Choose the perfect plan for your AI Agent, MCP Platform, or Automations needs
Custom voice minutes per month
10,000 voice minutes per month
20,000 voice minutes per month
50,000 voice minutes per month
Monthly totals are pre-GST, 18% GST is added at checkout. Minutes convert to credits on payment; every plan can run voice, WhatsApp, and automations from the same pool.
We'll match you to the right plan
Connect your AI to 304+ apps with 1,500+ implemented tools
No separate subscription, automations run on the same credit pool as everything else, metered per use with live cost tracking in the dashboard.
1 credit = $0.01 (≈ ₹0.96). Every event is metered and visible per run.
One-time purchase · never expire · GST added at checkout
Find answers to common questions about our products
An AI Agent is an intelligent voice assistant that can handle customer conversations across phone, WhatsApp, and SMS. It uses natural language processing to understand customer queries and respond appropriately, 24/7 without human intervention.
AI Agent supports voice calls (phone), WhatsApp Business, and SMS. You can deploy the same agent across all three channels simultaneously, providing consistent customer experience everywhere.
AI Agent supports 40+ languages including English, Spanish, French, German, Hindi, Mandarin, Arabic, and many more. You can configure multi-language support for your specific needs.
Yes! You can fully customize the agent's voice characteristics, tone, personality, and conversation style to match your brand. Choose from multiple voice options and define how the agent should respond in different scenarios.
Standard inbound or outbound agents can be configured and tested in days; custom CRM and compliance wiring adds setup time — we'll scope it with you.
Yes! AI Agent integrates with popular CRMs like Salesforce, HubSpot, Zendesk, and custom systems via API. It can read customer data, create tickets, update records, and sync conversation history.
You can configure escalation rules. When the agent encounters a question it can't handle, it can transfer to a human agent, create a support ticket, or collect information for follow-up. You have full control over escalation logic.
You can review call transcripts and intent outcomes in the dashboard, then refine prompts and flows to improve accuracy over time.
Yes! You get real-time access to all conversation transcripts, recordings (for voice), analytics dashboards showing resolution rates, customer satisfaction, common queries, and performance metrics.
AI Agents excel at: customer support, order tracking, appointment scheduling, lead qualification, surveys, reminders, FAQs, payment collections, and any repetitive customer interaction that follows predictable patterns.
Voice plans are per-minute subscriptions: Pilot lets you choose 3,000–9,000 minutes at ₹4.75/min plus a ₹5,000 platform fee; plans scale up to Enterprise at 100,000 minutes for ₹3.60/min. WhatsApp and automations draw from the same credit pool, see the plans above for exact monthly totals.
Each call runs in an isolated pipeline, scaling with your telephony provider channels and model quotas; high-concurrency campaign sizing is supported on Growth and Enterprise plans.
Yes! AI Agents can make outbound calls for appointment reminders, payment collections, surveys, follow-ups, and campaigns. You can schedule bulk campaigns or trigger calls based on specific events.
Audio and transcripts are encrypted in transit and at rest, credentials are encrypted at rest, and your conversations are never used to train public models. We follow SOC 2-aligned controls.
Yes! You can create multiple conversation variants and test them with different customer segments. Analytics show which flows perform better, helping you optimize for higher satisfaction and resolution rates.
AI Agents handle voice, WhatsApp, and SMS with natural conversation capabilities. Unlike basic chatbots, they understand context, handle complex queries, integrate deeply with your systems, and provide human-like interactions.
Agents send payment links via WhatsApp or SMS during or after the call using payment connectors (Stripe, PayPal, Razorpay), keeping card entry inside the payment provider's PCI-compliant checkout.
You provide documentation, FAQs, product information, and sample conversations. Our team helps configure the agent with your business logic, policies, and knowledge base. Review transcripts and refine prompts to improve the agent over time.
Yes! You can pause, modify conversation flows, update responses, add new capabilities, or adjust settings anytime through the dashboard. Changes can be deployed instantly or scheduled.
All plans include email support; Professional and Growth plans include priority support; Enterprise plans include a dedicated account manager and custom SLA.
Each call runs in an isolated pipeline, so an agent scales with your telephony provider channels and model quotas rather than a fixed cap; high-concurrency campaign sizing is supported on Growth and Enterprise plans.
Track metrics like: support cost reduction, increased customer satisfaction (CSAT), faster response times, higher resolution rates, reduced no-shows, and staff time saved. You can estimate payback for your exact call volume and agent minutes using our interactive ROI Calculator at /ai-agent-roi-calculator.
MCP Platform provides instant access to 304+ apps through a single integration. Instead of building custom integrations for each app, your AI connects to our MCP server and gets access to 1,500+ implemented tools across all major platforms.
MCP Platform offers 304+ native app integrations with 1,500+ implemented workflow-oriented tools. This includes popular apps like Slack, GitHub, Notion, Salesforce, Google Workspace, and hundreds more.
Simply add our MCP server URL to your AI client (Cursor, Claude, Windsurf, Cline), connect your apps via OAuth, and you're ready to go. Managed OAuth handles the heavy lifting — connect an app with your credentials and it's ready to use.
No! MCP Platform handles all credential management automatically. One-click OAuth for each app, automatic token refresh, encrypted storage, and server-side injection. You never expose credentials to your AI.
Connect any MCP client to our Streamable HTTP endpoint at /api/v1/mcp on your workspace host, using your X-API-Key header. Once connected, your client gets access to all connected apps and tools.
MCP Platform uses credit-based pricing. Each tool call costs credits based on complexity. You can see credit costs in X-Credit-Cost headers. Spend scales with your tool-call volume — credits are billed at ₹0.96 per credit.
MCP Platform works with any MCP-compatible client including Cursor, Claude Desktop, Windsurf, Cline, and custom implementations. If your client supports the MCP protocol, it works with our platform.
Yes! You can connect multiple accounts for the same app (e.g., multiple GitHub organizations, Slack workspaces). Each connection is managed separately with its own credentials.
All credentials are encrypted at rest, stored securely, and never exposed to your AI. OAuth tokens are automatically refreshed and rotated. We follow SOC 2-aligned controls and industry security best practices.
We monitor all 304+ apps and update tools automatically when APIs change. You don't need to worry about breaking changes or maintenance - we handle all updates server-side.
Yes! You can request new apps anytime through our dashboard — we regularly add integrations and prioritize based on demand.
MCP Platform automatically resolves context like tenant IDs, workspace IDs, cloud IDs, and other required parameters. Your AI doesn't need to know these details - we inject them server-side based on your connected apps.
Instead of raw API endpoints, we provide high-level tools designed for common workflows. For example: 'create_jira_ticket', 'send_slack_message', 'read_notion_page' - tools that accomplish complete tasks, not just API calls.
Yes — MCP Platform is built for production with automatic token refresh, exponential-backoff retries, and execution logging. Contractual SLAs are available on Enterprise plans.
Every API response includes an X-Credit-Cost header showing credits consumed. Your dashboard provides real-time usage analytics, cost breakdowns by app, and alerts when approaching limits.
Custom integrations take weeks per app, require ongoing maintenance, and expose credentials. MCP Platform gives you 304+ apps instantly, handles all maintenance, and keeps credentials secure - without building integrations yourself.
Yes! You can connect or disconnect any app instantly through the dashboard. Disconnecting revokes OAuth tokens immediately and removes access. No long-term commitments required.
Yes! Many integrations support webhooks for real-time notifications. You can configure webhooks to trigger actions in your AI when events occur in connected apps.
We implement intelligent rate limiting per app, automatic retry with exponential backoff, and queue management. Your AI gets clear error messages if rate limits are hit, with suggested retry times.
Yes! MCP Platform works with any AI that supports the MCP protocol. Whether you're using GPT-4, Claude, Llama, or custom models, as long as your client supports MCP, it works.
You get detailed analytics on: tool usage by app, credit consumption trends, most-used integrations, error rates, response times, and cost projections. Export data for custom analysis.
You get 50 free credits on signup, enough to connect an app and make real tool calls. After that, top up anytime with credit packs that never expire.
We provide documentation, video tutorials, and API reference. All plans include email support; Professional and Growth plans include priority support; Enterprise plans include a dedicated account manager and custom SLA.
5 credits per run (≈ ₹4.80) plus 1 credit for each integration or tool call the run makes. Test chat turns cost 0.1 credits; builder chat turns cost 0.5 credits. 1 credit = $0.01, and every charge is visible per run in the dashboard.
No. Every subscription plan, Pilot through Enterprise, already includes the automation channel, so runs draw from the same credit pool as calls and WhatsApp. You can also top up with credit packs, which never expire.
Steps retry automatically with backoff, per-app circuit breakers isolate failing integrations, and anything that still can't complete lands in a dead-letter queue you can inspect and replay, nothing is silently dropped.
Yes, that's the point. An automation's output can feed a calling or WhatsApp campaign directly, and automations can also run before a call (to inject CRM context) or after one (to write results back).