Historic SQL Pattern Extractor

A ktx-specific skill that mines historic SQL query shards for recurring cross-table analytical intents and emits verified pattern evidence.

Data & AnalyticsAdvanced★ 1,566⑂ 101AI score 8/10Last updated: Jul 19, 2026

What it does

Reads a bounded SQL template shard (patterns-input/part-0001.json or legacy patterns-input.json) and identifies recurring analytical intents spanning two or more tables.
Packages each intent into a typed pattern evidence object (slug, title, narrative, definitionSql, tablesInvolved, slRefs, constituentTemplateIds) via emit_historic_sql_evidence.
Enforces a strict identifier-verification protocol: discover_data to reuse existing wikis/sources, entity_details to confirm schemas, tables, columns and both sides of every join key, and sql_execution probes for literal values not present in sampleValues.
Unresolvable identifiers must be marked [unverified - from <rawPath>] or logged with the failing probe error in emit_unmapped_fallback.
Clear boundaries: no wiki_write, no SL source writes/edits, no single-table pattern pages, no copying credentials or unredacted literals.

Who it's for

Teams running ktx's memory_agent pipeline to auto-build warehouse knowledge wikis.
Data platform engineers documenting frequently used join/aggregation patterns from query history.
Anyone who needs an LLM workflow that refuses to invent table and column names.

Examples

Order lifecycle pattern: merge several templates joining orders and customers into one order-lifecycle-analysis pattern and emit it once.
Join key verification: before describing c.id = o.customer_id, resolve both columns through entity_details and inspect FK/PK plus sample values.
Fictional identifier guard: if a candidate table won't resolve, run SELECT 1 FROM <ref> LIMIT 0; an error means the identifier is fictional and the error text goes into the fallback record.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/Kaelio/ktx/HEAD/packages/cli/src/skills/historic_sql_patterns/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)

Download the ZIP with the button below.
In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.

↓ Download ZIP

Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the packages/cli/src/skills/historic_sql_patterns folder from the GitHub repo Kaelio/ktx into my ~/.claude/skills/historic-sql-patterns/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/Kaelio/ktx.git && mkdir -p ~/.claude/skills/historic_sql_patterns && cp -r ktx/packages/cli/src/skills/historic_sql_patterns/* ~/.claude/skills/historic_sql_patterns/

⚠ This is a third-party skill. Check the source repository before installing.

Clone the repository: git clone https://github.com/Kaelio/ktx.git
Create the skill folder: mkdir -p ~/.claude/skills/historic_sql_patterns
Copy the files: cp -r ktx/packages/cli/src/skills/historic_sql_patterns/* ~/.claude/skills/historic_sql_patterns/
Confirm ~/.claude/skills/historic_sql_patterns/SKILL.md exists.
Restart Claude Code and check that historic_sql_patterns shows up in your skill list.
Note: the skill only functions where ktx CLI tools such as read_raw_file, entity_details and emit_historic_sql_evidence are available, so set up the ktx pipeline as well.

View source on GitHub ↗License: Apache-2.0