From e415404a54d508dcd39967554d2655480a0b098d Mon Sep 17 00:00:00 2001 From: chris harris <283105756+Chris-ObtuseAI@users.noreply.github.com> Date: Sun, 26 Jul 2026 16:45:37 -0500 Subject: [PATCH] Standardize portfolio presentation --- README.md | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 41590b8..fbc67e5 100644 --- a/README.md +++ b/README.md @@ -2,7 +2,7 @@

Evidence-gated prediction-market intelligence for crypto & sports.
- An autonomous organism that gathers point-in-time public evidence, calibrates + A governed research system that gathers point-in-time public evidence, calibrates competing forecasts, earns every source its trust from settled outcomes, and explains and records each paper decision in an auditable ledger.

@@ -11,7 +11,7 @@ Python 3.11, 3.12, 3.14 More than 5,000 maintained tests collected Public release v1.0.0 - 45 autonomous loops + 45 governed research loops Shadow paper mode Human-gated Fail-closed @@ -50,6 +50,24 @@ --- +## Product truth + +| Contract | Current public state | +| --- | --- | +| **Portfolio role** | **Prediction-Market Intelligence** — calibrated forecasting, paper twins, settlement learning, and human-gated authority across crypto and sports. | +| **Maturity** | **Public-source research release** | +| **Engineering evidence baseline** | [`caffe3db`](https://github.com/ObtuseAI/dummy/commit/caffe3db1767847b70ee5f2cc3a5b7089fe693a8) · configured [tests](https://github.com/ObtuseAI/dummy/actions/workflows/tests.yml), supply-chain workflow, and [live product presentation](https://obtuseai.github.io/dummy/) | +| **Proved now** | The public release implements point-in-time evidence intake, competing forecasts, calibration, settlement grading, paper allocation, risk gates, replay, and an artifact-backed operator board. | +| **Authority ceiling** | Paper-only and human-gated. No forecast, model vote, dashboard state, or research loop has capital, credential, broker, deployment, or execution authority. | +| **Clean demonstration** | Use the [operator quickstart](#operator-quickstart) or inspect the hosted presentation and committed artifact views. | +| **Known limit** | Current launch status is `NO-GO`: elapsed retention, backup, canary, grading, execution-policy, and kill-drill evidence remain incomplete. Public statistics must remain bound to their named data and commit. | + +**Designed for:** prediction-market, sports, and crypto research teams that need forecast competition, calibration, settlement learning, and risk firewalls to remain inspectable. + +[Explore the ObtuseAI portfolio](https://github.com/ObtuseAI) · +[Open the live presentation](https://obtuseai.github.io/dummy/) · +[Start a technical conversation](https://github.com/ObtuseAI/dummy/issues) + > **Current launch status: NO-GO.** The hardened code path is implemented, but > live operations still fail the retention/WAL/deadline gates and lack the > required elapsed backup, canary, grading, execution-policy, and kill-drill