From Swantje
Connect a ClickHouse instance to Swantje — collects host/database config and guides env var setup for credentials
How this skill is triggered — by the user, by Claude, or both
Slash command
/swantje:connect-clickhouseThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Guide the user through connecting their ClickHouse instance.
Guide the user through connecting their ClickHouse instance.
Ask the user for each of these. Mark which are required:
Non-secret (write to .swantje/config.json):
host (required) — e.g. abc123.us-east.clickhouse.cloudport (default: 8443 for HTTPS, 8123 for HTTP)database (required) — default database nameusername (required) — usually defaultsecure (default: true) — whether to use HTTPS/TLSSecret (env vars only, never written to files):
CLICKHOUSE_PASSWORD — the password for the username aboveTell the user to set:
export CLICKHOUSE_PASSWORD="your-password-here"
Suggest they add it to their shell profile (~/.zshrc or ~/.bashrc) or a .env file (gitignored) for persistence.
Once you have the values, read .swantje/config.json and update the connectors.clickhouse block:
"clickhouse": {
"enabled": true,
"host": "<value>",
"port": <value>,
"database": "<value>",
"username": "<value>",
"secure": true
}
Write the updated config back to .swantje/config.json.
[STUB — v0.0.1] Tell the user that connection verification is not yet implemented. They can test manually with:
curl "https://<host>:<port>/?query=SELECT+1" \
-u "<username>:$CLICKHOUSE_PASSWORD"
Tell the user ClickHouse is now configured in Swantje and they can use /swantje:analyst to query it.
npx claudepluginhub datumlabsio/swantje --plugin swantjeProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.