SQL Table Reference Extractor

Overview

You pass a SQL statement and we report tables, common table expressions, column references (when requested), and write targets with role and access mode. We do not resolve schemas or expand SELECT *. Do not use this when you need column lineage or schema-qualified expansion.

Regex table lists miss writes and CTEs. We walk the parse tree so you see what the statement actually references.

When to use it

Scope

Non-goals

• Does not resolve schemas or expand wildcards
• Does not prove read-only intent at runtime
• Does not build full column lineage graphs

Properties

Capabilities

Cost & billing

Flat rate per invocation

File handling

How we handle your data

Input schema

Output schema

API contract

curl -sS -X POST "$BASE/v1/execute" -H "Authorization: Bearer $KEY" -H 'Content-Type: application/json' -d '{"skillKey":"database.sql_extract_tables","input":{"sql":"SELECT 1","dialect":"postgresql"}}'

import { Squillify } from "@squillify/sdk";

const sq = new Squillify({ apiKey: process.env.SQUILLIFY_API_KEY });

const { output } = await sq.execute("database.sql_extract_tables", {"sql":"SELECT u.name FROM users u JOIN orders o ON o.user_id = u.id","dialect":"postgresql"});

Errors

Other limits

Limitations

• Does not resolve schemas or expand wildcards
• Does not prove read-only intent at runtime
• Does not build full column lineage graphs
• Schema-qualified column expansion
• Semantic data lineage across pipelines

How to call

POST /v1/execute (requires authentication)

{
  "skillKey": "database.sql_extract_tables",
  "input": {
    "sql": "SELECT u.name FROM users u JOIN orders o ON o.user_id = u.id",
    "dialect": "postgresql"
  }
}

Examples

{
  "sql": "SELECT u.name FROM users u JOIN orders o ON o.user_id = u.id",
  "dialect": "postgresql"
}

{
  "unresolvedColumns": [],
  "statementType": "SELECT",
  "writes": [],
  "tables": [
    {
      "name": "orders",
      "access": "READ",
      "role": "SOURCE"
    },
    {
      "name": "users",
      "access": "READ",
      "role": "SOURCE"
    }
  ],
  "columns": [],
  "findings": [],
  "ctes": [],
  "_meta": {
    "referenceCount": 2,
    "resolveAliases": true,
    "parserProfile": "jsqlparser-5.3"
  },
  "hasWildcard": false
}

{
  "sql": "INSERT INTO audit_log (msg) VALUES ('x')",
  "dialect": "postgresql"
}

{
  "unresolvedColumns": [],
  "statementType": "INSERT",
  "writes": [
    {
      "name": "audit_log",
      "access": "WRITE"
    }
  ],
  "tables": [
    {
      "name": "audit_log",
      "access": "WRITE",
      "role": "TARGET"
    }
  ],
  "columns": [],
  "findings": [],
  "ctes": [],
  "_meta": {
    "referenceCount": 1,
    "resolveAliases": true,
    "parserProfile": "jsqlparser-5.3"
  },
  "hasWildcard": false
}

Recommended use cases

• Gate read-only agents by checking the writes array is empty
• Inventory tables touched by a migration script
• Pre-flight review before granting query access

Avoid for

• Schema-qualified column expansion
• Semantic data lineage across pipelines

Related skills

Composable with

Safety & privacy

• Runs without calling any external service.
• Your payloads are never stored permanently.
• Telemetry captures metadata only — never your payload contents.
• We never train models on your payloads.
• No human reviews your payloads.

Tags