Complete governed NL2SQL platform: benchmark, demos, multi-DB adapters, review fixes - #19
Merged
Merged
Conversation
…atabase adapters - Phase 6: agent benchmark (32 gold tasks, 3 schemas, isolated evaluator, ablation, tier 1/2/3 gates), end-to-end demos (docs/demo, 5 narrated+asserting scripts), database adapter contract + DuckDB and PostgreSQL backends. - Independent review fixes: gateway/terminal contracts, transport allowlist, retrieval-scope governance, concurrent usage accounting, session governance, planner observability (spans/usage/latency), Studio SSE consumption. - Honest gaps recorded: no real-model accuracy numbers; PostgreSQL backend implemented but unverified without a live server. - Remove development-process docs (optimization plan/prompts/acceptance records).
…s/demo The offline-acceptance jobs install only core dependencies, so demo D (REST) and demo E (HTTP upload) failed with ModuleNotFoundError instead of skipping, and the integration job still invoked the removed scripts/demo_data_agent.py.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Completes the QueryForge optimization programme and hardens the platform.
Phase 6
docs/demo/— five narrated, asserting scripts (make demo) covering upload→publish→query, semantic catching of wrong SQL, multi-step analysis, transports/refusal/recovery, and the API/repair/attribution loop.DatabaseAdaptercontract plus a verified DuckDB backend and a PostgreSQL backend (implemented; server-backed conformance is DSN-gated and not run in this environment).Independent-review fixes
Gateway terminal contracts,
/analyzetransport allowlist, CLI index policy, cancellation/terminal persistence, retrieval-scope governance (write/read agreement, fail-closed filtering, source-type candidate constraints, delete guards), concurrent usage accounting, session governance (REST/CLI wiring + version refs), planner observability (spans/usage/latency), Studio SSE consumption.Verification
python -m unittest discover -s tests -q: 852 tests OK (24 skipped) — the 24 skips are the PostgreSQL DSN-gated conformance (no live server here).python scripts/run_acceptance.py --full: 13/13.make demo: 5/5, 80 assertions.check_repository.py: 0 errors; semantic drift: 82/82.Honest gaps (not hidden)
Development-process docs (optimization plan, per-step prompts, acceptance records) were removed from the repository at the request of the author.