What's New & Next

Dashboard Performance Overhaul

The dashboard now loads significantly faster. We rewrote analytics queries to aggregate data at write time instead of computing it on every page load, consolidated redundant subscriptions, and added proper database indexes across the board.

Also fixed an issue where execution counters could drift out of sync, and resolved a bug that left some executions stuck in a "processing" state.

Webhooks, Reliability Insights & Advanced Filtering

Webhooks now show delivery status directly in the bots table—green when healthy, red when failing—so you can spot integration issues at a glance. Webhook configuration got smarter too: reuse existing configs across bots and pick your data path from a dropdown instead of typing it manually.

A new reliability column shows each bot's success rate over recent runs. The bot registry also gained a multi-dimension filter popover so you can slice by status, tags, and categories simultaneously.

Dashboard header stats now reflect real data instead of placeholder values.

Clipboard & Proxy Improvements in Browser Sessions

Browser sessions (Browse Now and Shared Sessions) now support full clipboard operations—paste, copy, select-all, and text selection all work as expected. The URL bar updates in real time as you navigate, and proxy configuration is now automatically derived from the bot's target URL.

Bot Categories & Scheduler Reliability

Bots are now automatically grouped by target domain into categories, making large bot libraries much easier to navigate. Combine categories with tags and status filters for precise organization.

Fixed an issue where scheduled runs could get stuck in a pending state. The scheduler now bases parallelism on actual runner capacity instead of a fixed limit, and long-running bots no longer get prematurely killed.

Sessions Hub, External Sessions & Webhooks

Sessions Hub—share live browser sessions with your team. Multiple people can view the same stealth browser in real time, perfect for collaborative debugging and demos.

External Session Access—when a bot hits a step that requires human judgment (2FA, CAPTCHA review, manual approval), it pauses and sends a notification. Respond from a unique URL without needing a hidettp account.

Bot Data Webhooks—configure webhooks to POST extracted data to your systems when executions complete. Includes full request/response debugging so you can troubleshoot delivery issues.

Also new: file download capture during bot runs, and org-level tags with color-coded filter chips on the dashboard.

Browse Now & Human-in-the-Loop

Browse Now—launch a stealth browser session directly from the dashboard. Browse any site with anti-detection enabled, proxy support, and full remote access. Useful for testing selectors, verifying bot behavior, or manual data collection.

Request Human Step—bots can now pause mid-execution and request human intervention. The assigned person receives a notification, takes over the browser session, completes the action, and the bot resumes automatically.

Selectors are now more resilient with multi-locator fallback strategies, and native HTML <select> dropdowns are properly supported.

Action Bots, Scheduling & Multi-Runner

Parameterized Action Bots—define parameters in your bot's manifest (dates, counts, text inputs) and pass different values for each execution. The AI generates the script logic to handle parameters automatically.

Bot Scheduling—set any bot to run on a recurring interval (minutes, hours, or days). The system handles queuing, capacity management, and retry logic.

Horizontal Scaling—deploy multiple runner instances and they automatically register, load-balance, and handle failover. A new infrastructure dashboard shows runner status and capacity in real time.

Also new: step replay for editing recorded scripts, and an API playground with bot selector and parameter validation.

Proxy Support, CAPTCHA Handling & Security

Added a proxy bridge for seamless authenticated proxy connections without exposing credentials to the target site. Bots can now detect and handle CAPTCHAs during execution, with support for third-party solver integrations.

New assessment system verifies that bot actions had the intended effect (e.g., confirming a purchase went through, not just that the button was clicked). Plus security hardening across the runner and API layers.

Step Pipeline, Data Extraction & REST API

Step Pipeline—record multi-step workflows and execute them as reproducible pipelines. Each step reports its status, timing, and output independently, so you can see exactly where things succeed or fail.

Visual Data Extraction—point and click on page elements to define what data to extract. An AI-powered formatter structures the output automatically, and a built-in data viewer shows results in a clean table.

REST API v1—full programmatic access to bots, executions, scripts, manifests, and schedules. Authenticated with API keys, cursor-based pagination, and standard JSON responses.

Visual Recorder & Live Browser View

Record bot automations by interacting with a real browser—clicks, inputs, navigation, and scrolling are captured as a Python script. Watch bots run in real time through a live browser view, and review video recordings of past executions for debugging.

Teams & Role-Based Access

Create organizations, invite team members, and assign roles. Four permission levels—Owner, Admin, Member, Viewer—control who can create bots, trigger executions, manage settings, and view results. All data is strictly scoped to the organization.

Initial Release

Bot dashboard with filtering and search, manual bot creation, execution tracking with detailed logs, AI chat panel for script generation, and OTP-based email authentication. Deployed to production.