sql-dynamic-pattern-analyzer

Analyzing SQL Dynamic Patterns

Analyzes Dynamic SQL occurrences flagged by SnowConvert (issue code SSC-EWI-0030), classifies them against a dialect-specific pattern catalog, scores complexity, and records migration considerations. All operations go through the scai assessment sql-dynamic command — there are no Python helpers, no CSV/JSON scripts, no bash loops.

Sub-Agent Mode

When invoked from a parent skill (e.g., assessment/SKILL.md) as a sub-agent, the parent provides a context block with the fields below. The review_mode field controls whether the per-occurrence review loop runs.

On entry: call the configure MCP tool with project_dir from the context block. Snowflake credentials are not required — scai assessment sql-dynamic reads only local CSVs and source files.

Branching by review_mode:

• generate-only — run only scai assessment sql-dynamic generate (Workflow Step 1) and return. The output JSON will contain PENDING occurrences; that is expected.
• auto-review-all — generate, then loop show-code-unit → analyze → update --status REVIEWED for every PENDING occurrence (Workflow Steps 2–6). Run stats at the end and confirm zero PENDING. The single-record / unique-analysis rule from Critical Rules (NO BATCH UPDATES) still applies — each occurrence gets its own analysis.
• skip — return immediately with "status": "skipped". The parent should not have dispatched in this case; this branch is defensive.

On completion, return JSON only:

{
  "sub_skill": "analyzing-sql-dynamic-patterns",
  "status": "ok",
  "output_json": "<abs path to sql_dynamic_analysis.json>",
  "summary": "<one-line: total occurrences, REVIEWED count, PENDING count>",
  "error": null
}