# Check Point Avanan MCP Server

An [MCP](https://modelcontextprotocol.io) server for the **Check Point Harmony Email &
Collaboration** API (formerly Avanan), reached through the Check Point Infinity Portal. It lets
an AI agent triage security events, search scanned emails and files, quarantine or restore
them, and manage allow / block exception rules.

- Transport: MCP Streamable HTTP at `POST /mcp`, health probe at `GET /health`
- Stateless and multi-tenant: credentials arrive on every request as HTTP headers and are
  never stored; the access token, region and scopes are resolved per request and not cached
- Python 3.12, [uv](https://docs.astral.sh/uv/), FastMCP, httpx
- License: Apache-2.0, Copyright 2026 MSPbots.ai

For the older SmartAPI (`smart-api-production-*.avanan.net`, MSP tenant management), use the
separate Avanan (legacy) server.

<div align="center">

## MSPbots Connector

**The MCP Gateway built for MSPs: visible, controlled, and your data stays where it belongs.**

[**Get started at connect.mspbots.ai →**](https://connect.mspbots.ai)

</div>

---

MSPbots Connector gives your AI clients one MCP URL to reach your entire MSP stack (RMM, PSA, documentation, security, finance), and gives you control over who can call which tool and a record of everything that was called.

**Hosted service: [connect.mspbots.ai](https://connect.mspbots.ai)** · Free to use.

### Why Connector

![MSPbots Connector](https://stgoss.mspbots.ai/files/mcp/mspbots-mcp.png)

| | |
|---|---|
| **Visible** | Every tool call is logged: who, which vendor, when. Usage dashboard, CSV export, SIEM streaming. |
| **Controlled** | Role-based, per-tool allowlist. Deny by default. |
| **Protected** | Credentials encrypted per record. AI clients never see your vendor passwords or tokens. |
| **MSP-native** | Built for the MSP stack only: RMM, PSA, documentation, security, finance. |

### Quick start

1. **Sign up** at [connect.mspbots.ai](https://connect.mspbots.ai).
2. **Connect your vendors.** Pick from 150+ connectors and authorize each one on the vendor's own page, or bring your own API keys. No extra vendor accounts needed.
3. **Set access.** Create roles and decide which tools each role may call. Nothing is allowed until you allow it.
4. **Add your MCP URL to your AI client.** Copy the URL from your Connector dashboard into Claude, ChatGPT, Cursor, VS Code, Gemini, Codex, or any MCP-capable client.
5. **Ask your AI to do real work**, and watch every call appear in the audit log.

### Features

**Unified gateway**
- One MCP URL for all your AI clients
- 150+ connectors across the MSP stack
- Custom MCP servers supported, free
- Authorization happens on each vendor's own page; the AI never sees passwords or tokens
- Bring your own credentials: use the vendor API keys you already have
- Multiple accounts per app; connect once, reuse across AI clients
- Tool search: finds the right action for the task
- One sign-in; vendor credentials are injected behind the scenes and OAuth tokens refresh automatically

**Visibility**
- Audit log for every tool call (who, which vendor, when)
- Usage dashboard: call volume, top vendors, error rate, per-user activity
- Export audit logs to CSV, or stream to Loki, Graylog, or LogScale (SIEM)
- Optional recording of tool arguments and result summaries (off by default, enabled per organization)

**Control**
- Per-role, per-tool allowlist, deny by default
- owner / admin / member roles
- SCIM provisioning with Entra ID, Okta, JumpCloud, and Google Workspace

**Security**
- Credentials encrypted one record at a time (AES-256-GCM, per-record key)
- AI clients authenticate with OAuth 2.1 + PKCE
- Vendor responses are scanned for prompt-injection patterns and returned as structured data. This is one layer of defense, not a guarantee against every attack.
- Static egress IP for customer firewall and API allowlisting

### Pricing

**Free.**

Links: [connect.mspbots.ai](https://connect.mspbots.ai) · [mspbots.ai](https://mspbots.ai)

---

## Authentication headers

| Header | Required | Type | Default | Allowed values | Description | Example |
|---|---|---|---|---|---|---|
| `X-Checkpoint-Client-Id` | Yes | string | none | - | Infinity Portal API key client id | `your-client-id` |
| `X-Checkpoint-Client-Secret` | Yes | string | none | - | Secret paired with the client id | `your-client-secret` |
| `X-Checkpoint-Region` | No | string | detected | `eu`, `us`, `ca`, `au` (`ap`), `uk`, `uae` (`me`), `in` | Data region of the account | `us` |

Where to get them:

- **Client id / secret**: in the Check Point Infinity Portal create an API key for the
  **Email & Collaboration** service (Global Settings -> API Keys). See the Check Point Portal
  Administration Guide.
- **Region**: keys are regional. Without `X-Checkpoint-Region` the server exchanges the key for
  a token at the EU gateway, uses the region carried in the token, and if that region returns
  no scopes it tries the other regional gateways.

A request to `/mcp` without the required headers gets HTTP 401 listing them. An unknown region
does not block `tools/list`; tools then return an `invalid_argument` error naming the allowed
regions.

## Tools

All tools return a JSON string. List tools return `{items, count, total, has_more, scroll_id}`;
pass `scroll_id` back as `scrollId` for the next page. The API has no page size: if a batch is too large for one response, the first records
that fit are returned without `scroll_id` (following it would skip the rest of the batch), and the note
says to narrow the query. Empty values are left out of list records. With a key that covers several tenants,
queries run across all of them and action tools need a `scope` (`farm:tenant`).

| Tool | Purpose | Parameters (required in **bold**) |
|---|---|---|
| `hec_query_events` | Search security events | eventTypes, eventStates, severities, startDate, endDate, saas, eventIds, scrollId |
| `hec_get_event` | One event | **eventId** |
| `hec_search_emails` | Search secured entities (emails, files) | **saas**, **startDate**, endDate, saasEntity, filters (`[{saasAttrName, saasAttrOp, saasAttrValue}]`), scrollId |
| `hec_get_email` | One entity | **entityId** |
| `hec_quarantine_events` | Quarantine the emails behind events (**destructive**, reversible) | **eventIds**, scope |
| `hec_restore_events` | Restore them (**write**) | **eventIds**, scope |
| `hec_quarantine_emails` | Quarantine emails by entity id (**destructive**, reversible) | **entityIds**, entityType, scope |
| `hec_restore_emails` | Restore emails (**write**) | **entityIds**, entityType, scope |
| `hec_get_task_status` | Progress of a quarantine / restore task | **taskId**, scope |
| `hec_list_exceptions` | Allow / block rules | **excType** (`whitelist`, `blacklist`) |
| `hec_add_exception` | Add a rule (**write**) | **excType**, senderEmail, senderDomain, senderName, recipient, subject, attachmentMd5, comment, senderEmailMatching, senderDomainMatching, subjectMatching |
| `hec_update_exception` | Update a rule (**write**) | **excType**, **excId**, same fields |
| `hec_delete_exception` | Delete a rule (**destructive**) | **excType**, **excId** |

Errors come back as a JSON envelope instead of an exception:

```json
{"error": {"code": "unauthorized", "message": "Check Point rejected the client id / secret (token request failed)", "retryable": false}}
```

`code` is one of `not_configured`, `unauthorized`, `not_found`, `invalid_argument`,
`rate_limited`, `upstream_error`.

## Run locally

```bash
uv sync
uv run python -m avanan_mcp          # listens on 0.0.0.0:8080
```

or with Docker:

```bash
docker build --platform linux/amd64 -t avanan-mcp:dev .
docker run --rm -p 8080:8080 avanan-mcp:dev
```

Configuration (environment variables, all optional): `MCP_HTTP_HOST` (default `0.0.0.0`),
`MCP_HTTP_PORT` (default `8080`), `LOG_LEVEL` (default `INFO`). Credentials are never read from
the environment.

## Testing example

```bash
# list tools
curl -s http://localhost:8080/mcp \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -H "X-Checkpoint-Client-Id: your-client-id" \
  -H "X-Checkpoint-Client-Secret: your-client-secret" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}'

# phishing events since a date
curl -s http://localhost:8080/mcp \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -H "X-Checkpoint-Client-Id: your-client-id" \
  -H "X-Checkpoint-Client-Secret: your-client-secret" \
  -d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"hec_query_events","arguments":{"eventTypes":["phishing"],"startDate":"2026-01-01T00:00:00.000Z"}}}'
```

Unit tests (no network access needed):

```bash
uv run pytest
```

## Delivery notes

- **Tool names** keep their existing `hec_` prefix because they are an existing contract.
- **Write tools.** Quarantine, restore and exception changes affect users' mailboxes and mail
  filtering. They are annotated (`readOnlyHint=false`; quarantine and delete also
  `destructiveHint=true`) and their descriptions tell the agent to act only on an explicit
  request. `hec_list_exceptions` is read-only.
- **Optional additions.** `X-Checkpoint-Region` and the `scope` argument on action tools are
  optional and do not change behaviour for existing connections.

## License

Apache License 2.0 - see [LICENSE](LICENSE) and [NOTICE](NOTICE).