mcp-percona-pg
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PERCONA_MODE | No | Access mode: read-only, read-write, or admin. Over-privileged tools are never registered. Defaults to read-only. | read-only |
| PERCONA_DRY_RUN | No | Set to 'true' to run in validate-only/dry-run mode. | false |
| PERCONA_AUDIT_LOG | No | Set to 'true' to write a JSON audit line per guarded operation. | false |
| PERCONA_NAMESPACE | No | Namespace allowlist (used in quickstart examples; may be an alias for PERCONA_NAMESPACE_ALLOWLIST). | |
| PERCONA_ALLOW_DELETE | No | Set to 'true' to allow delete operations (requires admin mode). | false |
| PERCONA_ALLOW_RESTORE | No | Set to 'true' to allow restore operations (requires admin mode). | false |
| PERCONA_ALLOW_UPGRADE | No | Set to 'true' to allow upgrade operations (requires admin mode). | false |
| PERCONA_CLUSTER_ALLOWLIST | No | Comma-separated list of clusters the agent is allowed to access. | |
| PERCONA_PROTECTED_CLUSTERS | No | Comma-separated list of clusters that are readable but never mutated, restored, or deleted. | |
| PERCONA_NAMESPACE_ALLOWLIST | No | Comma-separated list of namespaces the agent is allowed to access. | |
| PERCONA_REQUIRE_CONFIRMATION | No | Set to 'true' to require echoing the cluster name for high-impact operations. | false |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_contextsA | List the contexts (clusters) available in the loaded kube-config. |
| list_clustersA | List PerconaPGCluster resources (PostgreSQL + PgBouncer). Omit |
| get_clusterA | Summary of a PerconaPGCluster: state, PostgreSQL size, PgBouncer, version, standby, host. |
| get_cluster_statusB | The full |
| get_connection_infoA | Connection endpoints for a cluster: the primary/replica Service hosts, port, and the declared users (names only — passwords live in Secrets and are never returned). |
| get_pgbouncer_configB | The PgBouncer settings for a cluster: replicas, pool_mode, and the global pool tunables. |
| get_pg_parametersB | The tuned PostgreSQL parameters from |
| list_backupsB | List PerconaPGBackup resources in a namespace, with their state, repo, and completion time. |
| list_restoresB | List PerconaPGRestore resources in a namespace, with their target cluster and state. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 9 tools
get_cluster and get_cluster_status both describe cluster status and could be confused (summary vs full .status). list_clusters and list_contexts also overlap in listing clusters, though descriptions hint at a distinction. Other tools are clearly distinct.
All tools follow a consistent snake_case verb_noun pattern: list_* and get_*. No naming deviations or mixed conventions.
9 tools is well-scoped for a Kubernetes operator inspection server. Each tool maps to a distinct resource or config aspect, and nothing feels redundant by count.
The server is clearly read-only, but within that scope it lacks singular detail tools like get_backup or get_restore, and has no operations for cluster management (create/update/delete) or triggering backup/restore. Core inspection needs are covered, but notable gaps remain.