing the next version.
Use cases
- Developers automate record updates in legacy web-based CRMs that do not offer a public API for integration.
- Support teams use the agent to check order statuses in custom vendor portals during a customer conversation.
- Operations managers trigger automated follow-up actions in web tools when a customer call ends or details change.
Specifications
- Platforms
- Web-based software
- Integrations
- CRMs, calendars, custom APIs, and browser-based software
- Deployment
- Cloud
- Target Systems
- Browser-based software without APIs
- Integration Type
- Direct browser interaction
- Platform Foundation
- Model-agnostic AI infrastructure
- Testing Support
- Scenario-based simulations
- Analytics
- Tool activity and success criteria tracking
How it works
- Identify Target Systems: Identify the browser-based software or legacy system that lacks an API where tasks need to be automated.
- Configure Agent Actions: Configure the Browser Agent within the telli platform to navigate the web interface and perform specific actions like updating records.
- Test and Verify: Use the platform's simulation tools to test the agent's actions against real-world scenarios and verify outcomes.
- Deploy to Workflows: Deploy the agent into live workflows where it can automatically trigger based on call ends or customer detail changes.
FAQ
What is the purpose of the telli Browser Agent?
The Browser Agent is designed to complete tasks directly in browser-based software that lacks a traditional API, ensuring end-to-end automation across all systems.
What kind of tasks can the Browser Agent perform?
It can be used to check a status, submit a request, or update a record in systems where native integrations or custom APIs are not available.
Can I test the actions the Browser Agent takes?
Yes, telli allows you to test the sequence of actions an agent takes in your systems, including those performed via the Browser Agent, to verify expected outcomes.
Does the Browser Agent work with different AI models?
The Browser Agent is part of the telli platform, which provides a model-agnostic foundation, meaning it can work as you evaluate and change supported AI models.
Added by AntSpace from public sources (telli.com).