Skip to content

feat: prepare v0.9.0 reviewed Sol evidence #6

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
iarjunganesh merged 1 commit into from
Jul 17, 2026
Merged

feat: prepare v0.9.0 reviewed Sol evidence #6

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 9 additions & 3 deletions AGENTS.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,7 @@ PostgreSQL was verified at migration `0003` through its public TCP proxy; an
eight-source capture produced six verifier-passed drafts, and four were
published after human review (Transformers v5.14.1, vLLM v0.25.1, NCCL
v2.30.7-1, TensorRT 11.1). The hosted `v0.6.1` app was then verified through
`/health`, `/briefing` (now returning the four reviewed Insights), branded
`/health`, `/briefing` (now returning the five reviewed Tier.FINAL Insights), branded
`/docs`, a Vercel-to-Railway CORS preflight, and provider-backed `/search` and
`/chat` over that reviewed set. The Vercel HTML also references the canonical
API-served banner pair. This is a small, bounded reviewed set — not broad or
Expand All @@ -41,9 +41,14 @@ coverage; `main` branch protection requires the CI quality gate and the Codecov
`pytest` upload is confirmed (both verified 2026-07-17).
The 2026-07-17 freeze-plan audit and documentation synchronization is recorded
under Codex session `019f7190-912d-70e3-be6d-fcc81bf8e203`.
The v0.9.0 evidence cleanup and session synchronization is recorded under
Codex session `019f7213-be19-7e50-92ac-a48bd5ecaacb`.
The source patch `v0.8.1` contains the Ask DRIFT grounding/citation fix and is
tagged from `feature/v0.9.0-final-evidence`; hosted verification remains
`v0.8.0` until a deployment is independently checked.
at the deployed `v0.8.0` app build until a v0.9.0 redeploy is independently
checked. The live Railway store was updated on 2026-07-17 and `/briefing`
verified exactly five reviewed Tier.FINAL (`gpt-5.6-sol`) Insights: 10, 11, 13,
15, and 16; Luna IDs 3, 6, 7, and 8 are draft.

## Key commands

Expand Down Expand Up @@ -187,7 +192,8 @@ Railway FastAPI service built from the repository root with `Dockerfile` and
`railway.json`. The public frontend is
`https://dr1ftless.vercel.app` and the API is
`https://drift-api-prod.up.railway.app`. The current verified hosted application
is `v0.8.0` in `DRIFT_MODE=live`. On 2026-07-17, Railway `/health` reported
build is `v0.8.0` in `DRIFT_MODE=live`; the current source version is `v0.9.0`.
On 2026-07-17, Railway `/health` reported
`0.8.0`, `/docs` returned `200`, the public Vercel page rendered Ask DRIFT, and
a Vercel-origin CORS preflight allowed `GET, POST`. `v0.7.0` is the prior
verified release; `v0.6.1` was the revision before that. On 2026-07-16, `/health` reported
Expand Down
132 changes: 96 additions & 36 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,45 +8,23 @@ Format: [Keep a Changelog](https://keepachangelog.com/en/1.1.0/) · Versioning:
The `0.1.0` entry is the initial repository baseline published on GitHub as
the annotated `v0.1.0` tag.

## [0.8.1] - 2026-07-17

### Fixed

- Made live Ask DRIFT responses structured and grounding-aware. The model now
reports the Insight IDs it actually used, the API filters citations and
`grounded_insight_ids` to that subset, and stray or malformed model IDs can
no longer cause unrelated source links to be presented for a question.
- Added regression coverage for subset citation behavior, malformed structured
responses, blank answers, router schema forwarding, and lifespan integration.

### Release boundary

- This is a source patch release for the Ask DRIFT grounding fix, tagged from
`feature/v0.9.0-final-evidence`. Hosted Railway/Vercel verification remains
at `v0.8.0` until a deployment is independently checked.

## Targeted releases — planned, not released

These targets are planning records only. They do not change the current
verified release (`v0.8.0`) or claim that the paid Tier.FINAL rerun has run.
verified app deployment (`v0.8.0`) or claim that the `v0.9.0` app build has
been redeployed. The `v0.9.0` Tier.FINAL evidence pass and live-store update
are recorded in the `[0.9.0]` entry below.

### `v0.9.0` — final-tier evidence and selected polish
### `v0.9.1` — Terra grounded-run evidence

Targeted scope:

- rerun selected reviewed examples through Tier.FINAL (Sol), initially the
four currently published sources: Transformers, vLLM, NCCL, and TensorRT;
- preserve source bytes, exact claim spans/hashes, upstream references, model
identifiers, verifier results, model-run audits, and meaningful human review
notes for every accepted result;
- publish only verifier-passed results after human review, retaining the
existing bounded-evidence and spend/retry guardrails; and
- optionally include narrowly scoped polish that improves judge clarity or the
demo path, provided it does not delay or blur the evidence boundary.

The Tier.FINAL run is still pending operator execution and budget validation.
The reported remaining account balance is planning context, not release
evidence; the configured spend guard remains authoritative.
Planned, not released. After the Tier.FINAL Sol evidence is complete, run a
small, explicitly bounded Tier.LIVE / `gpt-5.6-terra` capture against the
reviewed store. Preserve retrieved Insight IDs, structured answer audits,
citations, provider metadata, spend/retry records, and updated screenshots as
inspectable evidence. Verify that Terra answers remain retrieve-first and cite
only the evidence used; do not expand the product surface or imply broad
continuous monitoring. Publish this release only after the run, artifacts,
tests, documentation, and intended deployment path are independently checked.

### `v1.0.0` — final submission release

Expand All @@ -61,11 +39,92 @@ Targeted scope:
`v1.0.0` is the submission-final target, not permission to add new feature
surfaces such as MCP, tool calling, IDE integration, or a release timeline.

## [0.9.0] - 2026-07-17

### Tier.FINAL evidence pass — 2026-07-17

- Follow-up Codex session `019f7213-be19-7e50-92ac-a48bd5ecaacb` completed the
evidence cleanup and release-boundary synchronization on 2026-07-18.
- Executed the paid Tier.FINAL capture (`gpt-5.6-sol`) through the DRIFT Manual
Run notebook. All eight configured primary sources were sampled (one item
each); eight clusters produced verifier-passed drafts (IDs 9–16), each drafted
and separately verified at `gpt-5.6-sol`.
- After source-excerpt human review, published five Insights: JAX v0.11.0 (10),
Transformers v5.14.1 (11), vLLM v0.25.1 (13), NCCL v2.30.7-1 (15), and
TensorRT 11.1 (16). This grows the reviewed corpus to five sources, adding
JAX v0.11.0 — a substantive release with real breaking changes — alongside the
four sources first published at the dev / `gpt-5.6-luna` tier.
- Held three verifier-passed drafts at the human-review gate rather than
publishing them: a single PyTorch trunk commit (not a release), CUTLASS 4.6.1
(thin evidence with an interpretive expansion of an upstream typo), and an
out-of-tree Triton `gfx950` tutorial pin (not a mainline release).
- Archived the reviewed evidence and SHA-256 integrity manifest to
`assets/evidence/2026-07-17-all-sources-reviewed.json`
(sha256 `df32b3d4315b09fb4a6dbd508d381ca2c8095e25e16f88c4741ce7bc3055a337`).
- Recorded a display-only results notebook at
`notebooks/drift_manual_run.sol.results.ipynb` (Markdown-only; no executable
cells, operator configuration, provider/budget logs, or review notes), kept as
a companion to the earlier `gpt-5.6-luna` results record rather than replacing
it.

### Operational note

- A Tier.FINAL (`gpt-5.6-sol`) capture needs a higher per-attempt model timeout
than the 20-second default; the heavier draft stage otherwise exceeds the
timeout, exhausts its retries, and aborts the run (a non-`ValueError` stops the
capture rather than skipping one cluster). Raising `DRIFT_MODEL_TIMEOUT_SECONDS`
to the 120-second maximum let the eight-source capture complete. Local spend for
the reviewed Sol run stayed within the configured spend guard.

### Changed

- Rewrote the `submission/DRIFT_FREEZE_PLAN.md` score matrix as an honest
current-vs-target readiness view and added a dated audit addendum; reclassified
MCP, tool-calling, and IDE items as optional future work that is not shipped.
Updated `docs/INITIATIVES.md`, `docs/CODEX_PROMPTS.md`, and
`submission/SUBMISSION.md`, and removed the superseded
`submission/NEXT_STEPS.md`.
- Renamed the tracked UI screenshots to a kebab-case set led by the reviewed
briefing (`01-briefing`), claim evidence (`02-briefing-claim-evidence`), the
Ask DRIFT box (`03-ask-drift`), two grounded-answer examples
(`04.1-ask-drift-grounded-answer-nccl-example`,
`04.2-ask-drift-grounded-answer-tensorRT-example`), and the branded API docs
(`05-api-docs`); updated the README gallery image paths to match.

### Release boundary

- This is a source and reviewed-evidence release. The live Railway store now
serves exactly five reviewed, verifier-passed Tier.FINAL Insights (IDs 10,
11, 13, 15, and 16), all produced at `gpt-5.6-sol`; the four Luna Insights
(IDs 3, 6, 7, and 8) were retracted to draft on 2026-07-17. Hosted app
verification remains at `v0.8.0` pending redeploy: the deployed Railway
`/health` and Vercel frontend are not being described as `v0.9.0`. The
reported remaining account balance is planning context, not release
evidence; the configured spend guard remains authoritative.

## [0.8.1] - 2026-07-17

### Fixed

- Made live Ask DRIFT responses structured and grounding-aware. The model now
reports the Insight IDs it actually used, the API filters citations and
`grounded_insight_ids` to that subset, and stray or malformed model IDs can
no longer cause unrelated source links to be presented for a question.
- Added regression coverage for subset citation behavior, malformed structured
responses, blank answers, router schema forwarding, and lifespan integration.

### Release boundary

- This is a source patch release for the Ask DRIFT grounding fix, tagged from
`feature/v0.9.0-final-evidence`. Hosted Railway/Vercel verification remains
at `v0.8.0` until a deployment is independently checked.

## [0.8.0] - 2026-07-17

### Documentation audit — 2026-07-17

- Recorded Codex session `019f7190-912d-70e3-be6d-fcc81bf8e203` as the additive
- Recorded Codex sessions `019f7190-912d-70e3-be6d-fcc81bf8e203` and
`019f7213-be19-7e50-92ac-a48bd5ecaacb` as additive
freeze-plan audit and documentation-synchronization initiative.
- Corrected `submission/DRIFT_FREEZE_PLAN.md` so unimplemented release-timeline,
MCP, tool-calling, and IDE features are not marked as shipped; aligned the
Expand Down Expand Up @@ -253,7 +312,7 @@ surfaces such as MCP, tool calling, IDE integration, or a release timeline.
endpoint-level regression tests.
- Sanitized `notebooks/drift_manual_run.ipynb` to a clean, output-free template
(reset capture trigger, publish/archive IDs, and review notes), and made
`notebooks/drift_manual_run.results.ipynb` a Markdown-only display record;
`notebooks/drift_manual_run.luna.results.ipynb` a Markdown-only display record;
it contains no runnable cells, operator configuration, provider/budget logs,
or review notes. A regression test protects that boundary.
- Full suite: 149 tests at 100% backend coverage.
Expand Down Expand Up @@ -647,6 +706,7 @@ with explicit live-path architecture and publication-ready quality gates.
`019f61fc-c32e-7d92-9d2e-0bd9083d08e7`.
- Full scope and submission guidance: [`docs/INITIATIVES.md`](docs/INITIATIVES.md).

[0.9.0]: #090---2026-07-17
[0.8.1]: #081---2026-07-17
[0.8.0]: #080---2026-07-17
[0.7.0]: #070---2026-07-16
Expand Down
43 changes: 25 additions & 18 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -72,10 +72,10 @@ Built for **OpenAI Build Week 2026 · Developer Tools**.

## Try DRIFT in 60 Seconds

**Hosted — nothing to install.** The live briefing serves four human-reviewed
Insights from one real bounded capture (Transformers v5.14.1, vLLM v0.25.1,
NCCL v2.30.7-1, TensorRT 11.1); `/briefing`, `/search`, and `/chat` were verified
provider-backed on 2026-07-16.
**Hosted — nothing to install.** The live Railway store serves five human-reviewed
Tier.FINAL Insights (JAX v0.11.0, Transformers v5.14.1, vLLM v0.25.1, NCCL
v2.30.7-1, TensorRT 11.1); `/briefing?top_n=10` was verified on 2026-07-17.
The deployed app build remains `v0.8.0` pending redeploy.

| Surface | Link |
| --- | --- |
Expand Down Expand Up @@ -201,6 +201,7 @@ the bounded local capture path.
| Submission audit and frontend evidence presentation | `019f6a46-e3eb-7de2-81b1-91515ae80043` | Handwritten-next-step audit, explicit briefing states, system-theme presentation, canonical API-served banners, and status synchronization |
| Reviewed-evidence release hardening and hosted verification | `019f6a78-6fa2-7121-9059-85ac8ceb9904` | Evidence-byte integrity, database-only review notes, display-only results artifact, `v0.7.0` hosted verification, and the `v0.8.0` grounded-chat / verifiable-fixture hosted release |
| Freeze-plan audit and documentation synchronization | `019f7190-912d-70e3-be6d-fcc81bf8e203` | Audited the frozen scope against tracked implementation, corrected unshipped MCP/timeline/IDE claims, aligned demo requirements, and synchronized records |
| v0.9.0 evidence cleanup and session synchronization | `019f7213-be19-7e50-92ac-a48bd5ecaacb` | Retracted superseded Luna Insights through the audited review helper, verified the five-record Sol briefing, synchronized release status, and made the Luna results artifact explicit |

See the full [project initiative record](docs/INITIATIVES.md).

Expand All @@ -220,7 +221,9 @@ submission-audit/frontend-presentation follow-up is
hosted-verification, and `v0.8.0` release session is
`019f6a78-6fa2-7121-9059-85ac8ceb9904`.
The freeze-plan audit and documentation-synchronization session is
`019f7190-912d-70e3-be6d-fcc81bf8e203`.
`019f7190-912d-70e3-be6d-fcc81bf8e203`; the v0.9.0 evidence-cleanup and
session-synchronization follow-up is
`019f7213-be19-7e50-92ac-a48bd5ecaacb`.

GPT-5.6 is used only when an operator explicitly enables `DRIFT_MODE=live` and
provides an API key. The local capture job routes embeddings, classification,
Expand Down Expand Up @@ -261,7 +264,7 @@ Agent code must not hard-code provider model names. The intended tiers are:
| --- | --- | --- |
| `dev` / Luna | Classification, clustering, and prompt iteration | Produced the four reviewed Insights published 2026-07-16; run only with an explicit live key |
| `live` / Terra | Retrieve-first grounded chat | Serves hosted grounded chat over reviewed pgvector rows; verified provider-backed 2026-07-16 |
| `final` / Sol | Three to five reviewed demo insights | Targeted for the planned `v0.9.0` evidence pass; the 2026-07-16 demo set remains the saved `dev`/Luna evidence until rerun and review |
| `final` / Sol | Three to five reviewed demo insights | Produced the five reviewed Tier.FINAL Insights now served by the live Railway store; app redeploy remains pending |

Every live insight must preserve:

Expand Down Expand Up @@ -319,9 +322,9 @@ in [Try DRIFT in 60 Seconds](#try-drift-in-60-seconds):
| Surface | Status |
| --- | --- |
| **Current hosted release** | `v0.8.0` `live` — `/health` reports `0.8.0`, `/docs` returns `200`, the public Vercel page renders Ask DRIFT, and Vercel-origin CORS allows `GET, POST` |
| **Current source patch** | `v0.8.1` — structured Ask DRIFT grounding IDs restrict question citations to the Insights actually used; tagged on `feature/v0.9.0-final-evidence`, not yet hosted-verified |
| **Current source release** | `v0.9.0` — Tier.FINAL reviewed evidence, audited retraction helper, and structured Ask DRIFT grounding; app redeploy remains pending |
| **Prior hosted release** | `v0.7.0` `live` — `/briefing` (four reviewed Insights, review notes redacted), `/openapi.json`, and the deployed Vercel bundle's `top_n=10` request were verified |
| **Grounded retrieval** | `/briefing`, `/search`, and `/chat` verified provider-backed from an eight-source capture — six verifier-passed drafts, four published after human review |
| **Live reviewed store** | `/briefing?top_n=10` verified 2026-07-17 with exactly five reviewed, verifier-passed Tier.FINAL Insights: 10, 11, 13, 15, and 16 |
| **Database schema** | Railway PostgreSQL at `0003_claim_evidence_review_gate`, verified through its public TCP proxy |
| **Branding** | Swagger banner frame and canonical API-served banner pair follow the same system light/dark preference |
| **Historical** | `v0.5.1` served one unreviewed vLLM Insight through `/briefing` on 2026-07-15 — retained as pre-gate evidence only |
Expand All @@ -334,30 +337,34 @@ and **Chat** sections so reviewers can navigate the API by job.

## Screenshots & Evidence

These are live-state captures of the hosted app after four human-reviewed
Insights were published — not mockups. Click any image to open it full size.
These are live-state captures of the hosted app and reviewed Tier.FINAL evidence
— not mockups. Click any image to open it full size.

| Briefing — four reviewed Insights | Inspecting frozen claim evidence |
| Briefing — five reviewed Insights | Inspecting frozen claim evidence |
| --- | --- |
| [![DRIFT briefing showing four reviewed Insights](assets/screenshots/03-briefing.png)](assets/screenshots/03-briefing.png) | [![DRIFT briefing expanded to show each claim's primary-source excerpt](assets/screenshots/04-briefing-claim-evidence.png)](assets/screenshots/04-briefing-claim-evidence.png) |
| [![DRIFT briefing showing five reviewed Insights](assets/screenshots/01-briefing.png)](assets/screenshots/01-briefing.png) | [![DRIFT briefing expanded to show each claim's primary-source excerpt](assets/screenshots/02-briefing-claim-evidence.png)](assets/screenshots/02-briefing-claim-evidence.png) |

The Next.js briefing view exposes each record's status label, confidence,
model/audit label, rationale, bounded action, source links, and—when present—
claim-type evidence. On 2026-07-16, the hosted UI was verified against the
review-gated API; reviewed evidence has since been published, so `/briefing`
now serves four reviewed Insights instead of the intentional empty state.
now serves the five reviewed Tier.FINAL Insights in the live store; the public
app UI remains the deployed `v0.8.0` build pending redeploy.
`v0.7.0` deployed the review-note redaction and ten-item briefing request;
Railway, CORS, public-contract, and Vercel-bundle checks are recorded in the
changelog. `v0.8.0` deployed the grounded Ask DRIFT UI and tag-pinned synthetic
fixture evidence; Railway health/docs, the public UI, and CORS were verified.
Neither rollout re-ran paid provider chat.

The branded Swagger contract and the themed architecture assets are also
available as visual evidence:

| API documentation | Architecture (light / dark) |
| Ask DRIFT — grounded answer | Ask DRIFT — cited example |
| --- | --- |
| [![DRIFT branded Swagger API documentation](assets/screenshots/02-api-docs.png)](assets/screenshots/02-api-docs.png) | [![DRIFT architecture light](assets/architecture/arch-pipeline-light.png)](assets/architecture/arch-pipeline-light.svg)<br/>[![DRIFT architecture dark](assets/architecture/arch-pipeline-dark.png)](assets/architecture/arch-pipeline-dark.svg) |
| [![DRIFT Ask DRIFT grounded chat box](assets/screenshots/03-ask-drift.png)](assets/screenshots/03-ask-drift.png) | [![DRIFT Ask DRIFT grounded NCCL answer](assets/screenshots/04.1-ask-drift-grounded-answer-nccl-example.png)](assets/screenshots/04.1-ask-drift-grounded-answer-nccl-example.png) |

The branded Swagger contract is also available as visual evidence:

| API documentation |
| --- |
| [![DRIFT branded Swagger API documentation](assets/screenshots/05-api-docs.png)](assets/screenshots/05-api-docs.png) |

The scrubbed hosted capture evidence is stored separately in
[`assets/evidence/`](assets/evidence/), including the verified unreviewed vLLM
Expand Down
Loading
Loading