Skip to content
Agentic Control Plane Beta
Setup in 5 minutes

Your team is using AI with your data.
Who is watching?

Agentic Control Plane (ACP) ties every AI request to a real user — with identity, permissions, and audit trails.

See what secure access to your CRM, finance, and ops data through AI looks like

Choose a role

ACP enforces per-user access control. Each role below can only reach the tools it's been granted. This is one of six governance layers ACP applies to every request.

Your AI has a blind spot

Today's AI stacks have a three-party problem: the user, the LLM, and your backend. The user authenticates with the LLM — but when the LLM calls your tools, it forwards a shared API key. Your backend has no idea who made the request. Identity is lost, and nobody can enforce policy, attribute cost, or audit who did what.

The User Has identity, roles, permissions JWT The LLM Knows the user Forwards with shared API key API key Your Backend Can't verify who called Can't filter, enforce, or audit Identity is lost between the LLM and your backend

How ACP solves this:

User Authenticated LLM Calls tools Bearer token Agentic Control Plane Identity & access control PII detection & redaction Policy & rate limits Full audit trail x-user-uid Backend Verified identity Every request is verified, scoped, and attributed to a real user
An Agentic Control Plane solves this by sitting between the LLM and your backend — verifying the user's identity, enforcing policies, and forwarding every request with verified user context. One governance layer for your entire AI stack.

The biggest companies confirmed the problem

Every major AI and cloud vendor is racing to add governance. ACP gives you the same capabilities today — open source, no vendor lock-in, deployed in minutes.

Cisco · State of AI Security 2026
83% of organizations deployed agentic AI. Only 29% felt ready to do it securely — a 54-point readiness gap in the "connective tissue" of the AI economy.
Microsoft · Cyber Pulse Report 2026
80% of Fortune 500 companies are using active AI agents — most without governance.
Microsoft launched AI Foundry and Copilot Studio governance controls to address the gap
CSA + Google Cloud · State of AI Security & Governance 2025
"Governance maturity is the single strongest predictor of secure, scalable AI adoption."
Google urges a "full-stack, AI-driven approach to security"

What you get

Who accessed what, and when
Every AI tool call is tied to a real user. Enforce scopes, roles, and ABAC policies per tool.
Cost and usage guardrails
Set per-user rate limits, cost budgets, and agent guardrails. Prevent runaway usage before it starts.
PII protection, automatically
Detect and redact PII in inputs and outputs. Block requests that exceed risk thresholds.
Multi-tenant
Isolate teams into workspaces. Each tenant gets its own connectors, policies, users, and audit log.
Works with every AI client
ChatGPT Claude Desktop Claude Code Cursor Windsurf Cline Continue Zed Any MCP client

Built-in connectors. Custom APIs. One URL.

Start with pre-built connectors for popular services, or point ACP at any HTTP API and it becomes an MCP tool — with auth, validation, and audit built in.

GitHub
Repos, issues, PRs, code search
Salesforce
Records, queries, objects
Slack
Channels, messages, threads
Jira
Issues, projects, boards
Confluence
Pages, spaces, search
Linear
Issues, projects, teams
Notion
Pages, databases, blocks
HubSpot
Contacts, deals, companies
Google
Drive, Calendar, Gmail
Custom APIs
Any REST endpoint, your schema

Custom connectors are defined in the dashboard — point to any URL, define the input schema, and ACP handles authentication forwarding, SSRF protection, and request validation.

Set up in minutes, not months

1
Create a workspace
Sign in with email or Google and pick a slug for your team.
2
Connect your login system
Link Google, Okta, or any identity provider. ACP verifies your users automatically.
3
Enable connectors
Turn on GitHub, Salesforce, or add custom API endpoints from the dashboard.
4
Give AI the URL
Point any MCP client at your workspace URL. Identity, governance, and audit are handled.

Built for regulated industries

Healthcare
Without
Shared API keys. No way to verify which clinician accessed patient data.
With ACP
Per-physician identity binding, PII redaction, HIPAA-ready audit trails.
Fintech
Without
SSNs in prompts. No per-officer spend limits. No regulatory audit trail.
With ACP
Identity-scoped access, PII masking, per-user budgets, compliance logs.
Legal
Without
Junior associates access partner-level cases. Privilege blind spots.
With ACP
Matter-scoped access, privilege flagging, attorney-level audit trails.
SaaS & Operations
Without
Shared service accounts. No way to attribute AI actions to individuals.
With ACP
Per-user identity, usage guardrails, multi-tenant workspaces, full audit trail.

Start free. Managed from $10/day.

Open Source

$0
  • All 6 governance layers
  • Unlimited everything
  • Self-host anywhere
View on GitHub

Free

$0
  • 1,000 calls/month
  • 2 connectors
  • Unlimited seats
Start free
Most Popular

Pro

$10 /day, billed annually
  • 250,000 calls/month
  • Unlimited connectors
  • Full PII enforcement
Start with Pro

Enterprise

Custom
  • Unlimited everything
  • Custom PII patterns
  • Dedicated support
Contact us

See full pricing details →

Give your team AI — without the risk

Identity, permissions, and audit trails from day one. Start safe. Scale fast.