Skip to content

fix: mcp docs #132

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
levkk merged 1 commit into from
Nov 11, 2026
Merged

fix: mcp docs #132

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
36 changes: 18 additions & 18 deletions docs/enterprise_edition/mcp.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,11 +8,11 @@ icon: material/robot

MCP is a new feature. Please test it before deploying to production.

MCP (model context protocol) allows coding agents and harnesses to fetch real time PgDog metrics, logs and deployment information to help debug production incidents or configuration issues.
MCP (model context protocol) allows coding agents to fetch real time PgDog metrics, logs and deployment information to help debug production incidents or configuration issues.

## How it works

MCP is an HTTP protocol and most coding harnesses, e.g., Codex, Pi, Claude Code support it out of the box. It's disabled by default, so to enable MCP you need to set the following config in the [control plane](control_plane/index.md):
MCP is an HTTP protocol and most harnesses, e.g., Codex, Pi, Claude Code support it out of the box. It's disabled by default, so to enable MCP you need to set the following config in the [control plane](control_plane/index.md):

=== "Helm chart"

Expand All @@ -30,16 +30,16 @@ MCP is an HTTP protocol and most coding harnesses, e.g., Codex, Pi, Claude Code
enabled = true
```

When enabled, the MCP server will be running on the same host and port as the control plane, under the `/mcp` path.
When enabled, the MCP server will be running on the same host and port as the control plane, under the `/mcp` URL.

### Authentication
## Authentication

Requests to the MCP server require Bearer token authentication. Unauthenticated requests will be rejected automatically. To configure authentication, you have two options:

1. Hardcode a set of tokens in the configuration
2. Enable [Raft](control_plane/ha.md) and add tokens via the web dashboard

#### Configuring tokens
### Configuring tokens

A fixed set of Bearer tokens can be configured as follows:

Expand All @@ -65,7 +65,7 @@ A fixed set of Bearer tokens can be configured as follows:

If you're running our [Helm chart](control_plane/installation.md), you can also load a comma-separated list of tokens from a Kubernetes Secret.

#### Create a Secret
### Using a Secret

Create the Secret in the same namespace as the control plane:

Expand All @@ -91,19 +91,19 @@ control:
tokensKey: tokens
```

#### Dynamic tokens
### Dynamic tokens

Managing tokens via configuration can be tedious, so the control plane also supports storing authorized tokens in its [Raft](control_plane/ha.md) log. Stored tokens are checked on each request to the MCP server, so tokens can be created and deleted at any time without configuration changes.
Managing tokens via configuration can be tedious, so the control plane also supports storing Bearer tokens in its [Raft](control_plane/ha.md) log. Those are checked on each request to the MCP server, so they can be created and deleted at any time without configuration changes.

<center>
<img src="/images/ee/mcp_system.png" width="100%" alt="MCP Add token">
</center>

### Configuring agents
## Configuring agents

Most agent harnesses support MCP out of the box, so configuring PgDog's MCP server should be pretty straightforward.
Most harnesses support MCP out of the box, so configuring PgDog's MCP server should be pretty straightforward.

#### Codex
### Codex

To add the MCP server to Codex, you can add it to either `.codex/config.toml` or `~/.codex/config.toml` as follows:

Expand All @@ -122,13 +122,13 @@ http_headers = { Authorization = "Bearer <TOKEN>" }

The MCP server comes with four tools:

| Tool | Description |
| ----------------------- | ------------------------------------------------------------------------------------------------------------------------- |
| `pgdog.proxy_instances` | Show an up-to-date list of PgDog instances currently running. | |
| `pgdog.proxy_metrics` | Fetch real-time PgDog metrics. The tool allows you to filter by database/user, by Kube instance name, and by time window. |
| `pgdog.proxy_logs` | Fetch real-time logs from Kube containers running PgDog. |
| `pgdog.proxy_config` | Fetch latest configuration loaded into each PgDog instance. |
| `pgdog.auth_check` | A no-op tool that allows you to verify if the harness has the correct Bearer token. |
| Tool | Description |
| ----------------------- | ----------------------------------------------------------------------------------------------------------------------------------- |
| `pgdog.proxy_instances` | Show an up-to-date list of currently running PgDog instances. |
| `pgdog.proxy_metrics` | Fetch real-time PgDog instance metrics. This tool allows you to filter by database/user, by Kube instance name, and by time window. |
| `pgdog.proxy_logs` | Fetch real-time logs from Kube containers running PgDog. This tool works only if PgDog is deployed in Kube. |
| `pgdog.proxy_config` | Fetch latest configuration from each PgDog instance. |
| `pgdog.auth_check` | A no-op tool that allows you to verify if the harness has the correct Bearer token. |

An agent typically doesn't need special instructions to use these tools. For example, a common prompt to debug a production incident could be:

Expand Down
Loading