MCPg : Production Grade Postgresql Mcp Server

MCP数据服务

提供方 devopam

数据服务 未标注 会话登录 社区 / 第三方 Claude DesktopClaudeCursorGemini CLI

A safe-by-default PostgreSQL Model Context Protocol server for AI agents.

一句话选型结论

按资料字段整理

MCPg : Production Grade Postgresql Mcp Server 属于数据服务类 MCP 服务器,传输方式未标注,会话登录,可直接接入 Claude Desktop、Claude、Cursor 等,没有现成安装命令,需按官方文档接入。接入前先确认权限范围;涉及写入、删除或付款的操作,建议先在测试环境验证。

分类数据服务 传输方式未标注 鉴权会话登录 部署方式未标注
GitHub 星标 未标注 仓库热度,无公开仓库时显示未标注
工具数 未标注 暴露给模型的工具数量
资源 / 提示词 0 / 0 资源与提示词模板数量
维护方 社区 / 第三方 核验等级 仅资料核对

工具与能力

数据服务
数量来自来源目录的标注,具体工具清单以官方文档为准
工具数量未标注暴露给模型可调用的工具个数
资源数量0可读取的资源条目
提示词数量0内置提示词模板数量

能力标签

支持客户端

适用场景

mcpserverpostgresql

安装与配置

未标注 会话登录
上游未给出传输方式
  1. 1
    准备运行环境

    按官方文档确认运行环境(常见为 Node.js 或 Python),装好后再执行安装命令。

  2. 2
    安装服务

    复制上面的安装命令执行;命令里带占位符的位置,换成自己的密钥或路径。

  3. 3
    写入客户端配置

    在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,让新配置生效。

  4. 4
    验证是否可用

    连接成功后随便调一个只读工具,确认能返回结果,再接入正式流程。

典型使用场景

3 类
按分类与标签整理的常见用法,具体以官方文档为准
数据库查询

用自然语言查库、取数,不用手写 SQL。

数据同步与清洗

把多来源数据拉到一处做整理。

报表取数

按口径快速拉出统计结果。

选择建议

选之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 已适配 Claude Desktop、Claude、Cursor,换客户端不用重新研究配置。

先想清楚的地方

  • 没有现成安装命令,需要按官方文档自己填启动命令和参数。
  • 使用前要准备好会话登录对应的凭据,并按最小权限配置。
  • 资料没标注工具数量,实际能力要装好后确认。
  • 版本更新与行为细节以官方仓库为准。

更适合

  • 需要在数据服务场景里让模型直接调用外部能力的人。
  • 已经在用 Claude Desktop、Claude 的用户,接入路径最短。
  • 想先快速试一下、能接受在本地或自有环境跑服务的人。

不太适合

  • 不方便管理密钥或第三方授权的人。
  • 需要厂商级 SLA 和技术支持的生产环境。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
MCP 服务器工具数传输方式部署方式安装命令最后核验
MCPg : Production Grade Postgresql Mcp Server当前 未标注 未标注 未标注 无 未标注
Memory 9d25f64e 9 未标注 未标注 无 2026-05-17
Getzep MCP 9 stdio · 本地进程 本地部署 有 2026-05-17
anything-llm 未标注 stdio · 本地进程 本地部署 有 2026-04-11
Prisma MCP Server 未标注 stdio · 本地进程 本地部署 有 2026-05-17
Astr Bot 未标注 stdio · 本地进程 本地部署 有 2026-05-17

服务器档案

数据服务
字段来自项目仓库与公开索引,中文说明由本站标注
提供方devopam上游标注的维护方 devopam
分类数据服务本站按上游分类展示
连接方式未标注上游未给出传输方式
鉴权方式会话登录远程服务建议用最小权限凭据
维护方社区 / 第三方核验等级 仅资料核对
包名 / 运行时未标注 · Cross-platform本地安装时用到的包标识与运行时
部署类型未标注上游给出的部署形态
最近核验未标注上游最后核验时间

详细介绍

32.6 KB
来自项目仓库与上游目录的原文,未做删改

MCPg

A production-grade Model Context Protocol
server for PostgreSQL.
It lets AI agents safely inspect, query, operate, and
tune a Postgres database — 279 tools spanning catalog introspection,
query intelligence, natural-language SQL, structural diffs, hybrid search,
graph queries, data movement, live ops, and more.

> Try it live: point an MCP client — or the MCP Inspector — at the hosted, read-only demo endpoint https://devopam-mcpg-demo.hf.space/mcp. It serves read tools against throwaway demo data; for real use, run MCPg next to your own database (see Quick start).

📍 Listed On

  • Official MCP Registry
  • mcp.so
  • mcpservers.org
  • Smithery
  • Glama
  • MCP ServerList

Aspect MCPg
Safety Read-only default + AST validation
Transport stdio + HTTP/SSE
Install pip install mcpg
Postgres Versions 14–19
Key Differentiator Production observability + multi-tenancy

Why MCPg

  • Safe by default. Read-only access mode. Every user-supplied SQL
    statement parses through a validated AST allowlist before execution.
    Names that reach PostgreSQL as SQL identifiers are encoded with
    delimited-identifier quoting (see Identifier policy),
    not rejected merely for containing hyphens or other legal characters.
    Capabilities like DDL, shell, and LISTEN/NOTIFY are off until you
    opt in. Every tool publishes MCP ToolAnnotations (readOnlyHint,
    openWorldHint) derived from those same gates, so clients can
    auto-approve reads and gate writes without guessing.
  • One server, broad surface. Application data access (queries, search,
    cursors, NL→SQL) and DBA-grade operations (health checks, index tuning,
    EXPLAIN analysis, locks, vacuum, dumps, replicas, migrations) in a
    single MCP server. Agents don't have to switch tools to switch tasks.
  • PostgreSQL-native everything. No ORM, no abstraction tax — uses
    psycopg3 directly, speaks every pg_* system view, integrates with
    TimescaleDB, pgvector, PostGIS, Apache AGE, and pg_stat_statements
    where they're available, and degrades gracefully when they aren't.
  • Production-shaped, not demo-shaped. Connection pooling, per-request
    SET ROLE multi-tenancy, read-replica routing with degraded-host
    detection, server-side cursors with dedicated connections,
    rate-limiting, audit trail with regex redaction, PG TLS enforcement
    on startup, OIDC JWT bearer auth, per-session statement / lock
    timeouts.
  • Observability built in. Prometheus /metrics endpoint on the
    HTTP transport surfaces mcpg_tool_calls_total{tool,status} +
    mcpg_tool_duration_seconds. Every tool call records a structured
    audit event with credential-redacted arguments.
  • Test-driven, multi-version. 2,500+ unit tests plus an integration
    suite that runs against a real PostgreSQL container in CI — matrix
    covers PG 14, 15, 16, 17, 18 on every push, plus PG 19 (beta)
    as an experimental (non-blocking) entry tracked under issue #120.

Install

From PyPI (recommended)```bash

pip install mcpg

or, in an isolated venv exposed globally:

uv tool install mcpg


Verify:```bash
mcpg --version

Docker

Pull the pre-built image from the GitHub Container Registry (published
on every tagged release — :latest tracks the newest, or pin a version
like :0.6.5):```bash
docker pull ghcr.io/devopam/mcpg:latest
docker run --rm --name mcpg -p 8000:8000 \
-e MCPG_DATABASE_URL=postgresql://user:pass@host:5432/db \
-e MCPG_ACCESS_MODE=read-only \
ghcr.io/devopam/mcpg:latest


On **Windows PowerShell** replace the trailing `\` with a backtick `` ` ``
(or put the command on one line); the installation
guide has ready-to-copy
Linux/macOS, PowerShell, and Command Prompt blocks.

Or build it yourself from source:```bash
docker build -t mcpg https://github.com/devopam/MCPg.git

Multi-stage image: runtime stage drops the build toolchain, runs as
uid=10001 / gid=10001 with nologin shell, application files
root-owned and read-only to the runtime user.

From source (developers)```bash

git clone https://github.com/devopam/MCPg && cd MCPg
uv sync


`uv sync` creates a venv with all runtime + dev dependencies and exposes
the `mcpg` console script.

More detail in the Installation Guide.

---

## Quick start

**One-click installs:**

— setup for Windsurf, JetBrains, Zed, Cline, Antigravity, Qwen Code, Perplexity,
ChatGPT, Copilot Studio, Continue, and HTTP
clients in the integrations guide.

### One-click install in Claude Desktop (.mcpb)

Download `mcpg-.mcpb` from the
latest release and
double-click it (or drag it into Claude Desktop's Settings →
Extensions). You'll be prompted for your PostgreSQL connection URL —
stored in the OS keychain — and an access mode (defaults to
read-only). That's the whole install: the bundle is ~2 kB and the
host resolves the pinned `mcpg` release from PyPI for your platform.

### Or wire it up manually (stdio transport)

Drop this into your `claude_desktop_config.json` (macOS:
`~/Library/Application Support/Claude/claude_desktop_config.json`;
Windows: `%APPDATA%\Claude\claude_desktop_config.json`):```json
{
  "mcpServers": {
    "mcpg": {
      "command": "uvx",
      "args": ["mcpg"],
      "env": {
        "MCPG_DATABASE_URL": "postgresql://user:pass@localhost:5432/mydb"
      }
    }
  }
}

Restart Claude Desktop. The MCPg toolset is now available to the model.
You can ask Claude things like:

> "What schemas exist in this database? For each one, summarise the
> three biggest tables."

> "Why is this query slow?
> SELECT * FROM orders WHERE customer_id = 42 ORDER BY created_at DESC"

No interesting data yet? Seed the demo dataset```bash

MCPG_DATABASE_URL=postgresql://... mcpg --demo


One command seeds a small, curated e-commerce dataset (3,000 orders,
900 product reviews, deliberately planted flaws) into an `mcpg_demo`
schema — engineered so the index advisor, query-plan analysis,
full-text search, PII audit, and graph projection all have something
real to find on your first try. See the
guided tour for a captured walkthrough, and remove it
any time with `mcpg --demo-drop`.

### Run as an HTTP server (for IDE integrations, web apps, etc.)```bash
MCPG_DATABASE_URL=postgresql://user:pass@localhost:5432/mydb \
MCPG_TRANSPORT=streamable-http \
MCPG_HTTP_PORT=8000 \
mcpg

Then point any MCP-aware client at http://localhost:8000/mcp (or
/sse for the SSE transport). The HTTP transport refuses to start
unless it's authenticated — set MCPG_HTTP_AUTH_TOKEN=... for a static
bearer, or MCPG_AUTH_MODE=oidc for full JWT validation against an
OIDC issuer. To deliberately run without auth (not recommended), set
MCPG_HTTP_ALLOW_UNAUTHENTICATED=true.


Configuration

MCPg is configured entirely through environment variables — no
config file, no flags (the CLI's --version / --demo / --demo-drop
are one-shot commands, not configuration). The only required one is
MCPG_DATABASE_URL; everything else has a safe default.

Common scenarios

Scenario Set
Local exploration, read-only MCPG_DATABASE_URL
Read-write app data access MCPG_ACCESS_MODE=restricted
DBA toolkit (DDL, vacuum, etc.) MCPG_ACCESS_MODE=unrestricted + MCPG_ALLOW_DDL=true
HTTP transport with bearer auth MCPG_TRANSPORT=streamable-http + MCPG_HTTP_AUTH_TOKEN=…
Multi-tenant SaaS MCPG_DEFAULT_ROLE=tenant_a + MCPG_ALLOWED_ROLES=tenant_a,tenant_b,…
Read-replica fan-out MCPG_REPLICA_URLS=postgresql://…?sslmode=require,postgresql://…?sslmode=require
NL→SQL — single provider Set any one vendor key (ANTHROPIC_API_KEY, OPENAI_API_KEY, GEMINI_API_KEY, XAI_API_KEY, GROQ_API_KEY, HF_TOKEN, … — 22 built-in providers). MCPg auto-picks the default.
NL→SQL — multiple providers, caller picks Set all vendor keys you want active. Each call to translate_nl_to_sql can pass provider="…" (any configured built-in or custom).

Full reference

Core

Variable · Default · Description
MCPG_DATABASE_URL · required · Primary PostgreSQL DSN. Supports URI (postgresql://…) and keyword (host=… user=…) forms. Remote hosts require sslmode=require (or stronger).
MCPG_ACCESS_MODE · read-only · read-only \ · restricted (allows write tools) \ · unrestricted (also unlocks DBA tools when paired with the gate vars).
MCPG_TRANSPORT · stdio · stdio (default, for Claude Desktop) \ · streamable-http \ · sse.
MCPG_LOG_LEVEL · INFO · DEBUG \ · INFO \ · WARNING \ · ERROR \ · CRITICAL.
MCPG_HTTP_HOST · 127.0.0.1 · Bind address for HTTP transports. Set to 0.0.0.0 inside containers.
MCPG_HTTP_PORT · 8000 · Listen port for HTTP transports (1–65535).

Capability gates (opt-in for higher-blast-radius tools)

Variable Default Description
MCPG_ALLOW_DDL false Expose DDL tools (run_ddl, create_graph, drop_graph, hypertable tools, migration tools). Requires MCPG_ACCESS_MODE=unrestricted.
MCPG_ALLOW_SHELL false Expose subprocess-backed tools (dump_database, restore_database, run_pg_binary). Required PG client binaries must be on PATH.
MCPG_ALLOW_TIMESCALEDB_ADMIN false With unrestricted mode + MCPG_ALLOW_DDL, expose the destructive / heavy TimescaleDB tools (drop_chunks, compress_chunk, decompress_chunk, reorder_chunk, merge_chunks, refresh_continuous_aggregate, run_job, alter_job, delete_job); each needs dry_run then confirm=true + confirmation_token. See docs/timescaledb.md.
MCPG_ALLOW_ALTER_SYSTEM false Expose alter_system, alter_system_reset_all, reload_config (unrestricted mode only; two-step confirm=true flow — see docs/alter-system.md). MCPG_ALTER_SYSTEM_ALLOWLIST / MCPG_ALTER_SYSTEM_DENYLIST (comma-separated GUC names) narrow or extend the built-in denylist.
MCPG_ALLOW_LISTEN false Expose LISTEN/NOTIFY tools (subscribe_channel, poll_notifications, unsubscribe_channel, list_notification_subscriptions).

Authentication (HTTP transports only)

Variable · Default · Description
MCPG_AUTH_MODE · static · static (compare bearer to MCPG_HTTP_AUTH_TOKEN) \ · oidc (full JWT validation).
MCPG_HTTP_AUTH_TOKEN · — · Required bearer token when MCPG_AUTH_MODE=static. Constant-time compare. The HTTP transport refuses to start (ConfigError) unless this, MCPG_AUTH_MODE=oidc, or MCPG_HTTP_ALLOW_UNAUTHENTICATED=true is set.
MCPG_HTTP_ALLOW_UNAUTHENTICATED · false · Explicit opt-out of the HTTP transport's fail-closed auth check. Loudly logged on every startup when set; not recommended.
MCPG_OIDC_ISSUER · — · OIDC issuer URL (required when MCPG_AUTH_MODE=oidc).
MCPG_OIDC_AUDIENCE · — · Expected aud claim (required when MCPG_AUTH_MODE=oidc).
MCPG_OIDC_JWKS_URL · discovered · Override JWKS endpoint (auto-discovered from issuer's .well-known otherwise).
MCPG_OIDC_ROLE_CLAIM · — · JWT claim whose value becomes the per-request PG role (SET LOCAL ROLE). Composes with the tenancy driver.

HTTP hardening (HTTP transports only)

Variable Default Description
MCPG_HTTP_MAX_BODY_BYTES 1048576 (1 MiB) Request bodies above this get a 413. Counts streamed bytes, so a missing/lying Content-Length can't bypass it.
MCPG_HTTP_ALLOWED_ORIGINS — Comma-separated CORS allowlist. Unset = no CORS middleware (no cross-origin headers emitted).
MCPG_HTTP_HSTS_MAX_AGE 63072000 Strict-Transport-Security max-age (2 years, OWASP's current recommendation). 0 disables the HSTS header. Security headers (CSP, X-Frame-Options, X-Content-Type-Options, Referrer-Policy) are always added unless the app already set them.
MCPG_HTTP_REQUEST_TIMEOUT_SECONDS 0 Per-request wall-clock cap (504 on expiry). 0 = disabled. Leave off if you rely on long-lived SSE / streamable-http streams — a hard cap also severs those.
MCPG_HTTP_TRUSTED_HOSTS — Comma-separated list of allowed Host header values (wildcards like *.example.com supported). Unset = no host-header validation (current behaviour). When set, requests with a non-matching Host get a 400 via Starlette's TrustedHostMiddleware.

Multi-tenancy (SET ROLE)

Variable Default Description
MCPG_DEFAULT_ROLE — Static PG role applied to every query. Identifier-validated.
MCPG_ALLOWED_ROLES — Comma-separated allowlist. When set, the X-MCPG-Role header / OIDC role claim must be in this list.

Read replicas

Variable Default Description
MCPG_REPLICA_URLS — Comma-separated replica DSNs. force_readonly queries round-robin across healthy replicas; primary fallback on failure; 30 s degraded-replica retry window.

Multiple databases (read-only secondaries)

Variable Default Description
MCPG_SECONDARY_DATABASE_URLS — Comma- or newline-separated name=dsn entries naming additional read-only databases this one server can serve (e.g. analytics=postgresql://…?sslmode=require,reporting=postgresql://…?sslmode=require). Read-capable tools accept an optional database argument selecting a secondary by name; omit it for the primary. Secondaries are read-only — PostgreSQL-enforced (every query runs in a READ ONLY transaction), so writes / DDL / shell / migrate always target the primary. Names must be simple identifiers ([a-z0-9_]+), unique, and not primary (the reserved id of MCPG_DATABASE_URL). Same TLS rules as the primary DSN. Call list_databases to discover the configured ids and their reachability.

Pool / timeouts / TLS

Variable Default Description
MCPG_POOL_MIN_SIZE 1 Minimum pool connections.
MCPG_POOL_MAX_SIZE 5 Maximum pool connections. Must be ≥ MCPG_POOL_MIN_SIZE.
MCPG_STATEMENT_TIMEOUT_MS 30000 Per-session statement_timeout set on connection checkout. Runaway queries self-terminate.
MCPG_LOCK_TIMEOUT_MS 5000 Per-session lock_timeout. Hanging lock waits self-terminate.
MCPG_ENABLE_ANALYTICAL_QUERIES true Expose run_analytical_query (long-running reads on an isolated pool). Set false to withdraw the tool.
MCPG_ANALYTICAL_TIMEOUT_MS 120000 Default per-call budget for run_analytical_query (2 min).
MCPG_ANALYTICAL_MAX_TIMEOUT_MS 600000 Hard ceiling for run_analytical_query; a per-call timeout_ms is clamped to this (10 min). Must be ≥ MCPG_ANALYTICAL_TIMEOUT_MS.
MCPG_ANALYTICAL_MAX_CONCURRENCY 2 Size of the isolated analytical pool — max simultaneous run_analytical_query calls.
MCPG_ALLOW_INSECURE_TLS false Bypass the startup TLS check that refuses remote DSNs without sslmode=require (or stronger). Loopback hosts are always exempt.
MCPG_SHUTDOWN_DRAIN_SECONDS 30 On SIGTERM, wait up to this long for in-flight tool calls to finish before closing the pool and cursors.

Subprocess tools (MCPG_ALLOW_SHELL=true only)

Variable Default Description
MCPG_SHELL_TIMEOUT_SEC 60 Max wall-clock for pg_dump / pg_restore / psql invocations.
MCPG_SHELL_MAX_OUTPUT_BYTES 67108864 (64 MiB) Cap on captured stdout per subprocess call.
MCPG_SUBPROCESS_BIN_ALLOWLIST — Comma-separated absolute dirs the resolved pg_dump / pg_restore / psql must live under. Empty = trust PATH. Defeats a PATH-shim of these binaries.
MCPG_SUBPROCESS_CPU_SECONDS — Per-child RLIMIT_CPU (seconds). POSIX only; unset = inherit.
MCPG_SUBPROCESS_MEMORY_MB — Per-child RLIMIT_AS (MiB). POSIX only; unset = inherit.

LISTEN/NOTIFY (MCPG_ALLOW_LISTEN=true only)

Variable Default Description
MCPG_LISTEN_QUEUE_MAX 1000 Per-channel buffer; oldest notifications dropped on overflow.

Audit

Variable Default Description
MCPG_AUDIT_PERSIST false When true, every run_write / run_ddl call persists to a mcpg_audit.events table (auto-created idempotently).
MCPG_AUDIT_REDACT_KEYS — Comma-separated regex fragments added to the secret-name pattern (defaults already cover password, passwd, secret, token, api[_-]?key, bearer, authorization, database_url, dsn, conninfo).
MCPG_AUDIT_INTEGRITY false When true, each persisted event is signed with an HMAC chained over the previous event; the verify_audit_chain tool walks the chain and reports the first break. Requires MCPG_AUDIT_HMAC_KEY.
MCPG_AUDIT_HMAC_KEY — Secret key for the audit HMAC chain. Required when MCPG_AUDIT_INTEGRITY=true. Never appears in repr/logs.

Secrets backend

By default every secret is read straight from the environment. Set
MCPG_SECRETS_BACKEND=file to instead load API keys / bearer token /
HMAC key from a mounted file — a name in the file wins; anything absent
falls back to the env var, so partial files work.

Variable · Default · Description
MCPG_SECRETS_BACKEND · env · env (read every secret from the environment) \ · file (overlay a secrets file on top of the environment).
MCPG_SECRETS_FILE_PATH · — · Required when MCPG_SECRETS_BACKEND=file. Path to a flat name → value map: JSON always, or YAML (.yaml/.yml) when PyYAML is installed. Covers ANTHROPIC_API_KEY / OPENAI_API_KEY / GEMINI_API_KEY / GOOGLE_API_KEY / MCPG_NL2SQL_API_KEY, MCPG_HTTP_AUTH_TOKEN, and MCPG_AUDIT_HMAC_KEY.

Rate limiting

Variable Default Description
MCPG_RATE_LIMIT_ENABLED true Token-bucket per-tool rate limiting. Set to false to restore the pre-breaking-change unlimited behavior.
MCPG_RATE_LIMIT_MAX_REQUESTS 60 Global cap per window across all tools.
MCPG_RATE_LIMIT_WINDOW_SECONDS 60 Window length for the global quota.
MCPG_RATE_LIMIT_HEAVY_MAX 5 Cap for heavy tools (run_write, run_ddl, dump_database, etc.).
MCPG_RATE_LIMIT_HEAVY_WINDOW 60 Window length for the heavy-tool quota.

Caching & Feature flags

Variable Default Description
MCPG_CACHE_ENABLED true Enable or disable the adaptive cache layer.
MCPG_CACHE_TTL_SECONDS 300 Default cache Time-To-Live in seconds.
MCPG_CACHE_MAXSIZE 1024 Maximum LRU capacity bound for the memory cache.
MCPG_REDIS_URL — Optional Redis backend connection string for external, multi-node caching.
MCPG_ENABLE_HEAVY_DIAGNOSTICS true Toggle computationally heavy diagnostic, diagram, and advisor tools.
MCPG_ELICIT_CONFIRM_WRITES false When true, every write/DDL/shell/listen/migrate-tier tool call (any tool whose readOnlyHint annotation isn't true) requires an accepted interactive confirmation (ctx.elicit()) before running. Best-effort, not an enforcement boundary: it only engages for clients that both pass a request context and declare the elicitation capability during initialize — a client that omits either silently bypasses the gate and the tool runs as normal.

Natural-language SQL

MCPg auto-discovers every configured provider from the environment at
startup — set as many vendor keys as you have and each becomes callable.
Nineteen providers ship built in. Three are first-party (Anthropic,
OpenAI, Gemini); the other sixteen speak the OpenAI-compatible API with
vendor-preset endpoints: DeepSeek, Qwen, OpenRouter, Perplexity, xAI
(Grok), Groq, Mistral, Together, Fireworks, DeepInfra, Cerebras, Nebius,
Hugging Face, GitHub Models, SambaNova, and Moonshot (Kimi)
. Every
built-in is plug-and-play — set the vendor's conventional API-key env
var and it's auto-discovered — and any other OpenAI-compatible
vendor or local model server (Ollama, vLLM, LM Studio) is still
pluggable through configuration alone
via MCPG_NL2SQL_CUSTOM_PROVIDERS.
The whole built-in list is one declarative registry in nl2sql.py, so
adding a vendor or refreshing a retired default model is a one-line data
change.

When MCPG_NL2SQL_PROVIDER is unset, MCPg auto-picks the default in
registry order — anthropic → openai → gemini stay first so existing
deployments are unaffected. translate_nl_to_sql accepts an optional
provider="…" argument to route per call; get_server_info reports
which are configured.

Variable · Default · Description
_API_KEY · — · Setting a vendor's conventional key enables that provider. Standard slugs: ANTHROPIC_API_KEY, OPENAI_API_KEY, DEEPSEEK_API_KEY, OPENROUTER_API_KEY, PERPLEXITY_API_KEY, XAI_API_KEY, GROQ_API_KEY, MISTRAL_API_KEY, TOGETHER_API_KEY, FIREWORKS_API_KEY, CEREBRAS_API_KEY, NEBIUS_API_KEY, SAMBANOVA_API_KEY, MOONSHOT_API_KEY.
(keys that deviate) · — · A few vendors don't follow _API_KEY: Gemini → GEMINI_API_KEY or GOOGLE_API_KEY; Qwen → DASHSCOPE_API_KEY or QWEN_API_KEY; Hugging Face → HF_TOKEN; GitHub Models → GITHUB_TOKEN; DeepInfra → DEEPINFRA_TOKEN.
MCPG_NL2SQL_PROVIDER · auto-picked · Any built-in slug (listed above) or a custom name. Pins the default provider used when the tool is called without provider=. Unset + any vendor key present → MCPg auto-picks in registry order.
MCPG_NL2SQL_API_KEY · — · Explicit key for the configured MCPG_NL2SQL_PROVIDER. Overrides the vendor-conventional env var for that provider only. Requires MCPG_NL2SQL_PROVIDER to be set.
MCPG_NL2SQL_MODEL · provider default · Override the default model (e.g. claude-sonnet-4-6, gpt-4o-mini, grok-3-mini). Applies only to the default provider.
MCPG_NL2SQL_BASE_URL · — · Endpoint override for the default provider (private gateways / regional endpoints).
MCPG_NL2SQL_CUSTOM_PROVIDERS · — · Bring your own provider — no code change. Comma/newline-separated name=base_url\ · model entries declaring extra OpenAI-compatible providers beyond the built-ins (local Ollama / vLLM / LM Studio, or any niche vendor). Key from _API_KEY by convention, or append \ · KEY_ENV_VAR for ones that deviate; keyless allowed for loopback endpoints. Each name becomes callable via provider=.
MCPG_NL2SQL_MAX_TOKENS · 2048 · Cap on generated tokens (hard limit: 16384).


Usage examples

The MCP tools are invoked by the agent (Claude, Cursor, Continue, …)
in response to your natural-language instructions. A handful of
illustrative round-trips:

Inspect the schema

> You: What tables live in the public schema, and which ones are
> the biggest by row count?

> Agent (using list_tables + summarize_table × N):
> Six tables: customers (1.2M rows), orders (4.7M), line_items
> (18.3M), products (340), addresses (1.4M), audit_log
> (45.8M — the largest, accumulates without retention).

Diagnose a slow query

> You: Why does
> SELECT * FROM orders o JOIN line_items l ON o.id = l.order_id WHERE o.created_at > now() - interval '7 days'
> take 8 seconds?

> Agent (using why_is_this_slow + recommend_indexes):
> EXPLAIN ANALYZE shows a sequential scan over orders (4.7M rows)
> filtered by created_at. There's no index on orders.created_at.
> Recommended:
> CREATE INDEX CONCURRENTLY orders_created_at_idx ON orders (created_at DESC);
> Expected improvement: ~95% reduction (index scan touches ~33K rows
> instead of full table). Run validate_migration on it before applying
> to confirm there's no locking surprise.

Translate plain English to SQL

> You: Find customers who placed an order this month but never
> in the previous three months — show me their email and current
> lifetime spend.

> Agent (using translate_nl_to_sql):
>sql > SELECT c.email, SUM(l.price * l.quantity) AS lifetime_spend > FROM customers c > JOIN orders o ON o.customer_id = c.id > JOIN line_items l ON l.order_id = o.id > WHERE EXISTS ( > SELECT 1 FROM orders o2 WHERE o2.customer_id = c.id > AND o2.created_at >= date_trunc('month', now())) > AND NOT EXISTS ( > SELECT 1 FROM orders o3 WHERE o3.customer_id = c.id > AND o3.created_at >= date_trunc('month', now()) - interval '3 months' > AND o3.created_at < date_trunc('month', now())) > GROUP BY c.email; >

Picture the schema

> You: Draw me an ER diagram of the public schema.

> Agent (using generate_schema_diagram): Returns a Mermaid
> diagram you can paste straight into GitHub / Notion / Obsidian.

Audit the database

> You: How healthy is this database right now?

> Agent (using audit_database): Returns a graded report:
> Memory & I/O score 92 (GOOD), Transaction & Connection 78 (WARNING:
> rollback rate 0.4%, look at app logs), Concurrency & Locks 60
> (CRITICAL: 14 backends waiting), Cleanliness & Bloat 88 (GOOD), Slow
> queries 70 (WARNING: top query template runs 5000×, mean 90 ms —
> see optimize_query).

Run a guarded write

> You: Soft-delete every order older than 5 years.

> Agent (using run_write with MCPG_AUDIT_PERSIST=true): Validates
> the statement through the safe-SQL kernel, runs it inside a transaction,
> returns affected row count, persists the call (sql + arguments —
> with secrets regex-redacted — + status) to mcpg_audit.events for
> after-the-fact review.

For dozens more recipes — multi-tenant routing, RLS testing, NL→SQL,
hybrid vector + FTS search, Apache AGE Cypher, TimescaleDB, ORM schema
exports, server-side cursors — see docs/cookbook.md.


What's in the box

Compact category list. For the full, current tool reference see
docs/tools.md; for a guided walkthrough see
docs/tour.md.

  • Catalog introspection — schemas, tables, columns, indexes,
    constraints, views, functions, triggers, sequences, partitions,
    policies, roles, grants, enums, domains, composite types, FDWs,
    publications, subscriptions, extensions, generated columns.
  • Query intelligence — run_select, run_select_parallel,
    explain_query, analyze_query_plan, why_is_this_slow,
    recommend_indexes, analyze_workload, check_database_health,
    detect_n_plus_one, audit_database.
  • Search — fuzzy_search (trigram), full_text_search,
    vector_search, hybrid_search (pgvector + FTS via RRF),
    geo_search (PostGIS k-NN).
  • Natural language → SQL — translate_nl_to_sql (22 built-in
    providers — Anthropic, OpenAI, Gemini, xAI, Groq, Mistral, Hugging
    Face, … — plus any custom OpenAI-compatible endpoint; output passes
    through the same safe-SQL kernel as hand-written queries).
  • Visualisation — generate_schema_diagram (ER),
    generate_fk_cascade_graph (blast-radius of ON DELETE CASCADE),
    generate_graph_diagram (Apache AGE property graphs).
  • Structural diff & migrations — compare_schemas,
    validate_migration, staged prepare_migration /
    complete_migration / cancel_migration workflow.
  • Apache AGE graph + Cypher — list_graphs, describe_graph,
    run_cypher, create_graph, drop_graph, generate_graph_diagram.
  • Composite + advisor tools — summarize_table,
    find_unused_objects, find_sensitive_columns (PII heuristic),
    lint_naming_conventions, test_rls_for_role, list_locks,
    find_blocking_chains, read_pg_stat_io (PG16+),
    generate_test_data.
  • Live ops & maintenance — list_active_queries,
    verify_connection_encryption (TLS status of the live link),
    run_maintenance (VACUUM/ANALYZE), prune_audit_events
    (audit retention), cancel_query, terminate_backend,
    run_write, run_ddl, enable_extension.
  • Data movement — export_query / export_table (CSV/JSON),
    dump_database / restore_database, import_csv / import_json
    (COPY FROM STDIN), copy_table_between_databases.
  • Server-side cursors — open_cursor, fetch_cursor,
    close_cursor, list_cursors for pageable reads over millions
    of rows.
  • TimescaleDB — hypertable / chunk / job / continuous-aggregate
    introspection and stats, policy and dimension management, and gated
    chunk / job administration (drop_chunks, compress_chunk,
    refresh_continuous_aggregate, run_job, …). See
    docs/timescaledb.md.
  • ORM schema exporters — Prisma, Drizzle, SQLAlchemy, sqlc,
    Diesel, jOOQ, Ent, Ecto.
  • Event streams — subscribe_channel, poll_notifications,
    unsubscribe_channel, list_notification_subscriptions bridging
    PostgreSQL LISTEN/NOTIFY into the MCP poll model.
  • Observability — Prometheus /metrics endpoint +
    get_metrics_exposition tool for stdio; structured audit trail
    with regex-based credential redaction.

Documentation

  • docs/installation.md — install + configure
  • docs/tour.md — guided tool tour
  • docs/cookbook.md — practical agent recipes
  • docs/tools.md — complete tool reference
  • docs/architecture.md — how the pieces fit together
  • docs/scaling.md — pool sizing, replicas, performance
  • docs/security-hardening.md — security feature roadmap
  • docs/release-process.md — how releases ship to PyPI
  • docs/adr/ — architecture decision records
  • Browse at https://devopam.github.io/MCPg/

Security

  • Vulnerability reporting: see SECURITY.md. 90-day
    coordinated-disclosure window; reports to devopam@gmail.com.
  • Defence-in-depth: capability gates, SafeSQL kernel, identifier
    allowlist, audit redaction, PG TLS enforcement at startup,
    rate-limiting, OIDC JWT validation, per-session timeouts.
  • See docs/security-hardening.md for
    the living roadmap of shipped (✅) and queued (⬜) hardening items.

Privacy Policy

MCPg is self-hosted: your database contents never leave your

…(正文过长,页面只展示前面部分,完整内容见仓库)

常见问题

7 问
答案由本站依据条目字段整理,不引用上游原文
MCPg : Production Grade Postgresql Mcp Server 怎么安装?

资料没有给出安装命令,请到官方仓库或文档按说明接入;这类条目多数是生态应用,或尚未发布可安装包。

MCPg : Production Grade Postgresql Mcp Server 怎么在 Claude Desktop、Claude、Cursor 里配置?

它的传输方式未标注,在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,配置才会生效。

MCPg : Production Grade Postgresql Mcp Server 收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于服务本身和背后的接口额度,正式使用前按官方说明确认。

MCPg : Production Grade Postgresql Mcp Server 需要 API Key 吗?

资料标注需要会话登录认证,使用前准备好对应凭据,并按最小权限配置。

MCPg : Production Grade Postgresql Mcp Server 支持哪些客户端?

资料标注支持 Claude Desktop、Claude、Cursor、Gemini CLI、Codex、Zed;其他客户端只要能配置 MCP Server,也能用同样的命令或地址接入。

MCPg : Production Grade Postgresql Mcp Server 能提供哪些能力?

资料显示能力标签包括 mcp、server、postgresql。具体工具清单以官方文档为准。

MCPg : Production Grade Postgresql Mcp Server 需要什么运行环境?

运行时为 Cross-platform。

热门专题

按标签挑服务器
全部专题 →

服务器信息以作者仓库与官方文档为准;分类、传输方式与鉴权方式的中文说明为中文整理,最近核验 未标注。