Initialize persistent engine storage.

Insert one row into the persistent engine.

Upsert one row by order_id.

Create customer index for engine table.

Run EXPLAIN ANALYZE style output by customer filter.

Re-index this project dataset (engine_cli index).

Execute full MiniPG + MiniDatabricks + DuckDB end-to-end flow.

Explain one recorded run_id from the local trace store.

Run a traced demo flow, then explain the run immediately.

Run a traced failing flow, then explain the failed run immediately.

Run a traced semantic-corruption flow and explain the failed validation.

Run a traced idempotency-conflict flow through WriteCore.

Run a traced concurrency conflict plus failure-storm scenario.

Run regression checks and return explain output for each traced run.

Append one MCP tool trace record to local trace store.

Find semantically similar historical incidents.

Incrementally ingest new lines from source path into trace store.

Incrementally ingest project docs, code, and config files.

Upsert one operational memory entry for semantic recall.

Search memory entries by semantic similarity and metadata filters.

Run a quick MCP smoke flow and summarize status.

Benchmark MCP operations with SLO-style summary metrics.

Mixed workload load test: insert/upsert/explain/reindex/e2e and summary metrics.

Capture baseline KPI snapshot for ROI Phase 0.

Increment and persist split-logic drift bug counter.

Evaluate migration triggers and return pass/fail gate.

Describe project state roots, schemas, and supported regression primitives.

Return declared runtime capabilities and contract coverage.

Return the full MCP tool catalog with groups, entrypoints, and summaries.

Return default workspace, paths, runtime mode, and project metadata.

Run the explain-first regression bundle and return a unified verdict.

Capture a baseline snapshot and return a unified verdict envelope.

Compare current benchmark/scenario results against a stored baseline.

List declared project entities from the generic state store.

Load one declared project entity by identity key.

Create or update a declared entity in the generic state store.

Delete a declared entity from the generic state store.

Append a generic project event to the local event log.

Append one normalized trace record through the generic project ingest path.

Read one run explanation through the generic project explain entrypoint.

Export generic project state as a JSON snapshot.

List declared heuristic profiles available for generic project analysis.

Run one declared heuristic profile over a source and persist the analysis through project state.

Ingest DDL from file or raw text, validate with DuckDB, store metadata in project state (schemas/).

Run EXPLAIN for each profile in query_profiles_json (JSON: name -> SQL). Optional seed_sql_json (JSON array of SQL) runs before EXPLAIN. Writes explain_.txt.

Build verdict from schema metadata and EXPLAIN outputs; write evaluation_report.json and verdict.md. If query_profile_names empty, discovers explain_*.txt in artifacts.

One-shot: load schema → run EXPLAIN (from query_profiles_json/seed_sql_json) → evaluate. Generic: you supply queries and optional seed SQL.