SQL Benchmarks Lab

Query pre-computed SQL engine benchmarks.

Connector by rctruta · GitHub account verified by the MCP Registry

IndexedApache-2.0Local (stdio)★ 2

Chalk Labs

Need help deploying SQL Benchmarks Lab?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/rctruta-sqlbenchdag

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

About

Query pre-computed SQL engine benchmarks. Runs standalone, no server setup required.

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list, install config and code. Checked Oct 25, 2026.

Tools: 9 read · 1 write · 0 delete
Secrets: None asked for
Install scripts: None found
Runs as: Python package

Tools by what they do 10

Read from the source code; the running server may list different tools.

Read 9 · Look things up without changing anything.

analyze_experiment
get_experiment_status
get_template
list_categories
list_engines
list_results
list_suites
list_templates
recommend_engine

Write 1 · Create, change, send or run things.

submit_experiment

Secrets and settings it asks for

Other settings (environment variables or headers).

SB_API_BASE
SB_RESULTS_DIR

Network hosts it talks to 3

astral.sh · found in code
github.com · found in code
sqlbench.local · found in code

Files it touches

~/.ssh/benchmark-key-pair.pem
/mnt/data
/opt/Actian/VectorVW/ingres/bin/sql
/opt/Actian/VectorVW/ingres/bin/vwload
/root/sql-benchmarks-dagster
/tmp/a.parquet
/tmp/b.parquet
/tmp/benchmark_data
/tmp/benchmark_test_data
/tmp/data
/tmp/data/duckdb_isolation
/tmp/duckdb_data
/tmp/q
/tmp/s.parquet
/tmp/sc.parquet
/tmp/smoke_mcp_stderr.log
/tmp/test_data
/tmp/test_duckdb_data
/tmp/test.parquet
/tmp/x.parquet
/Users/ramona/Downloads/production_customers.csv
/Users/ramona/Downloads/production_orders.csv
/var/lib/postgresql/data
/var/lib/typedb/data
Your home directory (~)
~/.dagster
~/Projects/ai-agent-utils/templates/gates

Tools (10)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

analyze_experiment
Analyze a completed experiment. Choose the right intent for your question. Intents: - "summary": Compact digest (means + scaling + narrative). Start here. - "means": Mean duration per partition/engine. - "scaling": Pairwise scaling factors across partitions. - "stability": Std dev, min, max, CV for reliability questions. - "compare": Ranked cross-engine comparison. (Can filter by `partition`). - "compare_by_partition": Ranked comparison broken down per partition. - "raw": The full raw JSON payload (use ONLY if projections aren't enough). Args: experiment_id: 8-character experiment ID. …
get_experiment_status
Check the status of a submitted experiment. Status values: "queued", "running", "complete", "not_found". Args: experiment_id: 8-character experiment ID returned by submit_experiment().
get_template
Return the raw YAML content of a template by name. Args: name: Name of the template from list_templates().
list_categories
List the category taxonomy to narrow down test suites. Small payload — call this FIRST to narrow the suite search.
list_engines
List all available database engines (postgres, duckdb, actian) and the benchmark test suites each engine has SQL for.
list_results
List completed benchmark experiments. Args: suite: Optional filter by test suite name (e.g. "analytical_wall"). engine: Optional filter by engine name (e.g. "duckdb").
list_suites
List benchmark test suites for a specific category. Suites include: analytical_wall, group_by, joins, null_logic, null_sentinel, recursion, selectivity, tpch, acid_test. Args: category: Required filter by category (e.g. "scaling"). Use list_categories to find valid options. include_sql: If true, include the raw SQL per engine (can be large).
list_templates
List curated experiment templates. Each is a valid, human-authored config you can get_template(name) and adapt.
recommend_engine
Get an engine recommendation based on pre-computed benchmark data. Returns the fastest engine for the given workload with confidence level and reasoning. Args: suite: Test suite name representing the workload type (e.g. "analytical_wall", "joins", "selectivity"). scale: Partition key substring to filter by data scale (e.g. "large", "medium", "1000000").
submit_experiment
Submit a new benchmark experiment. The experiment will run asynchronously. Use get_experiment_status() to poll for completion, then get_result() to retrieve data. If this exact experiment has been run before (same config + SQL + code), it returns status "duplicate" and you can immediately call get_result(). CRITICAL: For dataset.tables.*.rows, you MUST use a string alias that maps to definitions.rows, NOT a literal integer. Literal integers will be rejected. Args: config_yaml: YAML string defining the experiment (dataset, engines, matrix).

Signals

Stars: 2
Forks: 0
Open issues: 0
Weekly downloads: 194
Last commit: Oct 21, 2026 (10 days ago)
Last release: Aug 10, 2026 (2 months ago)
Last package publish: Oct 20, 2026 (10 days ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name: io.github.rctruta/sqlbenchdag
Version: 0.1.2
Auth: No auth
Packages: pypi sqlbenchdag
License: Apache-2.0
Repository: github.com/rctruta/sql-benchmarks-dagster
Website: github.com/rctruta/sql-benchmarks-dagster
First seen: Oct 24, 2026
Last checked: Oct 24, 2026
Source: Official MCP Registry github-graphql

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped: 50+; launches shipped
media placements: 500+; media placements
founders on every account: 2; founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get SQL Benchmarks Lab running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Need help deploying SQL Benchmarks Lab?

Call Get help

Cite this page

SQL Benchmarks Lab MCP server. Agent Mart. https://agentmart.dev/connectors/rctruta-sqlbenchdag (last updated 2026-10-24)