From 657017a19b8a12203e46c4b77a91b885e0387a00 Mon Sep 17 00:00:00 2001 From: Om Patel Date: Sat, 25 Jul 2026 18:15:29 -0700 Subject: [PATCH] docs(readme): record Track 2 FAIL and Track 3 Wave-1 draft in the status table MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The root README told a reader Track 2 was "search in progress" and Track 3 was "placeholders only". Both were false: Track 2 is FAIL, adjudicated in docs/research/vertical-search/DECISION.md and accepted as ADR-0004, and the Wave-1 pitch pack landed in docs/pitch/. The first file anyone opens misstated the most important fact about where the project is. Status table now cites the artifact behind each claim. A line under it makes the dependency explicit — two consecutive vertical FAILs mean the company rests on the Track-1 mechanism number — citing README-narrative.md section 6. Counts are quoted from DECISION.md (2/70 Week-0, 2/75 Track-2, 0 DURABLE surfaces), not recomputed. Also: Quick start now says what `npm run ci` actually runs and that it must be green before any PR; Layout gains scripts/, tests/, artifacts/ and pointers to architecture.md, ROADMAP.md and DEVELOPMENT.md. The public-repo section is untouched — its secret-hygiene rules are load-bearing. No performance number, percentage, or gate number anywhere in the file. docs/README.md index row and its open question updated; INTEGRITY-AUDIT D-04 marked resolved, with the residual deck/objections lag left open under D-01/E-12. Closes #54 --- README.md | 24 ++++++++++++++++++++---- docs/INTEGRITY-AUDIT.md | 2 +- docs/README.md | 7 +++++-- 3 files changed, 26 insertions(+), 7 deletions(-) diff --git a/README.md b/README.md index 4f28452..384c2d5 100644 --- a/README.md +++ b/README.md @@ -18,11 +18,14 @@ on SaaS observability config, especially where the person doing the work is the | Track | Question | Status | | --- | --- | --- | | Track 1 | Do compiled trajectories survive site churn? | Harness in progress — **no gate number yet** | -| Track 2 | Is there a vertical where the counterparty hypothesis holds? | Search in progress | -| Track 3 | Narrative / pitch | Placeholders only until Track 1 + 2 land | +| Track 2 | Is there a vertical where the counterparty hypothesis holds? | **FAIL — no vertical locked** ([C5 DECISION](docs/research/vertical-search/DECISION.md), [ADR-0004](docs/decisions/ADR-0004-vertical-track2-fail.md)) | +| Track 3 | Narrative / pitch | Wave-1 draft in [docs/pitch/](docs/pitch/); all performance claims **[PENDING TRACK-1]** | -Week-0 census (**observability config**): **FAIL** — 2 survivors of 70; archived under -`docs/research/census-week0/`. +Two consecutive vertical FAILs — Week-0 observability config (2 survivors of 70, +archived under [docs/research/census-week0/](docs/research/census-week0/)) and Track-2 +counterparty (2 of 75, 0 DURABLE) — mean the company now rests on the Track-1 +mechanism number. See +[README-narrative.md §6](docs/README-narrative.md#6-where-evidence-stands-now). ## Public repo — write accordingly @@ -50,6 +53,12 @@ npm install npm run ci ``` +`npm run ci` is the one command that must be green before any PR: secret-scan, +contract validation, lint, typecheck, unit tests, then the end-to-end integration +test, in that order. The privacy canary (`npm run test:canary`) is merge-blocking +as a separate CI job. Full command list and the pre-PR checklist: +[docs/DEVELOPMENT.md](docs/DEVELOPMENT.md). + Contracts live in `contracts/`. Wave-1 agents build against those schemas — not against each other. @@ -59,10 +68,17 @@ against each other. contracts/ # JSON Schema — build against these src/testbed|recorder|compiler|runner|cache|metrics/ experiments/gate-v1/ # throwaway gate harness +scripts/ # secret-scan, contract validation, branch protection, testbed compose +tests/unit|canary|integration/ +artifacts/ # committed compiled bundles docs/ # map: docs/README.md archive/ # superseded scaffolds (Python hello) + preserved history ``` +How the six packages connect: [docs/architecture.md](docs/architecture.md). +Where the project is and what to pick up next: [docs/ROADMAP.md](docs/ROADMAP.md). +How to run, test, and ship a change: [docs/DEVELOPMENT.md](docs/DEVELOPMENT.md). + ## Stack TypeScript + Node 20+ + Playwright — see [ADR-0001](docs/decisions/ADR-0001-typescript-node-playwright.md). diff --git a/docs/INTEGRITY-AUDIT.md b/docs/INTEGRITY-AUDIT.md index ea53730..4447f0d 100644 --- a/docs/INTEGRITY-AUDIT.md +++ b/docs/INTEGRITY-AUDIT.md @@ -79,7 +79,7 @@ D2 consolidator pass (2026-07-25), **after** C4 + C5 merges. **Surfaces conflict | D-01 **FOUNDER** | Historical A7 / early Wave-1 “conditionally credible” seller-portal lead | C4 ALREADY_SOLVED + **C5 FAIL** (do not lock “because A7 said so”) | Adjudicated by C5; residual deck slides may still lag | **Critical** if old pitch used without C5 cite | | D-02 **FOUNDER** | PRD §8: Grafana Cloud + Datadog anchors | ADR-0003 OSS bed; A8/C5 reject those locks | Residual PRD body | **Critical** | | D-03 **FOUNDER** | Pitch “no gate number” / unset thresholds | PRD §9 proposed 80/90/~50/70 | Unset measured vs proposed | **High** | -| D-04 | Root README Track 3 “placeholders only” | Pitch pack + C5 on main | Stale status | Medium | +| D-04 | (was) Root README Track 2 “search in progress” / Track 3 “placeholders only” | C5 FAIL + ADR-0004; pitch pack on main | **Resolved** — status table now records Track 2 FAIL and Track 3 Wave-1 draft, each citing the artifact | Resolved in README; deck/objections prose may still lag (D-01, E-12) | | D-05 | Early D1 proof-points: PRD absent / C5 pending | Files + C5 FAIL on main | Stale diligence | Medium — register refreshed | | D-06 | PRD §8 “pivot if <~6 survivors” | A8/C5 executed ≤2 FAIL gates | Different fail lines | Medium — historical | | D-07 | (was) Objections denying two FAILs | A8 + C5 | **Updated in this PR** to present-tense two FAILs | Resolved in objections; deck may lag | diff --git a/docs/README.md b/docs/README.md index 4849998..bcfe019 100644 --- a/docs/README.md +++ b/docs/README.md @@ -38,7 +38,7 @@ Integrity surface: **[INTEGRITY-AUDIT.md](./INTEGRITY-AUDIT.md)**. | [architecture.md](./architecture.md) | How the six `src/` packages connect: loop diagram, package/artifact tables, real vs stubbed, invariants | draft | 2026-07-25 | Owns the **chain** between packages; `gate/*.md` own one hop each. Records two unwired hops (issue #52) | | [README-narrative.md](./README-narrative.md) | Story: thesis → census kill → pivot → two tracks → evidence now | draft | 2026-07-25 | Includes C5 FAIL | | [INTEGRITY-AUDIT.md](./INTEGRITY-AUDIT.md) | Unsourced claims, placeholders, LOW load-bearing, contradictions | draft | 2026-07-25 | Surfaces conflicts; does not pick winners | -| [../README.md](../README.md) | Repo root status | living | 2026-07-25 | May lag Track 3 / C5 — see INTEGRITY-AUDIT | +| [../README.md](../README.md) | Repo root status | living | 2026-07-25 (status table refreshed) | Track 2 **FAIL** / Track 3 Wave-1 draft now match C5 + ADR-0004; INTEGRITY-AUDIT D-04 resolved | | [../CONTRIBUTING.md](../CONTRIBUTING.md) | Hard rules + YAML frontmatter standard | living | 2026-07-24 | Doc-shape source of truth | --- @@ -140,7 +140,10 @@ precondition); M5 and M7 can run in parallel; M6 is blocked on the M4 gate numbe ## Open questions / what I could not verify -- Root README Track 2/3 status lines are still stale (INTEGRITY-AUDIT D-04) — tracked as issue #54. +- ~~Root README Track 2/3 status lines are still stale (INTEGRITY-AUDIT D-04) — tracked as issue #54.~~ + **Resolved** — the root README status table now records Track 2 as FAIL and Track 3 as a + Wave-1 draft, both citing in-repo artifacts. The residual deck/objections prose flagged by + D-01 and E-12 is a separate, still-open lag. - Founder post-Track-1 choice (intermediary reframe vs shut search) — C5 next action; not decided in docs. - Measured gate number — still unset; do not invent. - Whether to buy GitHub Secret Protection. Non-provider patterns and validity checks are