20 Best PostgreSQL MCP Servers, Compared (September 2026)
The short answer
Reach for DBHub (bytebase/dbhub) for most PostgreSQL work. It is the official example in Claude Code docs, had a commit 0 days ago, and supports read-only mode, row limits, and query timeouts for safe exploration. If your job is production PostgreSQL tuning and maintenance, choose Postgres MCP (crystaldba/postgres-mcp) instead: its index analysis, health checks, and explain-plan tools show what to optimize. Postgres MCP also has a tool description grade of A.
Whichever you choose, give it the narrowest access that still works (a read-only credential, a replica, a scratch account), and widen it only once you have watched what your agent actually asks for.
Glama operates the MCP registry these numbers are measured from, and sells MCP hosting and a gateway. No position on this page is paid for. How the registry is built.
Quick picks
- 1Postgres MCP : Production PostgreSQL tuning and maintenance: its index analysis, health checks, and explain-plan tools show what to optimize.
- 2pg-aiguide : AI coding assistants that need version-aware PostgreSQL knowledge while writing schemas: it offers semantic search over the official PostgreSQL, TimescaleDB, and PostGIS documentation.
- 3pgEdge Postgres MCP Server : Internal tools and developer workflows: it gives LLMs read access to PostgreSQL and a natural language agent for forming SQL queries through MCP-compatible clients.
- 4butterbase : A team self-hosting an AI-native Postgres backend whose agents need to operate the entire backend: its MCP server exposes every capability above as tools.
- 5DBHub : PostgreSQL from Claude Code: it is the official example in Claude Code docs and supports read-only mode, row limits, and query timeouts for safe exploration.
Which one, for your situation
| Your situation | What to use |
|---|---|
| I need safe read-only exploration of a production PostgreSQL database. | Use MCP PostgreSQL Operations (call518/MCP-PostgreSQL-Ops); it is for production databases needing monitoring and maintenance insight without write access. |
| I want my AI coding assistant to use PostgreSQL in Claude Code. | Use DBHub (bytebase/dbhub); it is the official example in Claude Code docs and supports read-only mode, row limits, and query timeouts. |
| I need to tune a production PostgreSQL database. | Use Postgres MCP (crystaldba/postgres-mcp); its index analysis, health checks, and explain-plan tools show what to optimize. |
| I have to query and inspect an unfamiliar PostgreSQL database. | Use MCP PostgreSQL Server (antonorlov/mcp-postgres-server); it lists schemas and tables and describes columns before querying. |
Top MCP servers for PostgreSQL
| Best for | Profile | ||||||
|---|---|---|---|---|---|---|---|
| 1 | Production PostgreSQL tuning and maintenance: its index analysis, health checks, and explain-plan tools show what to optimize. | Community favourite | 3,286 | +108 | 30 days ago | 81.6 | |
| 2 | AI coding assistants that need version-aware PostgreSQL knowledge while writing schemas: it offers semantic search over the official PostgreSQL, TimescaleDB, and PostGIS documentation. | Community favourite | 1,835 | +27 | 6 days ago | 79.2 | |
| 3 | Internal tools and developer workflows: it gives LLMs read access to PostgreSQL and a natural language agent for forming SQL queries through MCP-compatible clients. | Steady | 223 | +5 | today | 71.3 | |
| 4 | A team self-hosting an AI-native Postgres backend whose agents need to operate the entire backend: its MCP server exposes every capability above as tools. | Community favourite | 3,487 | +393 | today | 70.2 | |
| 5 | PostgreSQL from Claude Code: it is the official example in Claude Code docs and supports read-only mode, row limits, and query timeouts for safe exploration. | Community favourite | 3,493 | +168 | today | 70.0 | |
| 6 | Production environments where you want guardrails: it defaults to readonly mode and requires explicit security flags for any write or admin operation. | Community favourite | 199 | +1 | 83 days ago | 68.4 | |
| 7 | Production PostgreSQL databases needing monitoring and maintenance insight without write access: it exposes 34 read-only tools for locks, WAL, replication, bloat, vacuum, and I/O. | Steady | 161 | +2 | yesterday | 67.3 | |
| 8 | AI-agent-assisted Prisma 8 development on PostgreSQL: it installs agent skills that load the ORM contract and drive schema/query changes. | Community favourite | 47,603 | +67 | today | 67.1 | |
| 9 | VS Code users with existing DBCode database connections who want AI agents to query those databases: its MCP server exposes schemas and executes queries. | Community favourite | 347 | +17 | today | 59.6 | |
| 10 | A developer who needs one MCP server for many database types: it offers natural language queries and database management via MCP. | Community favourite | 19,200 | +4,291 | today | 57.5 | |
| 11 | AWS development in Claude Code or Cursor: its 22 tools surface missing indexes, DLQs, Lambda config issues, IaC drift, and security findings. | Community favourite | 21 | +1 | today | 55.5 | |
| 12 | PostgreSQL teams on Google Cloud that need schema discovery, SQL execution, and custom restricted tools: the README documents list_tables, execute_sql, and the custom tool framework. | Community favourite | 16,371 | +200 | today | 54.5 | |
| 13 | For AI coding agents that switch between clients and need persistent memory: it stores shared context in PostgreSQL and retrieves it with hybrid search. | Steady | 115 | +3 | 20 days ago | 53.3 | |
| 14 | For a developer using PostgreSQL from an AI assistant: it exposes schema discovery, query, EXPLAIN, diff, and write-confirmation tools. | Community favourite | 5,024 | +19 | today | 52.6 | |
| 15 | Ad-hoc natural-language querying on production PostgreSQL: its default read-only mode prevents accidental writes. That is a concrete situation and the specific capability. | Abandoned but popular | 929 | +20 | 181 days ago | 52.2 | |
| 16 | A production PostgreSQL database you need to tune without guessing: it explains slow queries, spots bloat and unused indexes, and tests hypothetical indexes first. | Dormant | 29 | +4 | 104 days ago | 52.2 | |
| 17 | For quick SQL querying and schema inspection of a PostgreSQL database through MCP: it documents a query tool plus table and schema resources. | Abandoned but popular | 180 | -1 | 315 days ago | 50.8 | |
| 18 | For teams wanting agents to maintain a permissioned, versioned knowledge base in PostgreSQL: it exposes 50 tools spanning hybrid search, SQL tables, and Git-backed history. | Emerging | 160 | +11 | today | 50.5 | |
| 19 | Read-only exploration of an unfamiliar PostgreSQL database: it lists schemas and tables and describes columns before querying. | Community favourite | 19 | +3 | today | 50.3 | |
| 20 | PostgreSQL teams that want governed, auditable AI database access: GraphJin compiles natural language to SQL, validates queries, and logs every answer. | Community favourite | 3,165 | +29 | 3 days ago | 50.0 |
The ranking, with the evidence
Each position is a weighted mean of adoption (40%), maintenance (24%), momentum (14%), tool description quality (13%) and trust (9%), multiplied by three attenuators: how directly the server is about PostgreSQL (named for it, declaring it, tagged with it, or merely mentioning it), whether its repository is still moving, and how much independent evidence of adoption it has. Open the score on any entry to see every number, including the ones marked ≈, which were imputed from the median of the other candidates rather than measured. The maintenance grade on each entry is mostly issue responsiveness, release recency and open security alerts rather than commits, so a recent commit beside a low grade is two different measurements rather than a contradiction.
- Abandoned but popular: People use it, but its default branch has stopped moving. Fine to keep running, risky to adopt.
- Community favourite: Widely adopted and still actively maintained.
- Dormant: Neither changing nor widely adopted. Here because it still matches the search.
- Emerging: Small audience, growing quickly, maintained. The bet with the most upside.
- Steady: Maintained, modest audience, no surprises in either direction.
Best for: Production PostgreSQL tuning and maintenance: its index analysis, health checks, and explain-plan tools show what to optimize.
The server exposes nine tools that cover schema inspection, query plans, index recommendations, database health checks, top-query reporting, and SQL execution. It requires database credentials and either Docker or Python 3.12 or higher.
GitHub stars3,286Stars / 30 days+108npm / typical weekShips no npm packagePyPI / typical weekno attributed packageTools exposed9Last commit30 days agoCommits / 12 weeks1Maintenance gradeDTool descriptionsAScore 81.6: show every number behind it
- Adoption88 / 100 · weight 40%
- GitHub stars88
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance71 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade30
- Commit cadence40
- Momentum77 / 100 · weight 14%
- Stars gained, relative to size73
- Stars gained, absolute82
- npm download trendnot measuredno download history for the selected registry
- Tool quality76 / 100 · weight 13%
- Tool description quality68
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 81.6
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 81.6
Best for: AI coding assistants that need version-aware PostgreSQL knowledge while writing schemas: it offers semantic search over the official PostgreSQL, TimescaleDB, and PostGIS documentation.
The README documents a public MCP server at mcp.tigerdata.com/docs for semantic search across PostgreSQL, TimescaleDB, and PostGIS documentation, and also offers installation as Agent Skills via npx or as a Claude Code plugin. It is a hosted endpoint, so the main requirement is network access to the remote server rather than a local PostgreSQL connection.
GitHub stars1,835Stars / 30 days+27npm / typical week49PyPI / typical weekno attributed packageTools exposednever inspectedLast commit6 days agoCommits / 12 weeks20Maintenance gradeBTool descriptionsNot gradedScore 79.2: show every number behind it
- Adoption84 / 100 · weight 40%
- GitHub stars82
- npm downloads18downloads show none of the weekday rhythm human traffic has; halved
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance92 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade80
- Commit cadence85
- Momentum40 / 100 · weight 14%
- Stars gained, relative to size48
- Stars gained, absolute59
- npm download trend5
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integrates100
- Weighted mean of the five
- 79.2
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 79.2
Best for: Internal tools and developer workflows: it gives LLMs read access to PostgreSQL and a natural language agent for forming SQL queries through MCP-compatible clients.
The server supports MCP clients over stdio or HTTP, adds a natural language agent that forms SQL from natural language, and documents MCP tools, resources, prompts, authentication, and configuration. The scope to know before choosing: it is not for public-facing applications, because it gives LLMs read access to the entire database schema and data.
GitHub stars223Stars / 30 days+5npm / typical weekShips no npm packagePyPI / typical weekno attributed packageTools exposednever inspectedLast committodayCommits / 12 weeks122Maintenance gradeATool descriptionsNot gradedScore 71.3: show every number behind it
- Adoption59 / 100 · weight 40%
- GitHub stars59
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum43 / 100 · weight 14%
- Stars gained, relative to size50
- Stars gained, absolute33
- npm download trendnot measuredno download history for the selected registry
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integrates100
- Weighted mean of the five
- 71.3
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 71.3
Best for: A team self-hosting an AI-native Postgres backend whose agents need to operate the entire backend: its MCP server exposes every capability above as tools.
Butterbase is a self-hostable backend-as-a-service whose built-in MCP server presents Postgres, RLS, auth, storage, serverless functions, durable objects, realtime, KV, AI gateway, and RAG as tools over HTTP at /mcp or stdio via npx @butterbase/mcp. Before choosing it, note that no npm package is published, so the README's npx stdio command has no published package behind it, and self-hosting requires Docker, Node 22+, and a clone with submodules.
GitHub stars3,487Stars / 30 days+393npm / typical weekShips no npm packagePyPI / typical weekno attributed packageTools exposednever inspectedLast committodayCommits / 12 weeks224Maintenance gradeBTool descriptionsNot gradedScore 70.2: show every number behind it
- Adoption89 / 100 · weight 40%
- GitHub stars89
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance95 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade80
- Commit cadence100
- Momentum100 / 100 · weight 14%
- Stars gained, relative to size100
- Stars gained, absolute100
- npm download trendnot measuredno download history for the selected registry
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integrates100
- Weighted mean of the five
- 90.0
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 70.2
Best for: PostgreSQL from Claude Code: it is the official example in Claude Code docs and supports read-only mode, row limits, and query timeouts for safe exploration.
DBHub is an MCP server connecting MCP clients to PostgreSQL, MySQL, MariaDB, SQL Server, and SQLite, and it documents tools for executing SQL, searching objects, explaining plans, checking health, and custom SQL operations. Before installing, note that it requires Node.js 22.5.0 or newer because it uses the built-in node:sqlite module.
GitHub stars3,493Stars / 30 days+168npm / typical weekdownloads not countedPyPI / typical weekno attributed packageTools exposednever inspectedLast committodayCommits / 12 weeks77Maintenance gradeATool descriptionsNot gradedScore 70.0: show every number behind it
- Adoption89 / 100 · weight 40%
- GitHub stars89
- npm downloadsnot measurednpm names no repository for dbhub, so its downloads cannot be attributed
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum90 / 100 · weight 14%
- Stars gained, relative to size90
- Stars gained, absolute90
- npm download trendnot measuredno download history for the selected registry
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 89.7
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 70.0
Best for: Production environments where you want guardrails: it defaults to readonly mode and requires explicit security flags for any write or admin operation.
It exposes 18 tools covering PostgreSQL analysis, schema management, queries, mutations, and monitoring, with a security-mode system documented in the README. The server starts in readonly mode, so write, DDL, or arbitrary SQL require passing --security-mode and --allow-destructive flags.
GitHub stars199Stars / 30 days+1npm / typical week2.2KPyPI / typical weekno attributed packageTools exposed18Last commit83 days agoCommits / 12 weeks11Maintenance gradeDTool descriptionsBScore 68.4: show every number behind it
- Adoption80 / 100 · weight 40%
- GitHub stars58
- npm downloads71
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance73 / 100 · weight 24%
- Last commit touching this server88
- Repository maintenance grade30
- Commit cadence85
- Momentum13 / 100 · weight 14%
- Stars gained, relative to size21
- Stars gained, absolute12
- npm download trend0
- Tool quality63 / 100 · weight 13%
- Tool description quality55
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 68.4
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 68.4
Best for: Production PostgreSQL databases needing monitoring and maintenance insight without write access: it exposes 34 read-only tools for locks, WAL, replication, bloat, vacuum, and I/O.
It exposes 34 read-only tools covering server info, schemas, relationships, locks, WAL, replication, connections, query statistics, sizes, configuration, indexes, vacuum and autovacuum, bloat, I/O, background writer, and user-defined functions. Before choosing it, note that it depends on a configured PostgreSQL connection and that advanced query analysis is enhanced when pg_stat_statements and optionally pg_stat_monitor are installed.
GitHub stars161Stars / 30 days+2npm / typical weekShips no npm packagePyPI / typical weekno attributed packageTools exposed34Last commityesterdayCommits / 12 weeks18Maintenance gradeBTool descriptionsAScore 67.3: show every number behind it
- Adoption55 / 100 · weight 40%
- GitHub stars55
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance92 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade80
- Commit cadence85
- Momentum26 / 100 · weight 14%
- Stars gained, relative to size31
- Stars gained, absolute18
- npm download trendnot measuredno download history for the selected registry
- Tool quality81 / 100 · weight 13%
- Tool description quality73
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 67.3
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 67.3
Best for: AI-agent-assisted Prisma 8 development on PostgreSQL: it installs agent skills that load the ORM contract and drive schema/query changes.
The README documents Prisma 8 as a TypeScript rewrite of the ORM and covers agent skills for Claude Code, Cursor, Copilot Agent, and Devin. The one thing worth knowing before choosing it is that Prisma 8 is currently a release candidate and requires Node.js 24 or newer.
GitHub stars47,603Stars / 30 days+67npm / typical weekdownloads not countedPyPI / typical weekno attributed packageTools exposednever inspectedLast committodayCommits / 12 weeks357Maintenance gradeBTool descriptionsNot gradedScore 67.1: show every number behind it
- Adoption100 / 100 · weight 40%
- GitHub stars100
- npm downloadsnot measurednpm names no repository for dev-env, so its downloads cannot be attributed
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance95 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade80
- Commit cadence100
- Momentum39 / 100 · weight 14%
- Stars gained, relative to size15
- Stars gained, absolute74
- npm download trendnot measuredno download history for the selected registry
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integrates100
- Weighted mean of the five
- 86.0
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 67.1
Best for: VS Code users with existing DBCode database connections who want AI agents to query those databases: its MCP server exposes schemas and executes queries.
DBCode is a VS Code extension whose built-in MCP server gives AI agents access to databases, schemas, and query execution over connections already configured in the extension. Before choosing it, know that it depends on the DBCode extension and those existing connections, rather than managing its own connections.
GitHub stars347Stars / 30 days+17npm / typical weekdownloads not countedPyPI / typical weekno attributed packageTools exposednever inspectedLast committodayCommits / 12 weeks71Maintenance gradeATool descriptionsNot gradedScore 59.6: show every number behind it
- Adoption64 / 100 · weight 40%
- GitHub stars64
- npm downloadsnot measurednpm names no repository for dbcode, so its downloads cannot be attributed
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum66 / 100 · weight 14%
- Stars gained, relative to size76
- Stars gained, absolute51
- npm download trendnot measuredno download history for the selected registry
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 76.4
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 59.6
Best for: A developer who needs one MCP server for many database types: it offers natural language queries and database management via MCP.
dbx is the MCP server for the DBX database manager, and its README lists desktop, Docker, CLI, built-in AI assistant, and MCP server as the interfaces. The MCP tool list has not been inspected, so confirm which tools it exposes before relying on it.
GitHub stars19,200Stars / 30 days+4,291npm / typical weekdownloads not countedPyPI / typical weekno attributed packageTools exposednever inspectedLast committodayCommits / 12 weeks4,210Maintenance gradeATool descriptionsNot gradedScore 57.5: show every number behind it
- Adoption100 / 100 · weight 40%
- GitHub stars100
- npm downloadsnot measurednpm publishes dbx from samt/node-dbx
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum100 / 100 · weight 14%
- Stars gained, relative to size100
- Stars gained, absolute100
- npm download trendnot measuredno download history for the selected registry
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 95.8
- × relevance: the keyword is tagged here
- 0.60
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 57.5
Best for: AWS development in Claude Code or Cursor: its 22 tools surface missing indexes, DLQs, Lambda config issues, IaC drift, and security findings.
Infrawise statically analyzes a codebase, cloud infrastructure, and database schemas and exposes the results as 22 MCP tools, including get_infra_overview, get_graph_summary, and postgres_index_suggestions. It requires Node.js 22 or later and is scoped to AWS infrastructure analysis, not provisioning or observability.
GitHub stars21Stars / 30 days+1npm / typical week406PyPI / typical weekno attributed packageTools exposed22Last committodayCommits / 12 weeks231Maintenance gradeATool descriptionsAScore 55.5: show every number behind it
- Adoption61 / 100 · weight 40%
- GitHub stars34
- npm downloads56
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum20 / 100 · weight 14%
- Stars gained, relative to size36
- Stars gained, absolute15
- npm download trend0
- Tool quality85 / 100 · weight 13%
- Tool description quality77
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 71.2
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 55.5
Best for: PostgreSQL teams on Google Cloud that need schema discovery, SQL execution, and custom restricted tools: the README documents list_tables, execute_sql, and the custom tool framework.
The server is an MCP server for databases that documents prebuilt generic tools such as list_tables and execute_sql, plus a framework for building custom tools with restricted access. The README's install command uses npx @toolbox-sdk/server, and no npm package is recorded as published, so that command needs verification before an install.
GitHub stars16,371Stars / 30 days+200npm / typical weekShips no npm packagePyPI / typical weekno attributed packageTools exposednever inspectedLast committodayCommits / 12 weeks238Maintenance gradeATool descriptionsNot gradedScore 54.5: show every number behind it
- Adoption100 / 100 · weight 40%
- GitHub stars100
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum64 / 100 · weight 14%
- Stars gained, relative to size45
- Stars gained, absolute93
- npm download trendnot measuredno download history for the selected registry
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 90.8
- × relevance: the keyword is tagged here
- 0.60
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 54.5
Best for: For AI coding agents that switch between clients and need persistent memory: it stores shared context in PostgreSQL and retrieves it with hybrid search.
It is a retrieval engine for shared memory that runs as an MCP server, documented with an init/health/search/serve CLI, a setup wizard, and stdio, streamable HTTP, and SSE transports. Note that it assumes a Postgres database (Supabase or Neon/self-hosted) and an embedding provider with an API key.
GitHub stars115Stars / 30 days+3npm / typical weekShips no npm packagePyPI / typical week63ogham-mcpTools exposednever inspectedLast commit20 days agoCommits / 12 weeks42Maintenance gradeATool descriptionsNot gradedScore 53.3: show every number behind it
- Adoption57 / 100 · weight 40%
- GitHub stars52
- PyPI downloads38
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum26 / 100 · weight 14%
- Stars gained, relative to size42
- Stars gained, absolute23
- PyPI download trend0
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 68.3
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 53.3
Best for: For a developer using PostgreSQL from an AI assistant: it exposes schema discovery, query, EXPLAIN, diff, and write-confirmation tools.
It exposes 12 MCP tools covering connections, schemas, tables, columns, queries, EXPLAIN, diffs, ERD metadata, audit checks, and write confirmations. The one thing worth knowing before choosing it: it ships as a CLI and Docker image, not as an npm or PyPI package.
GitHub stars5,024Stars / 30 days+19npm / typical weekShips no npm packagePyPI / typical weekno attributed packageTools exposed12Last committodayCommits / 12 weeks550Maintenance gradeATool descriptionsAScore 52.6: show every number behind it
- Adoption93 / 100 · weight 40%
- GitHub stars93
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum36 / 100 · weight 14%
- Stars gained, relative to size24
- Stars gained, absolute52
- npm download trendnot measuredno download history for the selected registry
- Tool quality98 / 100 · weight 13%
- Tool description quality90
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integrates100
- Weighted mean of the five
- 87.7
- × relevance: the keyword is tagged here
- 0.60
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 52.6
Best for: Ad-hoc natural-language querying on production PostgreSQL: its default read-only mode prevents accidental writes. That is a concrete situation and the specific capability.
Universal DB MCP exposes MCP stdio, SSE, Streamable HTTP, and REST interfaces for querying and schema analysis, with PostgreSQL among its supported databases. It requires Node.js 20 or higher, and while it defaults to read-only, writes become possible if that mode is disabled.
GitHub stars929Stars / 30 days+20npm / typical week126PyPI / typical weekno attributed packageTools exposednever inspectedLast commit181 days agoCommits / 12 weeks0Maintenance gradeDTool descriptionsNot gradedScore 52.2: show every number behind it
- Adoption81 / 100 · weight 40%
- GitHub stars74
- npm downloads45
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance46 / 100 · weight 24%
- Last commit touching this server68
- Repository maintenance grade30
- Commit cadence5
- Momentum41 / 100 · weight 14%
- Stars gained, relative to size56
- Stars gained, absolute53
- npm download trend0
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 66.9
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 52.2
Best for: A production PostgreSQL database you need to tune without guessing: it explains slow queries, spots bloat and unused indexes, and tests hypothetical indexes first.
It connects to a PostgreSQL database to read pg_stat_statements, run EXPLAIN ANALYZE, surface table and index bloat, and test hypothetical indexes via HypoPG before they are created. The database user needs read access to system catalogs, and some tools also require the pgstattuple and HypoPG extensions.
GitHub stars29Stars / 30 days+4npm / typical weekShips no npm packagePyPI / typical week125pgtuner-mcpTools exposed14Last commit104 days agoCommits / 12 weeks0Maintenance gradeDTool descriptionsAScore 52.2: show every number behind it
- Adoption50 / 100 · weight 40%
- GitHub stars37
- PyPI downloads45
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance57 / 100 · weight 24%
- Last commit touching this server88
- Repository maintenance grade30
- Commit cadence5
- Momentum47 / 100 · weight 14%
- Stars gained, relative to size60
- Stars gained, absolute27
- PyPI download trendnot measuredno download history for the selected registry
- Tool quality93 / 100 · weight 13%
- Tool description quality85
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 61.4
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: modest but real audience
- 0.85
- Composite score
- 52.2
Best for: For quick SQL querying and schema inspection of a PostgreSQL database through MCP: it documents a query tool plus table and schema resources.
It executes SQL queries through a query tool and exposes table and schema resources over stdio or HTTP. Write operations are disabled unless the DANGEROUSLY_ALLOW_WRITE_OPS environment variable is explicitly enabled.
GitHub stars180Stars / 30 days-1npm / typical week189PyPI / typical weekno attributed packageTools exposednever inspectedLast commit315 days agoCommits / 12 weeks0Maintenance gradeDTool descriptionsNot gradedScore 50.8: show every number behind it
- Adoption64 / 100 · weight 40%
- GitHub stars56
- npm downloads48
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance32 / 100 · weight 24%
- Last commit touching this server42dated from the last commit on the default branch, re-read from GitHub at publication; github.com shows a push 26 days ago, which counts every ref; the stored date would have published 26 days ago
- Repository maintenance grade30
- Commit cadence5
- Momentum0 / 100 · weight 14%
- Stars gained, relative to size0
- Stars gained, absolute0
- npm download trend0
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 50.8
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 50.8
Best for: For teams wanting agents to maintain a permissioned, versioned knowledge base in PostgreSQL: it exposes 50 tools spanning hybrid search, SQL tables, and Git-backed history.
AKB exposes 50 tools for vault-scoped knowledge management: writing and updating docs, creating and querying tables, hybrid search, relations, and Git history. The README points real users to self-host with Docker Compose or Kubernetes and warns that the public demo is ephemeral.
GitHub stars160Stars / 30 days+11npm / typical weekShips no npm packagePyPI / typical weekno attributed packageTools exposed50Last committodayCommits / 12 weeks568Maintenance gradeATool descriptionsAScore 50.5: show every number behind it
- Adoption55 / 100 · weight 40%
- GitHub stars55
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum65 / 100 · weight 14%
- Stars gained, relative to size79
- Stars gained, absolute44
- npm download trendnot measuredno download history for the selected registry
- Tool quality73 / 100 · weight 13%
- Tool description quality65
- Built and inspected by Glama100
- Trust0 / 100 · weight 9%
- License0
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 64.7
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 50.5
Best for: Read-only exploration of an unfamiliar PostgreSQL database: it lists schemas and tables and describes columns before querying.
It exposes five tools: query, execute, list_schemas, list_tables, and describe_table, and connects via a DATABASE_URL or PG_* variables. The execute tool is disabled by default: the server is read-only until PG_ALLOW_WRITE=true, and Node.js 20 or newer is required.
GitHub stars19Stars / 30 days+3npm / typical week620PyPI / typical weekno attributed packageTools exposed5Last committodayCommits / 12 weeks8Maintenance gradeBTool descriptionsAScore 50.3: show every number behind it
- Adoption62 / 100 · weight 40%
- GitHub stars33
- npm downloads58downloads capped: not corroborated by repository audience
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance88 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade80
- Commit cadence65
- Momentum31 / 100 · weight 14%
- Stars gained, relative to size53
- Stars gained, absolute23
- npm download trend0
- Tool quality96 / 100 · weight 13%
- Tool description quality88
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 71.9
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: small audience
- 0.70
- Composite score
- 50.3
Best for: PostgreSQL teams that want governed, auditable AI database access: GraphJin compiles natural language to SQL, validates queries, and logs every answer.
GraphJin exposes an MCP endpoint, a GraphQL API, REST endpoints, workflows, and a built-in agent that returns typed, evidence-backed answers. Before choosing it, know that the built-in agent mode requires a model API key in ./.env; without one, that agent is not enabled.
GitHub stars3,165Stars / 30 days+29npm / typical week518PyPI / typical weekno attributed packageTools exposednever inspectedLast commit3 days agoCommits / 12 weeks502Maintenance gradeATool descriptionsNot gradedScore 50.0: show every number behind it
- Adoption92 / 100 · weight 40%
- GitHub stars88
- npm downloads29downloads show none of the weekday rhythm human traffic has; halved
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum35 / 100 · weight 14%
- Stars gained, relative to size38
- Stars gained, absolute59
- npm download trend0
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integrates100
- Weighted mean of the five
- 83.3
- × relevance: the keyword is tagged here
- 0.60
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 50.0
Questions people ask
How do I choose between DBHub and Postgres MCP (crystaldba/postgres-mcp)?
DBHub is for safe exploration from an AI client: it is the official example in Claude Code docs and supports read-only mode, row limits, and query timeouts. Postgres MCP is for production tuning, with index analysis, health checks, and explain-plan tools. Postgres MCP had a commit 30 days ago and 1 commit in the last 12 weeks, while DBHub had a commit 0 days ago and 77 commits in the last 12 weeks.
Which server has the strongest guardrails for production use?
PostgreSQL MCP Server (HenkDz/postgresql-mcp-server) defaults to readonly mode and requires explicit security flags for any write or admin operation. It is labelled Community favourite and its last commit was 83 days ago. It had 11 commits in the last 12 weeks.
Which one is best for read-only monitoring of production PostgreSQL?
MCP PostgreSQL Operations (call518/MCP-PostgreSQL-Ops) is aimed at production databases needing monitoring and maintenance insight without write access. It exposes read-only tools covering locks, WAL, replication, bloat, vacuum, and I/O. It had a commit 1 day ago and 18 commits in the last 12 weeks.
Is Universal DB MCP still worth installing?
It is labelled Abandoned but popular, with a last commit 181 days ago and 0 commits in the last 12 weeks. Its default read-only mode is a stated feature, so it can still serve ad-hoc natural-language queries safely. For an actively maintained alternative, use DBHub (bytebase/dbhub) or MCP PostgreSQL Server (antonorlov/mcp-postgres-server).
What does Postgres MCP not cover if I use it for querying?
Postgres MCP (crystaldba/postgres-mcp) is described for production tuning and maintenance, not as a general multi-database client. Its focus is index analysis, health checks, and explain-plan tools. If you need to query many database types, use DBHub (bytebase/dbhub) or dbx (t8y2/dbx) instead.