Historic Sql Table Digest

This does one thing: it takes a single table's query history from the historic-sql adapter and distills it into structured evidence about how that table is actually used. You get frequency tier, common filters, joins, and group-bys packaged into a narrative that feeds schema projection downstream. The workflow is strictly constrained: read the raw JSON, produce exactly one table_usage evidence object, then stop. No wiki writes, no semantic layer edits. The identifier verification protocol is notably thorough, requiring you to confirm every schema.table reference resolves before emitting anything. Useful when you need deterministic table usage facts extracted from query logs without second-guessing or creative expansion.

GitHub stars: 1.1K
last updated: Jun 11, 2026

View on GitHub

Install

$ npx -y skills add kaelio/ktx --skill historic_sql_table_digest --agent claude-code

Installs into .claude/skills of the current project.

Files

Historic SQL Table Digest

Use this skill when the WorkUnit raw file is one tables/<schema>.<name>.json file from the historic-sql adapter.

Required Workflow

Read the WorkUnit notes first.
Call read_raw_file for the single tables/<schema>.<name>.json raw file.
Read manifest.json only if the table JSON omits the dialect or the WorkUnit notes are unclear.
Produce one concise usage narrative for this table from the staged table JSON.
Call emit_historic_sql_evidence exactly once with kind: "table_usage".
Stop after the evidence tool succeeds.

Identifier Verification Protocol

Before writing a wiki page or SL source on any topic:

discover_data({query: "<topic>"}) - see what wikis, SL sources, and raw tables already exist. Prefer updating existing pages over creating new ones.

Before emitting any schema.table or schema.table.column into a wiki body, SL source, tables: frontmatter, sl_refs, or emit_unmapped_fallback:

SKILL.md on GitHub