Model Context Protocol connects agents to tools and data. It is a Linux Foundation standard.
One OS. All your AI agents work inside it.
Marketing, HR, CRM, Operations, Community. Every agent has access limits. Every action is logged. You keep control.
// architecture
Two protocols. One controlled system.
AGENTS — talk to each other via A2A ↔
One door. Authentication, permissions, and logging in one place. MCP: agent → tools.
Agent2Agent Protocol lets agents delegate tasks to each other with a controlled handoff.
Both are open standards under the Linux Foundation. No vendor lock-in.
Pick your modules. Governance is always included.
MODULE
Agentic Marketing
Campaign ops, content pipeline, lead routing, daily reports.
MODULE
Agentic CRM
Pipeline, auto follow-up, scoring, invoices, and quotations.
MODULE
Agentic HR
Onboarding, leave, knowledge base, and review cycles.
MODULE
Agentic Operations
SOP runner, approval flows, task orchestration, and alerts.
MODULE
Agentic Community
Moderation, engagement, broadcasts, and sentiment reports.
ALWAYS INCLUDED
Governance + MCP Gateway
Strict RBAC, audit log, human review, and one access door for every agent.
Choose your scope. Review each tier.
Each tier defines its modules, access boundary, and handover.
OS Core
- 2 agentic modules
- Governance + MCP Gateway
- Team training
- Docs + SOP
OS Growth
- 4 agentic modules
- Governance + MCP Gateway
- Channel integrations
- Training + review
OS Enterprise
- All modules
- Local AI option
- SLA + priority support
- Security audit
| Tier | Price | Scope |
|---|---|---|
| OS Core | IDR 60M | 2 agentic modules · Governance + MCP Gateway · Team training · Docs + SOP |
| OS Growth | IDR 120M | 4 agentic modules · Governance + MCP Gateway · Channel integrations · Training + review |
| OS Enterprise | From IDR 250M | All modules · Local AI option · SLA + priority support · Security audit |
Monitoring, audit-log review, model and prompt updates, monthly report.
FAQ for humans & AI
04 / 04What is Rama Business OS?
One business operating system of AI agents for Marketing, HR, CRM, Operations, and Community. Strict RBAC, full audit log, and human review. Access uses MCP. Agents delegate through A2A.
What is the difference between MCP and A2A?
MCP connects agents to tools and data. A2A connects agents to other agents for task delegation. Both are Linux Foundation open standards.
How is my data protected?
Each agent follows RBAC. It reads only the data allowed by its role. Every action enters the audit log. Risky actions require human review.
How long is implementation?
Implementation takes 4–8 weeks, based on module count. Milestones are written. You can check progress each week.
Can I start small?
Yes. Start with AI Diagnostic at IDR 1.5M. The result defines the first useful module.
Why not use separate tools?
Separate tools scatter data and permissions. This OS unifies them under one control and one audit log.
Send one message. We start with a diagnosis.
AI Diagnostic at IDR 1.5M. The result becomes your OS scope.
Chat on WhatsApp// tool scope
Choose the work surface.
| Tool | Role | Status |
|---|---|---|
| OpenClaw | Agent environment | Scope review |
| Hermes | Agent environment | Scope review |
| Claude Code | Coding tool | Scope review |
| Codex | Coding tool | Scope review |
