Does ClickHouse MCP work with ChatGPT (Codex)?

Yes. ClickHouse MCP passed every step of our test in ChatGPT (Codex) on Oct 23, 2026: it installed, connected, its tools were visible, and an agent used it to finish a real task.

Result

ResultPassed
Date testedOct 23, 2026
Plan tested onChatGPT Plus
App versionCodex CLI 0.162.0-alpha.17.2
Server packagemcp-clickhouse 0.7.0
Connect timeconnect time not recorded
Modelgpt-6-luna (reasoning low)

Steps

  1. install OK
  2. connect OK
  3. tools visible OK
  4. agent task: file query OKread-only SQL over a local file with the embedded engine (chDB mode has a select-only tool, so the write-then-read roundtrip can't run)

From our own harness run, not the maker’s claims. How we test.

Versions tested

Codex CLI 0.162.0-alpha.17.2 · mcp-clickhouse 0.7.0 · connect time not recorded · tested Oct 23, 2026

Server package tested: mcp-clickhouse 0.7.0 (uv, Python 3.12.15; resolved 2026-10-23)

Direct and key dependencies as resolved: anyio 4.15.1 · cachetools 7.2.4 · clickhouse-connect 1.11.0 · fastmcp 4.0.11 · mcp 2.3.0 · pydantic 2.14.1 · pydantic-settings 2.16.1 · pydantic_core 2.50.1 · python-dotenv 1.2.4 · simplejson 4.2.0 · sse-starlette 3.5.0 · starlette 1.9.0 · truststore 0.10.5 · typing_extensions 4.16.0 · uvicorn 0.54.0

All 82 resolved dependencies
aiofile==3.12.3
annotated-types==0.8.0
anyio==4.15.1
attrs==26.1.0
Authlib==1.9.1
backports.zstd==1.8.0
beartype==0.23.1
cachetools==7.2.4
caio==0.12.10
certifi==2026.10.23
cffi==2.1.1
chdb==4.4.0
chdb-core==26.9.0
click==8.5.0
clickhouse-connect==1.11.0
cryptography==50.0.2
cyclopts==5.3.0
dnspython==2.9.0
docstring_parser==0.18.0
email-validator==2.3.0
exceptiongroup==1.3.1
fastmcp==4.0.11
fastmcp-slim==4.0.11
griffelib==2.4.1
h11==0.16.0
httpcore2==2.16.0
httpx2==2.16.0
idna==3.20
jaraco.classes==3.4.0
jaraco.context==6.1.2
jaraco.functools==4.6.0
jeepney==0.9.0
joserfc==1.7.7
jsonref==1.1.0
jsonschema==4.26.0
jsonschema-path==0.5.0
jsonschema-specifications==2025.9.1
keyring==25.7.0
lz4==4.4.5
markdown-it-py==4.2.0
mcp==2.3.0
mcp-types==2.3.0
mdurl==0.1.2
more-itertools==11.2.1
numpy==2.5.4
openapi-pydantic==0.6.0
opentelemetry-api==1.45.1
packaging==26.3
pandas==3.1.0
pathable==0.6.0
platformdirs==4.13.5
py-key-value-aio==0.4.6
pyarrow==26.0.0
pycparser==3.12
pydantic==2.14.1
pydantic-settings==2.16.1
pydantic_core==2.50.1
Pygments==2.21.0
PyJWT==2.15.1
pyperclip==1.11.0
python-dateutil==3.0.0
python-dotenv==1.2.4
python-multipart==0.0.32
PyYAML==6.0.3
referencing==0.37.0
rich==15.0.0
rich-rst==2.2.0
rpds-py==2026.9.1
SecretStorage==3.5.0
simplejson==4.2.0
six==1.17.0
sse-starlette==3.5.0
starlette==1.9.0
truststore==0.10.5
typing-inspection==0.4.4
typing_extensions==4.16.0
tzdata==2026.5
uncalled-for==0.4.1
urllib3==2.8.0
uvicorn==0.54.0
watchfiles==1.3.0
websockets==17.3

Captured with: uvx --from mcp-clickhouse[chdb] python (importlib.metadata in the same uvx env; equivalent to uv pip freeze).

What happened in this run

Install config we used

[mcp_servers.clickhouse-mcp]
command = "uvx"
args = ["--from", "mcp-clickhouse[chdb]", "mcp-clickhouse"]
env = {CHDB_ENABLED="true",CLICKHOUSE_ENABLED="false"}

Connection check

server answered tool calls during the task

Tools check

verified via task step: agent saw and called the server's tools

Agent task: file query

  • Tool calls, in order: clickhouse-mcp:run_chdb_select_query

Raw details from our harness log for this run. Secrets redacted.

Set it up in ChatGPT (Codex)

Copy-paste config for every app is on the ClickHouse MCP page. Not connecting? ChatGPT (Codex) MCP troubleshooting.

Other apps we tested it in: Claude Code (passed) · Cursor (passed).

All MCP servers tested in ChatGPT (Codex)

Sponsors