M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
P

pg-aiguide

Community
AI-optimized PostgreSQL expertise for coding assistants
How to install ↓ Compare GitHub source repository ↗
Category
Database #38 of 58
Stars
★ 1.9k Very popular
Transport
Streamable HTTP
Runtime
Hosted (no local runtime)
Credentials
No credential needed
License
Apache-2.0
Last commit
Tools
2
50FMRS · D

pg-aiguide significantly improves the quality of PostgreSQL code generated by AI coding tools by providing semantic search and best-practice skills. It is suitable for developers needing high-quality database code, but requires network access and relies on external services.

Strongest · Maintenance 12/20 Weakest · Reliability 8/20

Reliability
8/20
Security and permissions
9/20
Maintenance
12/20
Documentation
10/20
Setup experience
11/20
Why each score
Reliability 8/20
No server manifest or explicit local run evidence; tools declared as search_docs and view_skill but behavior not verified. Error handling and dependency control are unclear, so reliability is limited to a plausible happy path.
Security and permissions 9/20
Exposed as remote HTTP endpoint with no documented auth or authorization, potential unrestricted access. No malicious behavior or real secrets found. Lacks least-privilege and confirmation details, so score is moderate-low.
Maintenance 12/20
Repository is active with many stars and recent commits, but lacks release versions and changelog. License is Apache-2.0, no security response channel visible. Score reflects activity with governance gaps.
Documentation 10/20
README offers detailed install and usage examples for multiple clients, but lacks detailed API parameters, limits, and troubleshooting. Documentation is usable but with hidden assumptions.
Setup experience 11/20
Provides client config examples for many platforms, but relies on remote hosted service; local setup steps unclear. No execution evidence, so score is moderate-high.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

  • Developers using AI coding tools (Claude Code, Cursor, Codex, Gemini CLI).
  • Those wanting to ensure generated PostgreSQL code is modern, reliable, and performant.

Not for

  • Applications requiring direct database connections or query execution.
  • Environments needing local, offline operation.

Required permissions

  • As a remote HTTP server, it receives queries and returns documentation.
  • As skills, it may be invoked automatically by AI agents.

Risks and side effects

  • Public server may contain outdated or inaccurate information from docs.
  • Remote server exposed to network risks.
  • AI agents may generate plausible but unvalidated code.

Setup

Before you start

Runtime：Hosted (no local runtime)

Install as Agent Skills: run npx skills add timescale/pg-aiguide --skill postgres. As an MCP server, add https://mcp.tigerdata.com/docs to your client config (e.g., mcpServers JSON). As a Claude Code plugin: run claude plugin marketplace add timescale/pg-aiguide and claude plugin install pg@aiguide.

.mcp.json 
{
  "mcpServers": {
    "pg-aiguide": {
      "url": "https://mcp.tigerdata.com/docs"
    }
  }
}

Shown for Claude Code. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "pg-aiguide": {
      "url": "https://mcp.tigerdata.com/docs"
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add --transport http pg-aiguide https://mcp.tigerdata.com/docs

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

Check that the client's tool list shows search_docs and view_skill, then ask the agent to run search_docs on a PostgreSQL topic to confirm the connection works.

Troubleshooting

  1. Ensure network connectivity to `mcp.tigerdata.com`.
  2. Check the URL in your client's MCP config is correct.
  3. For skills, verify `npx skills` is installed and version compatible.
  4. For Claude Code plugin, ensure you have a supported Claude version.

Things to try

Once connected, you can ask your AI assistant things like:

  • Create a Postgres table schema for storing usernames and unique email addresses.
  • Search the official PostgreSQL manual for partial index documentation.
  • Show me the best-practice skill on indexing strategies.
  • Generate a Postgres schema for an IoT device company collecting temperature, humidity, and pressure data.

Tools 2

search_docs read-only
Unified search tool supporting semantic (vector similarity) and keyword (BM25) search across multiple documentation sources, including PostgreSQL, TimescaleDB, and PostGIS.
view_skill read-only
Exposes curated, opinionated PostgreSQL best-practice skills covering schema design, indexing strategies, data types, data integrity, naming conventions, performance tuning, and modern features.

Use cases

Provide deep PostgreSQL knowledge to AI coding agents for better schema generation.
Search PostgreSQL-related documentation across multiple sources.
Automate best practice application via skills.

Supported clients

Claude Code
Cursor
VS Code
Codex
Gemini CLI
Windsurf

Listed from the project's documentation, not tested by this site.

Overview

pg-aiguide helps AI coding tools write dramatically better PostgreSQL code. It provides semantic search across the official PostgreSQL manual (version-aware), AI-optimized 'skills' (curated, opinionated Postgres best practices used automatically by AI agents), and extension ecosystem docs (starting with TimescaleDB). It can be used as Agent Skills, a public MCP server, or a Claude Code plugin.

Similar servers

Compare with other MCP servers

Source revision 187be00d317a Data synced 2026-10-19 Read the FMRS scoring method