WebAgent

WebAgent is Qoni's web action layer. It lets agents perform search, extraction, browser operations, tracking, and controlled web tasks on the open web. After reading these docs, you can integrate WebAgent through the current Node SDK or REST API and stream execution results.

WebAgent is not a traditional crawler SDK. It is designed for the LLM agent execution workflow: developers provide an instruction, while WebAgent manages runtime resources, page state, retries, structured results, and the run lifecycle. The Console and SDKs are clients for the same API.

An agent that lives in your browser

Tell WebAgent what you want done, and it turns the request into a sequence of executable browser actions: open a page, log in, search, fill a form, click, read the result, and return it to your business system.

In enterprise production, about 90% of computer work happens in a browser. WebAgent lets AI take over that work: people no longer have to click, copy, and switch between pages one step at a time. Give it the goal, and the agent in your browser can carry the task through.

WebAgent is more than a click bot: it turns the browser into an execution surface that business systems can call:

• Faster: reuse sessions, Profiles, and browser context to avoid repeated startup, login, and waiting.
• More resilient: manage page state, retries, timeouts, and human confirmation together, so failures are visible and recoverable.
• More controllable: isolate permissions and spend by project, then follow every action through SSE events.
• Easier to integrate: create a run from one instruction while Console, OpenAPI, and raw HTTP share one contract.

Traditional browser vs. WebAgent

Comparison area	Traditional browser	WebAgent
Primary role	Displays pages and waits for user input	Understands a goal and executes it in the browser
How work happens	The user clicks, types, and switches pages step by step	Turns a request into executable actions and keeps going
Who does the work	The user completes every step	The agent handles repetitive work; the user confirms key points
Task continuity	State often has to be reconstructed after a page or session ends	Sessions, Profiles, and runs keep the context
Results	Results stay on the page	Structured results and SSE events return to the business system

Use cases

WebAgent is for the work people do in a browser every day, but should not have to repeat by hand:

• Enterprise marketing: collect channel and customer feedback, turn it into insights, and draft or publish content automatically.
• Sales and customer success: log in to the CRM, summarize customer activity, update follow-ups, and trigger the next action.
• Operations and commerce: collect prices, inventory, and promotions across sites, update back-office systems, and flag anomalies.
• Research and intelligence: search public sources, cross-check evidence, and keep watch on competitors and industry changes.
• Internal workflows: query systems, fill forms, download files, reconcile records, and write results back to the business system.

Toward the agentic web

Qoni's goal is to make Agents first-class citizens of the Web. WebAgent is not only a background click simulator, and not only a renderer for humans; it places Human, Agent, and Web in the same auditable collaboration layer, so an Agent can read, act, produce results, and render state back to people within an explicit permission boundary.

When to use WebAgent

• Your agent needs real-time web data instead of relying only on model training data or a fixed knowledge base.
• You need search, extraction, browser actions, and long-running tasks behind auditable APIs.
• You want the Console, SDKs, and backend services to share the same REST API contract.
• You need a session / run model for state, event streaming, or steps that require user confirmation.

When not to use WebAgent

• The task only needs your own backend APIs and does not need open web access.
• You need large-scale offline crawling, warehouse synchronization, or search-index construction.
• The target site's terms do not allow automated access and you do not have the required authorization.
• You have not defined API keys, project scope, run budget, and failure handling.

Core capabilities

Capability	Description
DoAnything API	Provide a natural-language instruction. WebAgent chooses tools, runs steps, and returns a result.
Shaped APIs	Dedicated API contracts for artifact-shaped workflows such as DeepResearch, WebSearch, and Track.
Session / run model	A session owns runtime resources. A run represents one instruction or follow-up action.
Event stream	Subscribe to run state, output chunks, errors, and user-confirmation requests through SSE.
SDK and raw HTTP	Node.js / TypeScript and cURL docs use the same API semantics; Python uses raw HTTP.

Documentation entry points

• What is WebAgent explains WebAgent's role, boundaries, and API shape.
• Quickstart runs the first run with Node.js or cURL.
• Authentication & API keys explains wa_ keys, project scope, and rotation.
• DoAnything explains the session, run, event, and profile lifecycle.
• Errors & Retries covers error codes, retry policy, and idempotency.
• API Reference covers base URL, auth, errors, rate limits, and pagination.
• Vibecoding shows how to give the docs and OpenAPI spec to an IDE-resident LLM.