Adapt the task family
-- We map benchmark prompts, proxy tasks, and official fixtures into a repo-local runner - plan with harness versioning, model routes, budget limits, and restartable ledgers. -
-diff --git a/public/app.js b/public/app.js index e51b69a..d4f8786 100644 --- a/public/app.js +++ b/public/app.js @@ -46,6 +46,9 @@ const payload = requestPayload(); modeLabel.textContent = payload.mode; preview.textContent = JSON.stringify(payload, null, 2); + status.textContent = payload.target + ? "Request preview ready. Prepare request opens an email draft." + : "Fill in the target to generate a request."; return payload; } diff --git a/public/index.html b/public/index.html index f02b368..078c7f9 100644 --- a/public/index.html +++ b/public/index.html @@ -11,207 +11,325 @@
-Managed proof runs for agentic apps
-- Give us a live URL or a codebase. We adapt benchmark-style tasks to your product path, - run the app under proof gates, and return receipts that separate product proof from - official benchmark scores. -
-How it runs
-- We map benchmark prompts, proxy tasks, and official fixtures into a repo-local runner - plan with harness versioning, model routes, budget limits, and restartable ledgers. -
-- The agent harness checks the task path without multiplying every item through a - browser. This is where free or low-cost OpenRouter model sweeps belong. -
-- Browser runs prove the real user path: load the public URL, interact through stable - selectors, capture traces, and verify visible outcomes on the deployed app. +
Managed proof runs for agentic apps
++ Give us a live URL or a codebase. We adapt benchmark-style tasks to your product path, + run the app under proof gates, and return receipts that separate product proof from + official benchmark scores.
- -How it runs
++ Benchmark prompts, proxy tasks, and official fixtures become a repo-local runner + plan with harness versioning, model routes, budget limits, and restartable ledgers. +
++ The agent harness checks the task path before every item is multiplied through a + browser. Free or low-cost OpenRouter model sweeps belong in this lane. +
++ Browser runs prove the real user path: load the public URL, interact through stable + selectors, capture traces, and verify visible outcomes on the deployed app. +
++ Final reports label each result as product-path proof, proxy benchmark proof, or + official scorer output. No runner report can blur those categories. +
+Evidence
+Start
+- The final report labels each result as product-path proof, proxy benchmark proof, or - official scorer output. No runner report is allowed to blur those categories. + This form creates a scoped run request you can email. The hosted site does not collect + secrets, tokens, or repository credentials in the browser.
- -What you get
-Start
-- This form creates a scoped run request you can email. The hosted site does not collect - secrets, tokens, or repository credentials in the browser. -
-