@adminide-stack/yantra-help-browser 12.0.16-alpha.6 → 12.0.16

package/lib/templates/content/ai-employees/ai-employees-overview.md

@@ -0,0 +1,97 @@
+---
+title: 'AI Employees Overview'
+---
+
+# AI Employees Overview
+
+Yantra AI employees are intelligent agents that integrate into your team's existing toolkit and operate as autonomous, always-available team members. They go beyond traditional chatbots by working proactively — monitoring, analyzing, and acting on information across your connected services.
+
+## What Makes an AI Employee Different
+
+### Traditional AI Chatbot
+
+- Waits for you to ask a question
+- Has no awareness of your business context
+- Cannot take actions in your tools
+- Forgets conversations between sessions
+- Operates in isolation from your workflow
+
+### Yantra AI Employee
+
+- Works proactively without waiting for prompts
+- Monitors your tools 24/7 and understands your business context
+- Updates CRMs, creates tickets, drafts messages, and logs notes
+- Maintains a persistent memory of every interaction and decision
+- Operates within your existing workflow seamlessly
+
+## Core Capabilities
+
+### 1. Observing
+
+Your AI employee watches everything happening across your connected tools:
+
+- Slack conversations and channel activity
+- GitHub commits, pull requests, and deployments
+- Error monitoring alerts and exception trends
+- CRM deal movements and customer interactions
+- Email threads and calendar events
+
+### 2. Understanding
+
+Raw events are processed into meaningful insights:
+
+- Correlating a customer complaint in Slack with a Sentry error spike
+- Linking a stale deal in HubSpot with a missed follow-up email
+- Connecting a deployment failure with related code changes
+- Identifying patterns across seemingly unrelated events
+
+### 3. Acting
+
+Based on observations and understanding, your AI employee takes action:
+
+- Drafts response messages with full context
+- Updates CRM fields and deal stages automatically
+- Creates project tickets with detailed descriptions
+- Sends alerts and escalations to the right people
+- Prepares meeting briefs and daily summaries
+
+### 4. Learning
+
+Over time, your AI employee gets better:
+
+- Adapts to your team's communication style and preferences
+- Refines alert thresholds based on your responses
+- Builds deeper relationships between data points in your organization
+- Improves action suggestions through feedback loops
+
+## Types of AI Employees
+
+Yantra supports specialized AI employees configured for different functions:
+
+| Type               | Primary Function                                     | Key Integrations                     |
+| ------------------ | ---------------------------------------------------- | ------------------------------------ |
+| **Engineering AI** | Bug tracking, deployment monitoring, context sharing | GitHub, Sentry, Linear, Slack        |
+| **Sales AI**       | CRM automation, lead management, call preparation    | HubSpot, Salesforce, Gmail, Calendly |
+| **Marketing AI**   | Content management, social monitoring, engagement    | LinkedIn, X/Twitter, Notion, HubSpot |
+| **Support AI**     | Ticket routing, customer context, issue resolution   | Slack, Zendesk, Sentry, CRM          |
+
+## Trust and Control
+
+Yantra is designed with a **gradual trust model**:
+
+1. **Supervised Mode**: Your AI employee drafts all actions but waits for your approval before executing
+2. **Semi-Autonomous Mode**: Routine actions execute automatically; novel or high-impact actions require approval
+3. **Autonomous Mode**: Your AI employee operates independently, escalating only exceptions
+
+You can adjust the autonomy level at any time, for any action type.
+
+## Powered by OpenClaw
+
+Every Yantra AI employee runs on the [OpenClaw](https://github.com/openclaw) open-source agent framework. This provides:
+
+- **Transparency**: The agent logic is open source and auditable
+- **Community support**: Backed by a global community of developers and enterprise users
+- **Extensibility**: Custom workflows and integrations can be built on the OpenClaw framework
+- **Security**: Battle-tested by thousands of deployments worldwide
+
+Learn more about the OpenClaw foundation in the [OpenClaw Overview](/help-center/openclaw/openclaw-overview).

package/lib/templates/content/ai-employees/getting-started-ai-employees.md

@@ -0,0 +1,119 @@
+---
+title: 'Getting Started with AI Employees'
+---
+
+# Getting Started with AI Employees
+
+Setting up your first Yantra AI employee takes approximately 5 minutes. This guide walks you through the complete setup process from tool connection to your first proactive insight.
+
+## Prerequisites
+
+- A Yantra account (free trial available)
+- Admin or owner access to the tools you want to connect
+- A communication channel where you'd like to receive updates (Slack, WhatsApp, Telegram, or email)
+
+## Step 1: Connect Your Tools (2 minutes)
+
+### One-Click OAuth Setup
+
+From your Yantra dashboard, navigate to **Integrations** and connect your tools:
+
+1. Click the tool you want to connect (e.g., Slack, GitHub, HubSpot)
+2. Authorize access through the OAuth flow — same as adding any other app
+3. Select the specific resources to monitor (channels, repos, projects)
+4. Repeat for each tool your team uses
+
+### Supported Tools
+
+- **Communication**: Slack, Microsoft Teams, Gmail
+- **Development**: GitHub, GitLab, Bitbucket
+- **Error Tracking**: Sentry, Datadog, PagerDuty
+- **Project Management**: Linear, Jira, Asana
+- **CRM**: HubSpot, Salesforce
+- **Documentation**: Notion, Confluence
+- **Social**: LinkedIn, X/Twitter
+- **Calendar**: Google Calendar, Calendly
+- **Video**: Zoom (for call transcription)
+
+No coding or command-line setup is required. Your AI employee receives the same level of access as a regular team member.
+
+## Step 2: Configure Preferences (3 minutes)
+
+### Define What to Watch
+
+Tell your AI employee what matters most:
+
+- **Error monitoring**: Set severity thresholds (e.g., alert on 10+ occurrences per hour)
+- **Customer tracking**: Define high-priority accounts or customer tiers
+- **Deal tracking**: Specify pipeline stages that need attention
+- **Channel monitoring**: Select which Slack channels to watch actively
+
+### Set Notification Preferences
+
+- **Primary channel**: Where you want to receive most communications
+- **Urgent channel**: Where critical alerts get delivered (e.g., SMS, WhatsApp)
+- **Quiet hours**: Time windows where non-critical notifications are batched
+- **Daily briefing time**: When you'd like your morning summary
+
+### Choose Your AI Employee Type
+
+Select the focus area for your AI employee:
+
+- **Engineering**: Optimized for development workflows, error correlation, and deployment tracking
+- **Sales**: Optimized for CRM management, call preparation, and lead follow-up
+- **Marketing**: Optimized for content management, social media, and brand monitoring
+- **General**: Balanced across all areas for smaller teams
+
+## Step 3: Start Supervised (Ongoing)
+
+Your AI employee begins in **supervised mode** by default:
+
+1. **Observation begins immediately**: Your AI employee starts processing events from connected tools
+2. **First briefing**: Within hours, you'll receive your first contextualized summary
+3. **Action drafts**: When the AI employee identifies something actionable, it drafts the response and asks for approval
+4. **Review and approve**: Check the draft, edit if needed, and approve to execute
+
+### Example First Interactions
+
+- A morning briefing summarizing overnight activity across your tools
+- A contextual alert linking a customer complaint to a recent error spike
+- A drafted CRM update based on an email exchange it detected
+
+## Step 4: Expand Autonomy (When Ready)
+
+As you build trust with your AI employee:
+
+1. **Review the action log**: See all past actions, their outcomes, and your approval history
+2. **Identify routine actions**: Find action types you consistently approve without modification
+3. **Enable auto-execution**: For those routine actions, switch from "approval required" to "execute automatically"
+4. **Monitor and adjust**: Continue reviewing the daily summary to ensure quality
+
+### Autonomy Levels
+
+| Level               | Description                           | Best For                       |
+| ------------------- | ------------------------------------- | ------------------------------ |
+| **Supervised**      | All actions require approval          | Initial setup, first 1-2 weeks |
+| **Semi-Autonomous** | Routine actions auto-execute          | Established trust, weeks 2-4   |
+| **Autonomous**      | Independent operation with escalation | Mature setup, month 2+         |
+
+## Troubleshooting Setup
+
+### Tool Connection Issues
+
+- Ensure you have admin access to the tool you're connecting
+- Check that your organization allows third-party OAuth integrations
+- Verify your firewall isn't blocking Yantra's API endpoints
+
+### No Events Appearing
+
+- Confirm the correct channels, repos, or projects were selected during setup
+- Check that the connected tool has recent activity
+- Allow up to 5 minutes for initial event ingestion
+
+### Notification Not Received
+
+- Verify your notification channel preference is correctly set
+- Check Slack permissions if using Slack as your primary channel
+- Ensure notifications aren't being filtered by your device settings
+
+For additional help, contact support or visit the [Troubleshooting Guide](/help-center/troubleshooting/common-issues).

package/lib/templates/content/ai-employees/how-ai-employees-work.md

@@ -0,0 +1,123 @@
+---
+title: 'How AI Employees Work'
+---
+
+# How AI Employees Work
+
+This guide explains the technical and operational mechanics behind Yantra AI employees — how they observe, reason, and act within your organization's toolchain.
+
+## Architecture Overview
+
+Each Yantra AI employee runs in a **dedicated, isolated environment** provisioned specifically for your organization. This ensures:
+
+- Your data is completely separated from other customers
+- Processing resources are dedicated to your workload
+- Custom configurations don't interfere with other environments
+
+### Event Pipeline
+
+```
+Your Tools → Yantra Gateway → Event Processor → Context Engine → Action Engine → Your Tools
+```
+
+1. **Yantra Gateway**: Secure API routing layer that receives events from all connected tools via OAuth 2.0 and API keys
+2. **Event Processor**: Normalizes and enriches raw events from different tool formats into a unified event stream
+3. **Context Engine**: Correlates events with historical data, customer profiles, and organizational knowledge
+4. **Action Engine**: Determines appropriate responses, drafts actions, and either executes or queues for approval
+
+## The Observation Layer
+
+### Event Collection
+
+Your AI employee receives real-time events through:
+
+- **Webhooks**: Instant notifications from tools like GitHub, Slack, and Sentry
+- **Polling**: Periodic checks for tools that don't support webhooks
+- **API subscriptions**: Real-time feeds from CRMs and project management tools
+
+### Event Types Processed
+
+- Messages and conversations
+- Code changes and deployments
+- Error logs and performance metrics
+- Customer interactions and deal updates
+- Calendar events and meeting recordings
+- Document changes and wiki updates
+
+## The Reasoning Layer
+
+### Context Graph
+
+Yantra maintains a **context graph** — a dynamic knowledge structure that maps relationships between:
+
+- **People**: Team members, customers, contacts
+- **Projects**: Repositories, deals, campaigns, tickets
+- **Events**: Incidents, deployments, meetings, decisions
+- **Artifacts**: Documents, code, configurations, policies
+
+When a new event arrives, the reasoning layer traverses this graph to understand:
+
+- Who is involved?
+- What systems or projects are related?
+- Has this happened before? How was it resolved?
+- Who should know about this?
+
+### Pattern Recognition
+
+Over time, the AI employee identifies patterns:
+
+- Recurring errors after specific types of deployments
+- Customer issues that correlate with feature releases
+- Sales cycles that follow predictable timelines
+- Marketing content that generates the most engagement
+
+## The Action Layer
+
+### Action Types
+
+| Action       | Description                   | Example                                    |
+| ------------ | ----------------------------- | ------------------------------------------ |
+| **Inform**   | Share a contextual update     | Morning briefing with overnight summary    |
+| **Alert**    | Notify someone about an issue | Critical Sentry error escalation           |
+| **Update**   | Modify a record in a tool     | Update HubSpot deal stage after a call     |
+| **Create**   | Generate a new item           | Create a Linear ticket for a reported bug  |
+| **Draft**    | Prepare content for review    | Draft a follow-up email after a sales call |
+| **Escalate** | Route to a human for decision | Novel issue requiring human judgment       |
+
+### Execution Modes
+
+**Supervised Mode**
+
+- All actions are drafted and presented for approval
+- You can edit, approve, or reject each action
+- Ideal for initial setup and building trust
+
+**Semi-Autonomous Mode**
+
+- Predefined routine actions execute automatically
+- Novel or high-impact actions require approval
+- Good for teams that want automation with oversight
+
+**Autonomous Mode**
+
+- All actions execute independently
+- Exceptions and anomalies are escalated
+- Best for mature setups with well-tuned configurations
+
+## Communication Channels
+
+Your AI employee communicates through the channels your team already uses:
+
+- **Slack**: Primary channel for most teams; supports threads, mentions, and rich messages
+- **WhatsApp**: Mobile-friendly for on-the-go updates and alerts
+- **Telegram**: Lightweight messaging for quick interactions
+- **Email**: Formal communications and detailed reports
+
+## Data Flow and Privacy
+
+- All data is encrypted in transit (TLS 1.2+) and at rest (AES-256)
+- Your AI employee only accesses tools and scopes you've explicitly authorized
+- Data from third-party APIs is never used to train AI models
+- Complete audit trail of every observation, decision, and action
+
+For detailed security information, see [Security Overview](/help-center/security-compliance/security-overview).

package/lib/templates/content/ai-employees/supervised-mode.md

@@ -0,0 +1,141 @@
+---
+title: 'AI Employee Supervised Mode'
+---
+
+# Supervised and Autonomous Modes
+
+Yantra AI employees support a spectrum of operational modes, from fully supervised to fully autonomous. This guide explains each mode, when to use it, and how to transition between them.
+
+## The Trust Spectrum
+
+Yantra's design philosophy centers on **progressive trust**. You start with full oversight and gradually delegate more responsibility as your AI employee proves its reliability.
+
+```
+Supervised → Semi-Autonomous → Autonomous
+(You approve everything)  (Routine auto-executes)  (AI works independently)
+```
+
+## Supervised Mode
+
+### How It Works
+
+In supervised mode, your AI employee:
+
+- **Observes** all events across connected tools freely
+- **Analyzes** and correlates data without restriction
+- **Drafts** all proposed actions (messages, CRM updates, ticket creation, etc.)
+- **Waits** for your explicit approval before executing any action
+
+### When to Use
+
+- During initial setup (first 1-2 weeks)
+- When connecting new tools or expanding scope
+- For high-stakes actions regardless of overall trust level
+- When training the AI employee on new processes
+
+### Approval Workflow
+
+1. AI employee identifies an actionable situation
+2. A draft action appears in your notification channel
+3. You review the draft with full context
+4. Options: **Approve**, **Edit and Approve**, **Reject**, or **Reject with Feedback**
+5. The AI employee learns from every decision
+
+### Example
+
+```
+[Yantra AI Employee - Draft Action]
+
+Situation: Marcus closed deal with TechStart via email 2 hours ago.
+Proposed Action: Update HubSpot deal stage from "Negotiation" to "Closed Won"
+Additional: Log email exchange as activity, set revenue to $48,000/yr
+
+[Approve] [Edit] [Reject]
+```
+
+## Semi-Autonomous Mode
+
+### How It Works
+
+In semi-autonomous mode, you define **action categories** and their trust levels:
+
+| Category                | Trust Level       | Behavior                           |
+| ----------------------- | ----------------- | ---------------------------------- |
+| CRM field updates       | Auto-execute      | Updates happen automatically       |
+| Ticket creation         | Auto-execute      | Tickets created without approval   |
+| Email drafts            | Requires approval | Drafts presented for review        |
+| Customer communications | Requires approval | Always human-approved              |
+| Alert escalation        | Auto-execute      | Alerts sent immediately            |
+| Financial actions       | Requires approval | Budget/pricing changes need review |
+
+### When to Use
+
+- After 1-2 weeks of supervised operation
+- When you've identified actions you consistently approve unchanged
+- For teams that want operational speed with selective oversight
+
+### Configuring Semi-Autonomous Rules
+
+1. Navigate to **AI Employee Settings → Autonomy Rules**
+2. Review the action log for your most common action types
+3. Set trust levels for each action category
+4. Define exceptions (e.g., "auto-update CRM except for Enterprise deals")
+
+## Autonomous Mode
+
+### How It Works
+
+In autonomous mode, your AI employee operates independently:
+
+- Executes all routine actions without approval
+- Handles standard workflows end-to-end
+- Escalates only **anomalies** and **genuinely novel situations** for human review
+- Provides a daily summary of all actions taken
+
+### When to Use
+
+- After 4+ weeks of reliable semi-autonomous operation
+- For well-defined, repetitive workflows
+- When the AI employee has sufficient training data and feedback
+
+### Safeguards
+
+Even in autonomous mode, important safeguards remain:
+
+- **Action limits**: Maximum number of actions per hour before requiring human confirmation
+- **Scope boundaries**: AI employee only acts within explicitly granted permissions
+- **Anomaly detection**: Unusual patterns trigger automatic escalation
+- **Kill switch**: Instantly revert to supervised mode at any time
+- **Audit trail**: Every autonomous action is logged with full context
+
+## Transitioning Between Modes
+
+### Moving to More Autonomy
+
+1. Review your **approval statistics** — look for action types with 95%+ unmodified approval rates
+2. Promote those action types to auto-execute
+3. Monitor the **action quality dashboard** for the first week
+4. Repeat for additional action types as confidence grows
+
+### Reverting to Less Autonomy
+
+- Click **Pause Autonomy** in the AI employee dashboard to immediately switch to supervised mode
+- Individual action categories can be demoted independently
+- No data or configuration is lost during transitions
+
+## Best Practices
+
+1. **Start narrow**: Begin with a small set of connected tools and expand gradually
+2. **Provide feedback**: When rejecting or editing actions, include reasons — this helps the AI improve
+3. **Review weekly**: Check the action summary to ensure quality matches expectations
+4. **Trust but verify**: Even in autonomous mode, regularly review the audit trail
+5. **Set clear boundaries**: Define which actions should never be automated (e.g., customer-facing communications for sensitive accounts)
+
+## Monitoring AI Employee Performance
+
+Track these metrics to guide your trust decisions:
+
+- **Action accuracy**: Percentage of actions approved without modification
+- **False positive rate**: Alerts that were dismissed as irrelevant
+- **Time saved**: Estimated hours saved per week through automation
+- **Escalation quality**: Whether escalated items genuinely required human judgment

package/lib/templates/content/blog/how-yantra-browser-automates-workflows.md

@@ -0,0 +1,120 @@
+---
+title: 'How Yantra Browser Automates Your Web Workflows'
+description: 'A deep dive into the technical architecture behind Yantra Browser — how CDP relay, service workers, and cloud AI work together to automate the web.'
+author: 'Yantra Team'
+date: '2025-03-21'
+category: 'Engineering'
+tags: ['browser-extension', 'cdp', 'architecture', 'engineering', 'automation']
+coverImage: '/static/img/blog/yantra-browser-architecture.png'
+readTime: '8 min read'
+slug: 'how-yantra-browser-automates-workflows'
+---
+
+# How Yantra Browser Automates Your Web Workflows
+
+Behind every seamless automation in Yantra Browser lies a carefully designed architecture that bridges your local Chrome browser with cloud-based AI intelligence. In this post, we take you behind the scenes to explain how it all works.
+
+## The Relay Architecture
+
+Yantra Browser uses a **relay architecture** — a pattern where the Chrome extension acts as a secure bridge between your browser tabs and the Yantra cloud platform.
+
+```
+┌─────────────────┐       ┌──────────────┐       ┌─────────────────┐
+│  Your Browser   │◄─────►│   Relay      │◄─────►│  Yantra Cloud   │
+│  (Chrome + Ext) │  CDP  │   (Service   │  WSS  │  (AI Assistant)  │
+│                 │       │    Worker)   │       │                  │
+└─────────────────┘       └──────────────┘       └─────────────────┘
+```
+
+### How It Flows
+
+1. **You give an instruction** — "Find the cheapest flight from SFO to JFK next weekend" in the Yantra chat.
+2. **Cloud AI plans the steps** — The AI breaks down the task into discrete actions: open a travel site, enter search criteria, compare results.
+3. **Commands relay to your browser** — Each action (navigate, click, type, extract) is sent via secure WebSocket to the extension's service worker.
+4. **Your browser executes locally** — The service worker uses CDP to interact with your actual browser tabs — clicking buttons, reading page content, and navigating links.
+5. **Results flow back** — Extracted data and page state are relayed back to the cloud AI, which decides the next step or presents results to you.
+
+## Chrome DevTools Protocol (CDP) In Action
+
+The extension leverages specific CDP domains for different types of automation:
+
+### Page Navigation and Interaction
+
+Using `Page` and `DOM` domains, the extension can navigate to URLs, wait for page loads, find elements by selector, and simulate clicks and keyboard input — all through the same APIs that Chrome DevTools uses internally.
+
+### Network Monitoring
+
+The `Network` domain allows the extension to observe API calls made by web applications. This is invaluable for understanding what data a page fetches, intercepting responses for extraction, and detecting when a page has fully loaded.
+
+### JavaScript Evaluation
+
+The `Runtime` domain enables safe execution of JavaScript expressions within page contexts. This powers everything from reading form values to triggering complex UI interactions that require programmatic access.
+
+### Screenshot and Visual Analysis
+
+The `Page.captureScreenshot` capability lets the AI "see" what's on screen, enabling visual analysis for tasks like verifying a form was submitted correctly or identifying UI elements that aren't easily addressable by selectors alone.
+
+## The Service Worker: Always Ready
+
+Yantra Browser's service worker is the heart of the extension. Built on Chrome's Manifest V3 architecture, it:
+
+-   **Maintains persistent WebSocket connections** to the Yantra relay server
+-   **Manages CDP sessions** across multiple browser tabs
+-   **Handles authentication** via secure token exchange
+-   **Queues commands** when the browser is briefly unavailable
+-   **Auto-reconnects** after network interruptions with exponential backoff
+
+The service worker lifecycle is carefully managed to stay within Chrome's resource guidelines while remaining responsive to incoming automation commands.
+
+## Security at Every Layer
+
+### Encrypted Communication
+
+All data between the extension and Yantra's servers travels over TLS-encrypted WebSocket connections. There is no plaintext communication at any point.
+
+### Minimal Permission Scope
+
+The extension requests only the permissions it needs:
+
+-   `debugger` — For CDP access to automate browser tabs
+-   `activeTab` — To interact with the currently focused tab
+-   `storage` — To persist authentication tokens locally
+-   `identity` — For secure OAuth sign-in
+
+Host permissions are scoped to Yantra's own domains by default. The extension cannot access arbitrary websites unless you explicitly grant permission through the automation flow.
+
+### No Remote Code Execution
+
+The extension bundles all its code at install time. No external scripts are fetched or evaluated at runtime. This is enforced by Chrome's Manifest V3 Content Security Policy and verified during Chrome Web Store review.
+
+## Real-World Performance
+
+In our internal benchmarks, Yantra Browser automation achieves:
+
+| Metric                      | Performance                                |
+| --------------------------- | ------------------------------------------ |
+| **Command Latency**         | < 100ms round-trip for typical actions     |
+| **Page Load Detection**     | Accurate within 200ms of visual completion |
+| **Multi-tab Orchestration** | Up to 10 concurrent tab automations        |
+| **Reconnection Time**       | < 2 seconds after network interruption     |
+
+These numbers mean that automation feels nearly as fast as a human clicking through pages manually — but with perfect consistency and zero fatigue.
+
+## What's Next
+
+We're actively working on expanding Yantra Browser's capabilities:
+
+-   **Visual Understanding** — AI-powered screen analysis for complex UIs without clean DOM structure
+-   **Workflow Templates** — Pre-built automation recipes for common tasks (expense reports, data migration, content publishing)
+-   **Team Collaboration** — Share automation workflows across your organization with role-based access control
+-   **Custom Actions API** — Let developers extend the extension with domain-specific automation modules
+
+## Try It Today
+
+Yantra Browser is available now on the Chrome Web Store. Install it, connect your Yantra account, and start automating in under a minute.
+
+[Get Yantra Browser →](https://chromewebstore.google.com/detail/yantra-browser)
+
+---
+
+_Want to dive deeper? Check out our [How the Extension Works](/help-center/browser-extension/how-it-works) documentation or reach out to our engineering team at [support@yantra.app](mailto:support@yantra.app)._