> ## Documentation Index
> Fetch the complete documentation index at: https://docs.aviatrix.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Aviatrix MCP for Enterprise Fabrics

> Connect an MCP-compatible AI agent to the Aviatrix MCP Server for authenticated, read-only access to Controller and CoPilot data: topology, policy, flow intelligence, and diagnostics.

Controller and CoPilot already hold the answer to most network and security
questions your team asks: what does this SmartGroup cover, is there an active
IPS alert, why is this S2C tunnel down. Getting that answer today means opening
a dashboard, navigating to the right view, and cross-referencing policy against
topology by hand. The Aviatrix MCP for Enterprise Fabrics server gives your AI
agent direct, authenticated access to that same data, so it can answer the
question in a single turn instead of sending you back to the UI.

## What This Is

The Aviatrix MCP Server is an Aviatrix-hosted MCP server at
`platform.mcp.aviatrix.com` that gives AI agents authenticated access to your
Controller and CoPilot data across topology, policy, flow intelligence, and
diagnostics. It calls the same Controller and CoPilot APIs the UI uses, so there
are no gateway changes, no Terraform changes, and no new version requirement to
adopt it.

<Info>
  This server is generally available. Aviatrix continues to expand its tool
  coverage (see [What's Next](#whats-next) for the current roadmap).
</Info>

## What Your Agent Can Do

The server exposes 51 tools across four domains:

| Domain                     | Tools | Representative capabilities                                                                                                                                                                           |
| -------------------------- | ----- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Inventory and topology     | 8     | Gateway list and detail, cloud workloads, S2C connections, FireNet inspection, AWS TGW attachments and domains                                                                                        |
| Policy and security        | 15    | DCF groups and policies, firewall log search and aggregation, IPS alerts, audit trail, policy hit stats, enforcement state, gateway performance, gateway throughput (bps)                             |
| Flow intelligence          | 6     | FlowIQ top talkers, spoke egress bandwidth, top egress destinations, IP classification, fabric CIDR enumeration, IP ownership lookup                                                                  |
| Diagnostics and operations | 22    | FlightPath (3 variants + instance discovery), ping, traceroute, BGP diagnostics (async), S2C diagnostics and logs, syslogs (sync and async), controller logs, stateful sessions, connectivity testing |

Ask your agent a question in natural language, for example "What DCF policies
cover the production VPCs, and are there any IPS alerts in the last 24 hours?"
The agent queries groups, policies, and intrusion data in a single turn and can
follow up with an action like running FlightPath between two spokes.

## Permission Scoping

API keys are issued with an explicit scope at creation time, and scope is
enforced server-side, not by agent instruction:

* **`controller:read`** (default): 50 read-only tools covering inventory,
  policy, flow intelligence, and diagnostics. This includes
  `aviatrix_run_bgp_diag`, which is visible to all keys but enforces write scope
  for mutating commands (debug, clear) at execution time.
* **`controller:write`**: adds one operational tool,
  `aviatrix_reset_s2c_connection`, and unlocks write-level BGP commands. A
  `controller:read` key cannot invoke `aviatrix_reset_s2c_connection` regardless
  of what the agent requests.

<Warning>
  Only issue a `controller:write` key to an agent or user that needs to reset
  S2C connections or run BGP diagnostics. Default to `controller:read` for all
  other use cases.
</Warning>

<Note>
  The set of tools gated behind `controller:write` is expected to grow as
  Aviatrix adds more operational capabilities to the server. Review the tool
  list for any key with write scope after an MCP server update, rather than
  assuming its capabilities are fixed at issuance time.
</Note>

## Security

* **Transport:** All communication between your MCP client and the Aviatrix MCP
  Server uses TLS over HTTPS (Streamable HTTP transport).
* **Authentication:** OAuth 2.1 with PKCE. Your agent authenticates on every
  session.
* **Credential storage:** Your Controller and CoPilot credentials are encrypted
  inside the Bearer token itself using AES-256. They are never stored in a
  database. The plaintext exists only in memory for the duration of an API call.
* **API keys:** Encrypted at rest using AES-256. Revoke a key at any time from
  the login portal.
* **Scope enforcement:** Unauthorized tools are absent from the tool list, not
  marked as "access denied." A `controller:read` key cannot discover write-only
  tools.
* **Session lifecycle:** CoPilot sessions expire after approximately one hour.
  OAuth-capable clients (Claude Code, Claude Desktop) re-authenticate
  automatically with no user action required.
* **Audit logging:** Every tool call is logged to CloudWatch as structured JSON,
  including tool name, parameters, timing, and outcome. The log group can be
  forwarded to your SIEM.

## Network Requirements

The Aviatrix MCP Server connects to your Controller and CoPilot from a static IP
address. Allow inbound HTTPS (port 443) from the following address on both your
Controller and CoPilot network security groups:

| Environment | Source IP                 |
| ----------- | ------------------------- |
| Production  | `3.134.16.45` (us-east-2) |

<Note>
  Both Controller and CoPilot must allow this IP. The MCP Server calls CoPilot
  for policy, flow, and topology data, and calls Controller directly for
  diagnostics such as syslog retrieval, ping, and traceroute.
</Note>

## Connect Your Agent

<Steps>
  <Step title="Request access">
    Contact your Aviatrix account team to enable the Aviatrix MCP Server for
    your Controller and CoPilot. Aviatrix allowlists your environment before
    any API key can be issued.
  </Step>

  <Step title="Provision an API key">
    Once your environment is enabled, go to the Aviatrix MCP login portal
    and create an API key. Choose `controller:read` unless your use case
    requires resetting S2C connections or running BGP diagnostics.
  </Step>

  <Step title="Configure your MCP client">
    Add the Aviatrix MCP Server endpoint to your MCP client configuration.
    In Claude Code, run:

    ```bash theme={null}
    claude mcp add --transport http aviatrix-fabric https://platform.mcp.aviatrix.com/mcp
    ```

    Any MCP-compatible client works, including Claude Code, Claude Desktop,
    and Cursor.
  </Step>

  <Step title="Authenticate">
    Your agent completes an OAuth 2.1 with PKCE authentication flow on
    connection. Approve the request when prompted.
  </Step>

  <Step title="Ask a question">
    Ask your agent about your environment, for example "What is the current
    Blast Radius for the production AI workload VPCs?" The agent selects and
    calls the relevant tools and returns a structured answer.
  </Step>
</Steps>

## What's Next

| Phase | Status  | Scope                                                                                             |
| ----- | ------- | ------------------------------------------------------------------------------------------------- |
| 1     | Shipped | DCF visibility: groups, policies, firewall logs, IPS alerts, audit trail                          |
| 2     | Shipped | Operational depth: diagnostics, S2C suite, BGP, FlightPath, syslogs, egress bandwidth             |
| 3     | Shipped | Flow intelligence: FlowIQ analysis, IP classification, fabric CIDR enumeration, egress annotation |
| 4     | Planned | Write operations with safety gates, RBAC, multi-tenant isolation                                  |

## Security FAQ

<AccordionGroup>
  <Accordion title="Does the AI model (Claude, GPT, etc.) see my Controller password?">
    No. Your credentials are encrypted inside the Bearer token and decrypted
    only by the Aviatrix MCP Server at runtime. The AI model receives
    structured tool results (gateway lists, policy data, diagnostic output),
    never credentials.
  </Accordion>

  <Accordion title="Where are my credentials stored?">
    Inside the encrypted Bearer token itself, not in a database. The token uses
    AES-256 encryption. The plaintext password exists only in Lambda memory for
    the duration of an API call (typically under one second), then is discarded.
  </Accordion>

  <Accordion title="Can the AI model modify my network configuration?">
    With a `controller:read` key (the default), no. Read-only keys expose 50 tools
    that can query data but cannot change state. The only state-changing tool
    (`aviatrix_reset_s2c_connection`) requires an explicit `controller:write` key
    and is hidden from read-only sessions.
  </Accordion>

  <Accordion title="What happens if I revoke an API key?">
    The key is invalidated immediately. Any in-flight or subsequent requests using
    that Bearer token will fail authentication. No grace period.
  </Accordion>

  <Accordion title="Does the MCP Server store conversation history or query results?">
    No. The MCP Server is stateless with respect to conversation content. Tool
    call metadata (tool name, timing, outcome) is logged to CloudWatch for audit
    purposes, but query results and conversation context are not retained by the
    server.
  </Accordion>

  <Accordion title="What network access does the MCP Server have to my environment?">
    The MCP Server connects to your Controller and CoPilot on port 443 only, from
    a single static IP (`3.134.16.45`). All other outbound ports are blocked by
    policy. The server cannot reach any other hosts in your environment.
  </Accordion>

  <Accordion title="What happens to data retrieved from my environment?">
    Tool results (gateway lists, firewall logs, diagnostic output) are
    returned to the MCP client over TLS in a single response. The Aviatrix
    MCP Server does not persist, cache, or index query results. Data exists
    in Lambda memory only for the duration of the request (typically under
    one second), then is discarded when the execution context ends. No
    customer environment data is written to disk, object storage, or any
    database at any point during processing.
  </Accordion>
</AccordionGroup>

## Related Reading

For guidance on securing the MCP servers your own team deploys (not the Aviatrix
MCP Server itself), see

<a href={"/docs/enterprise/" + "10.1" + "/solutions/security-for-ai/ai-agent-workload-security"}>AI Agent Workload Containment</a>
and
<a href={"/docs/enterprise/" + "10.1" + "/solutions/security-for-ai/mcp-server-egress-security"}>MCP Server Security Through Containment Architecture</a>
.