Open-source platform for 3D AI agents. Turn text or a photo into a rigged, animated GLB avatar, give it an LLM brain, memory and a wallet, and embed it anywhere with one web component. Remote MCP server, x402 payments, Three.js, browser-native.
JavaScript 3d-generationai-agentsanimationavatardigital-humangenerative-aiglbgltfimage-to-3dllmmcpmcp-servermodel-context-protocolsolanatext-to-3dthreejsweb-componentswebglwebxrx402
docs(production-100): record the Playwright result and the ninth red it turned up
nirholas362f5a01edb2
diff --git a/prompts/finish/_context/production-100-PROGRESS.md b/prompts/finish/_context/production-100-PROGRESS.mdindex 4e8275895..f16402352 100644--- a/prompts/finish/_context/production-100-PROGRESS.md+++ b/prompts/finish/_context/production-100-PROGRESS.md@@ -743,8 +743,19 @@ secrets across 2424 reachable modules, `audit:deploy` clean). `server/cloudbuildthree failures were `Test timed out in 120000ms` in two forge suites, under load average 30with 2036 files in flight; both files pass in 19s when run alone. The first run of the samesuite reported 18 failures and every one was the worktree's environment, not the commit,-which is what the prep:worktree work below fixes. Playwright was left to a window with no-peer run in flight.+which is what the prep:worktree work below fixes.++**Playwright: 257 passed, 0 failed, 5 skipped, exit 0** (10.3 min), after one more fix. The+first run reported three failures, all in `home-lifecycle` and `home-plan`. Neither is a+product defect: both are live journeys that need a real Home Assistant and two provisioned+accounts, which only `playwright.home.config.js` sets up, and both had drifted out of the+default config's `testIgnore` list. The general run drove them with no house to reach, so+journey 1 died on `Could not reach http://127.0.0.1:39611` and the plan page found none of+its quota rows. **`npm test` could not pass for anyone while that held.** The config states+the rule in prose right above the list (importing `./home-support.js` is what makes a spec+live), so the list is now read from the spec files instead of being that rule written a+second time: eight live specs, the original six plus the two that drifted. Both still run+under `npm run test:home:e2e`, so nothing loses coverage.**Eight reds fixed, all of them at HEAD and none of them mine:**