PostgreSQL Performance (read-only)

io.github.gouranshulv0.1.0更新于 Oct 10, 2026

Read-only PostgreSQL performance diagnostics: slow queries, EXPLAIN, index advice, locks, functions

已验证STDIO仅桌面Databases Data & Analytics Security & Monitoring

概览

AI 生成的概览

只读的 PostgreSQL 性能诊断：慢查询、EXPLAIN 计划、索引建议、表健康、锁与函数。

功能: 通过一个受限的只读接口，让助手从 pg_stat_statements、EXPLAIN 和 PostgreSQL 系统目录中获取真实数据。工具包括 top_slow_queries、explain_query、suggest_indexes、unused_indexes、table_health、blocking_sessions、slow_functions 和 explain_function，另有 schema 资源和引导式 diagnose_slow_database 提示。它只建议索引，从不创建索引，也不会取消或终止会话。
适用场景: 当数据库变慢，希望助手查看真实执行计划、表大小、现有索引和查询负载而不是凭空猜测时使用。适合在副本或预发布数据库上诊断慢查询、索引候选、表膨胀和锁争用。
运行要求: 运行已发布镜像需要 Docker，从源码构建需要 Java 25。需要一个 PostgreSQL 数据库，在 shared_preload_libraries 中启用 pg_stat_statements，并有一个具备 pg_read_all_stats 和 pg_monitor 的只读角色。需设置 PGPERF_DB_URL、PGPERF_DB_USER、PGPERF_DB_PASSWORD 和 MCP_API_KEY；客户端以 Authorization bearer 头发送该密钥。函数相关工具还需启用 track_functions 和 auto_explain。

安装前请注意

服务器持有数据库凭据（PGPERF_DB_PASSWORD）和 API 密钥（MCP_API_KEY，以 Authorization bearer 令牌发送），应避免写入 shell 历史。analyze=true 和 explain_function 会实际执行查询，建议使用副本或预发布数据库。索引建议和未使用索引的删除语句只是建议，必须人工审核后自行执行。

安装

在 SourceWeft 中

打开控制台中的 PostgreSQL Performance (read-only)，将其添加到工作区。
为需要使用其工具的对话启用该服务。

Desktop only，通过 STDIO。 STDIO 服务会启动本地进程，因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照仓库中的启动说明。

README

pg-perf-mcp

[CI] [License: MIT] [Java 25] [Spring Boot 4.1]

An MCP server that lets AI assistants diagnose PostgreSQL performance problems safely: slow queries, EXPLAIN plans, index suggestions, table health and lock contention. It is read-only by design.

[Claude Code using pg-perf-mcp: it finds the slow database functions, then diagnoses a slow catalog query and declines to create the index because the tools are read-only]

A real Claude Code session against the demo shop, sped up. First it finds which database functions are slow and why. Then it explains a slow catalog query and estimates the fix with a hypothetical index. Asked to create the index, it says it can't: every tool is read-only.

Why

When a page is slow, developers paste a query into a chat and ask "why is this slow?". The assistant then guesses, because it cannot see the plan, the table sizes, the existing indexes or which queries actually dominate the load.

pg-perf-mcp gives the assistant real numbers from pg_stat_statements, EXPLAIN and the catalogs, through a narrow, audited, read-only interface. It cannot change data, cannot run DDL, and never creates the indexes it recommends.

Demo

A condensed, illustrative session against the demo shop (docker compose up, then ./demo/workload.sh). Exact numbers vary from run to run:

text

You:    The shop database feels slow. What's going on?
Claude: [top_slow_queries orderBy=total_time]        The top statement by total time is          SELECT id, status, total, created_at FROM shop.orders WHERE customer_id = $1 ORDER BY created_at DESC        (thousands of calls, the largest share of total execution time).
        [explain_query sql="... WHERE customer_id = 42 ORDER BY created_at DESC"]        HIGH  SEQ_SCAN_ON_LARGE_TABLE  Seq Scan on shop.orders              reads all ~1.0M rows of shop.orders to apply filter (customer_id = 42); only a handful match.
        [suggest_indexes]        CREATE INDEX CONCURRENTLY idx_orders_customer_id ON shop.orders (customer_id);        hypopg: the planner would use it; estimated cost drops by well over 90%.        Not executed. Review it and run it yourself, ideally off-peak.
        [table_health table=shop.orders]        Autovacuum is disabled on shop.orders and ~20% of rows are dead. Re-enable autovacuum and VACUUM.
        Prioritized actions: 1) add the index above (orders.customer_id is a foreign key with no index),        2) re-enable autovacuum on shop.orders, 3) ...

You:    The shop database feels slow. What's going on?
Claude: [top_slow_queries orderBy=total_time]        The top statement by total time is          SELECT id, status, total, created_at FROM shop.orders WHERE customer_id = $1 ORDER BY created_at DESC        (thousands of calls, the largest share of total execution time).
        [explain_query sql="... WHERE customer_id = 42 ORDER BY created_at DESC"]        HIGH  SEQ_SCAN_ON_LARGE_TABLE  Seq Scan on shop.orders              reads all ~1.0M rows of shop.orders to apply filter (customer_id = 42); only a handful match.
        [suggest_indexes]        CREATE INDEX CONCURRENTLY idx_orders_customer_id ON shop.orders (customer_id);        hypopg: the planner would use it; estimated cost drops by well over 90%.        Not executed. Review it and run it yourself, ideally off-peak.
        [table_health table=shop.orders]        Autovacuum is disabled on shop.orders and ~20% of rows are dead. Re-enable autovacuum and VACUUM.
        Prioritized actions: 1) add the index above (orders.customer_id is a foreign key with no index),        2) re-enable autovacuum on shop.orders, 3) ...

Architecture

mermaid

flowchart LR    A["AI assistant<br/>(Claude Code, Claude Desktop, ...)"] -- "MCP over streamable HTTP<br/>Authorization: Bearer key" --> F["API key filter<br/>(constant-time)"]    F --> M["MCP tools / resources / prompt<br/>Spring AI 2.0 annotations"]    M --> R["ToolRunner<br/>audit log · metrics · safe errors"]    R --> G["SqlGuard<br/>PG lexer + JSqlParser"]    G --> E["ReadOnlyExecutor<br/>READ ONLY tx · always rollback<br/>statement/lock timeout · row cap"]    R --> AN["analysis<br/>PlanAnalyzer · IndexAdvisor<br/>(pure functions)"]    E --> P[("PostgreSQL 18<br/>role mcp_readonly<br/>pg_stat_statements · hypopg")]

flowchart LR    A["AI assistant<br/>(Claude Code, Claude Desktop, ...)"] -- "MCP over streamable HTTP<br/>Authorization: Bearer key" --> F["API key filter<br/>(constant-time)"]    F --> M["MCP tools / resources / prompt<br/>Spring AI 2.0 annotations"]    M --> R["ToolRunner<br/>audit log · metrics · safe errors"]    R --> G["SqlGuard<br/>PG lexer + JSqlParser"]    G --> E["ReadOnlyExecutor<br/>READ ONLY tx · always rollback<br/>statement/lock timeout · row cap"]    R --> AN["analysis<br/>PlanAnalyzer · IndexAdvisor<br/>(pure functions)"]    E --> P[("PostgreSQL 18<br/>role mcp_readonly<br/>pg_stat_statements · hypopg")]

Tools

Tool	What it does	Safety notes
`top_slow_queries`	Most expensive statements from `pg_stat_statements`: normalized text, calls, total/mean/max ms, rows, cache hit ratio, share of total time. Order by `total_time`, `mean_time`, `calls` or `rows`.	Fixed SQL; sort column chosen from an enum. Excludes the server's own statements.
`explain_query`	Summarized plan: 5 costliest nodes, seq scans on large tables, 10x row misestimates, disk sorts/hash spills, nested loops with many iterations. `analyze=true` adds real timings. `$1` placeholders get a generic plan.	SQL guard first. ANALYZE runs in a read-only transaction that is rolled back and cancelled after the timeout.
`suggest_indexes`	`CREATE INDEX CONCURRENTLY` candidates from filtered seq scans, join keys and `ORDER BY ... LIMIT`, each with a reason; validated with hypothetical indexes when `hypopg` is installed.	Never executed. Hypothetical indexes are reset in a savepoint so they cannot leak to other calls.
`unused_indexes`	Indexes with `idx_scan = 0`, largest first, excluding primary key, unique and constraint indexes; includes size, definition and a `DROP INDEX CONCURRENTLY` to review.	Nothing is dropped. Reports when stats were last reset.
`table_health`	Live/dead tuples, dead ratio, last (auto)vacuum/analyze, rows changed since analyze, seq vs index scans, sizes, with plain-language warnings.	Catalog reads only.
`blocking_sessions`	Blocked/blocking pairs via `pg_blocking_pids`, truncated query texts, wait time, lock type/mode/table, blocker state, root blockers.	Never cancels or terminates anything.
`slow_functions`	User-defined functions (PL/pgSQL, SQL, ...) ranked by total, self or mean time or calls, from `pg_stat_user_functions`, with language and volatility.	Catalog reads only. Needs `track_functions`.
`explain_function`	Looks inside the functions a query calls: time per function, every statement executed inside them with calls, time and its own analyzed plan, and findings such as a seq scan inside a function, a function called once per row, or a read-only function left `VOLATILE`. Includes the function source.	SQL guard first; runs in the same rolled-back read-only transaction as `explain_query`, so functions that write fail. `auto_explain` is switched on for that transaction only, with generic plans so no data values appear.

Resources: pg://schema/overview (tables, row estimates, sizes, indexes) and pg://schema/{table} (columns, indexes with scan counts, foreign keys flagged when unindexed). Prompt: diagnose_slow_database, a guided workflow: slow queries, then explain, suggest indexes, table health, functions, then a prioritized summary.

Why a separate tool for database functions

EXPLAIN shows a function call as one opaque expression, and charges its time to whichever plan node evaluates it. On the demo catalog page (20 products, each calling shop.product_rating), explain_query blames the index scan on products (184 ms self time) and finds nothing else serious. explain_function on the same query, against the full demo data:

text

shop.product_rating          plpgsql  20 calls  99% of execution time  SELECT avg(rating) FROM shop.reviews WHERE product_id = p_product_id                             20 calls  175 ms  96% of execution timeHIGH    FUNCTION_DOMINATES_QUERY  shop.product_rating takes 99% of the query's execution timeHIGH    FUNCTION_CALLED_PER_ROW   ran 20 times for one query (9.0 ms per call)HIGH    SEQ_SCAN_ON_LARGE_TABLE   inside shop.product_rating: Seq Scan on shop.reviews                                  reads all ~250k rows to apply filter (reviews.product_id = $1)

shop.product_rating          plpgsql  20 calls  99% of execution time  SELECT avg(rating) FROM shop.reviews WHERE product_id = p_product_id                             20 calls  175 ms  96% of execution timeHIGH    FUNCTION_DOMINATES_QUERY  shop.product_rating takes 99% of the query's execution timeHIGH    FUNCTION_CALLED_PER_ROW   ran 20 times for one query (9.0 ms per call)HIGH    SEQ_SCAN_ON_LARGE_TABLE   inside shop.product_rating: Seq Scan on shop.reviews                                  reads all ~250k rows to apply filter (reviews.product_id = $1)

How it works (ADR 0006): the query runs once with auto_explain enabled for that transaction only, so the plan of each nested statement comes back as a notice; per-function time comes from pg_stat_xact_user_functions and per-statement calls from pg_stat_statements, both as before/after differences.

Security model

Five independent layers stand between the assistant and a write (details in ADR 0001):

SQL guard: exactly one plain SELECT. It is checked by a PostgreSQL-faithful lexer and JSqlParser, and both must agree. Blocks DML/DDL, multiple statements, data-modifying CTEs, SELECT INTO, FOR UPDATE/SHARE, COPY, DO, and dangerous functions (pg_sleep*, *_file, lo_*, dblink*, set_config, pg_terminate_backend, query_to_xml, ...). It fails closed.
Read-only role: mcp_readonly has SELECT plus pg_read_all_stats/pg_monitor, with default_transaction_read_only = on and statement_timeout = 5s set on the role.
Read-only connections: Hikari readOnly=true and a read-only session default.
One rolled-back read-only transaction per call: statement_timeout/lock_timeout set locally, standard_conforming_strings forced on, always rolled back.
Row cap: every result is capped (default 200 rows).

Around those layers: a bearer API key (constant-time comparison; only /actuator/health is open), a JSON audit line per call (SQL stored as a SHA-256 hash plus an 80-character preview), error messages with no stack traces or connection details, and metrics mcp.tool.duration and mcp.guard.rejections.

Quickstart

Requirements: Docker. There are two ways to start:

Use it on your database: pull the published image. No clone or build.
Try the demo shop: clone the repo and get a seeded database with realistic problems to diagnose.

Either way, finish with connecting your AI client.

Use it on your database

1. Pull the image (public on GHCR, built for amd64 and arm64):

2. Prepare the database (once). Create a read-only role like mcp_readonly (see docker/init.sql) and make sure pg_stat_statements is in shared_preload_libraries. Prefer a replica or staging database: analyze=true and explain_function execute the query. For the function tools, also enable function tracking.

3. Run it. Put the settings in a file so secrets stay out of your shell history:

bash

cat > pg-perf.env <<'EOF'PGPERF_DB_URL=jdbc:postgresql://host.docker.internal:5432/mydbPGPERF_DB_USER=mcp_readonlyPGPERF_DB_PASSWORD=<the role's password>MCP_API_KEY=<32+ random characters>EOF

cat > pg-perf.env <<'EOF'PGPERF_DB_URL=jdbc:postgresql://host.docker.internal:5432/mydbPGPERF_DB_USER=mcp_readonlyPGPERF_DB_PASSWORD=<the role's password>MCP_API_KEY=<32+ random characters>EOF

bash

docker run --rm --env-file pg-perf.env -p 127.0.0.1:8080:8080 ghcr.io/gouranshul/pg-perf-mcp:0.1.0

docker run --rm --env-file pg-perf.env -p 127.0.0.1:8080:8080 ghcr.io/gouranshul/pg-perf-mcp:0.1.0

host.docker.internal reaches a database on your machine (Docker Desktop; on Linux add --add-host=host.docker.internal:host-gateway). The MCP endpoint is now http://localhost:8080/mcp.

4. Connect your AI client: see below.

Try the demo shop

To build and test from source you also need Java 25.

bash

git clone https://github.com/gouranshul/pg-perf-mcp && cd pg-perf-mcpcp .env.example .env        # then replace the placeholder secretsdocker compose up --build   # Postgres 18 + seeded demo shop (~4.5M rows) + the MCP server on :8080

git clone https://github.com/gouranshul/pg-perf-mcp && cd pg-perf-mcpcp .env.example .env        # then replace the placeholder secretsdocker compose up --build   # Postgres 18 + seeded demo shop (~4.5M rows) + the MCP server on :8080

Generate some slow-query statistics (optional, but makes top_slow_queries interesting):

bash

./demo/workload.sh          # 60s of deliberately bad queries via pgbench./demo/lock-scenario.sh     # holds a row lock for 2 minutes so blocking_sessions has something to show./demo/smoke-test.sh        # curl-based end-to-end check of health, auth and two tool calls

./demo/workload.sh          # 60s of deliberately bad queries via pgbench./demo/lock-scenario.sh     # holds a row lock for 2 minutes so blocking_sessions has something to show./demo/smoke-test.sh        # curl-based end-to-end check of health, auth and two tool calls

Connect from Claude Code

bash

claude mcp add --transport http pg-perf http://localhost:8080/mcp --header "Authorization: Bearer $MCP_API_KEY"

claude mcp add --transport http pg-perf http://localhost:8080/mcp --header "Authorization: Bearer $MCP_API_KEY"

In PowerShell, set the key first, or the header goes out empty and the server answers 401:

powershell

$env:MCP_API_KEY = "<your MCP_API_KEY>"claude mcp add --transport http pg-perf http://localhost:8080/mcp --header "Authorization: Bearer $env:MCP_API_KEY"

$env:MCP_API_KEY = "<your MCP_API_KEY>"claude mcp add --transport http pg-perf http://localhost:8080/mcp --header "Authorization: Bearer $env:MCP_API_KEY"

Then ask: "Use pg-perf to find out why the shop database is slow", or run the diagnose_slow_database prompt.

Connect from Claude Desktop

Claude Desktop launches stdio servers, so use the mcp-remote bridge (claude_desktop_config.json):

json

{  "mcpServers": {    "pg-perf": {      "command": "npx",      "args": ["-y", "mcp-remote", "http://localhost:8080/mcp", "--header", "Authorization:${AUTH_HEADER}"],      "env": { "AUTH_HEADER": "Bearer <your MCP_API_KEY>" }    }  }}

{  "mcpServers": {    "pg-perf": {      "command": "npx",      "args": ["-y", "mcp-remote", "http://localhost:8080/mcp", "--header", "Authorization:${AUTH_HEADER}"],      "env": { "AUTH_HEADER": "Bearer <your MCP_API_KEY>" }    }  }}

Connect from Cursor

~/.cursor/mcp.json (or .cursor/mcp.json in a project):

json

{  "mcpServers": {    "pg-perf": {      "url": "http://localhost:8080/mcp",      "headers": { "Authorization": "Bearer <your MCP_API_KEY>" }    }  }}

{  "mcpServers": {    "pg-perf": {      "url": "http://localhost:8080/mcp",      "headers": { "Authorization": "Bearer <your MCP_API_KEY>" }    }  }}

Connect from VS Code (Copilot agent mode)

.vscode/mcp.json. VS Code prompts for the key once and stores it securely:

json

{  "inputs": [    { "type": "promptString", "id": "pg-perf-key", "description": "pg-perf MCP_API_KEY", "password": true }  ],  "servers": {    "pg-perf": {      "type": "http",      "url": "http://localhost:8080/mcp",      "headers": { "Authorization": "Bearer ${input:pg-perf-key}" }    }  }}

{  "inputs": [    { "type": "promptString", "id": "pg-perf-key", "description": "pg-perf MCP_API_KEY", "password": true }  ],  "servers": {    "pg-perf": {      "type": "http",      "url": "http://localhost:8080/mcp",      "headers": { "Authorization": "Bearer ${input:pg-perf-key}" }    }  }}

Enable the function tools (optional)

slow_functions and explain_function need two extra settings on your own database. The demo shop already has them, and every other tool works without them:

sql

-- postgresql.conf: track_functions = 'pl' (or 'all' to include SQL functions), then reload.-- Load auto_explain for this role only (no restart needed); it stays inactive until a call enables it:ALTER ROLE mcp_readonly SET session_preload_libraries = 'auto_explain';-- PostgreSQL 15+: let the role change only these settings, only for its own transactions:GRANT SET ON PARAMETER auto_explain.log_min_duration, auto_explain.log_analyze,    auto_explain.log_nested_statements, auto_explain.log_format, auto_explain.log_level,    auto_explain.log_verbose, auto_explain.log_parameter_max_length TO mcp_readonly;

-- postgresql.conf: track_functions = 'pl' (or 'all' to include SQL functions), then reload.-- Load auto_explain for this role only (no restart needed); it stays inactive until a call enables it:ALTER ROLE mcp_readonly SET session_preload_libraries = 'auto_explain';-- PostgreSQL 15+: let the role change only these settings, only for its own transactions:GRANT SET ON PARAMETER auto_explain.log_min_duration, auto_explain.log_analyze,    auto_explain.log_nested_statements, auto_explain.log_format, auto_explain.log_level,    auto_explain.log_verbose, auto_explain.log_parameter_max_length TO mcp_readonly;

Without these, explain_function still reports what it can and says in notes what is missing.

Configuration

Variable / property	Default	Purpose
`MCP_API_KEY`	(required)	Bearer token clients must send; at least 16 characters. The server will not start without it.
`PGPERF_DB_URL`	`jdbc:postgresql://localhost:5432/shop`	JDBC URL
`PGPERF_DB_USER` / `PGPERF_DB_PASSWORD`	`mcp_readonly` / (empty)	Database credentials (use a read-only role)
`pgperf.query.statement-timeout`	`5s`	Per-call `statement_timeout` and `lock_timeout`
`pgperf.query.max-rows`	`200`	Row cap for every query
`pgperf.guard.max-sql-length`	`20000`	Longest SQL accepted
`pgperf.functions.nested-plan-threshold`	`1ms`	`explain_function` keeps the plan of a statement inside a function only for executions at least this long (every execution is still counted)
`OTEL_TRACING_ENABLED`	`false`	Export traces via OTLP/HTTP
`OTEL_EXPORTER_OTLP_ENDPOINT`	`http://localhost:4318/v1/traces`	OTLP traces endpoint
`OTEL_TRACING_SAMPLING_PROBABILITY`	`1.0`	Trace sampling ratio

Properties can also be set as environment variables (PGPERF_QUERY_STATEMENT_TIMEOUT=3s, ...). Metrics are at /actuator/prometheus (requires the API key).

Build and test

This runs unit tests, Testcontainers integration tests against PostgreSQL 18 (same image, init scripts and read-only role as compose; needs Docker), an end-to-end test with the MCP Java SDK client over HTTP, and a JaCoCo gate (80% lines on the guard and analysis packages).

Design decisions

Roadmap

OAuth 2.1 resource server per the MCP authorization spec (per-user identity instead of a shared key)
Multiple named database targets, with read replicas preferred
Workload-level index advice (weigh candidates across all top queries, not one at a time)
auto_explain log / pg_stat_kcache ingestion for plans of queries that already ran in production
Optional raw-plan output and per-tool rate limits
MCP elicitation to confirm analyze=true on expensive queries

License

来源：README.md，提交 2b97e2b

工具

工具元数据尚未被收录。

版本历史

v0.1.0最新Oct 10, 2026

PostgreSQL Performance (read-only)

概览

安装

在 SourceWeft 中

其他 MCP 客户端

README

pg-perf-mcp

Why

Demo

Architecture

Tools

Why a separate tool for database functions

Security model

Quickstart

Use it on your database

Try the demo shop

Connect from Claude Code

Connect from Claude Desktop

Connect from Cursor

Connect from VS Code (Copilot agent mode)

Enable the function tools (optional)

Configuration

Build and test

Design decisions

Roadmap

License

工具

版本历史

更多类似 PostgreSQL Performance (read-only) 的服务

Algeria

Free the Tools

Tewip