The read_image poisoning fix had no end-to-end guard because scenarios could
only send text. Add an upload_images step that renders and uploads visually
distinct images, let chat steps attach them the same way the WeChat channel
does, and assert the generated page actually embeds every upload.
Co-authored-by: Cursor <cursoragent@cursor.com>
Image turns already get a vision-model description injected into the prompt,
but the agent kept calling read_image to "confirm" the pictures. Those tool
results carry base64 image parts that Goose persists, so every later turn
against the text-only chat provider failed with `unknown variant image_url`
before the agent could write the page. WeChat page requests therefore fell
through to the fail-closed delivery message.
Drop read_image for the turn whenever a vision model handles the images, say
so explicitly in the injected prompt, and teach the poison scan to recognise
tool image parts so already-polluted sessions rotate instead of failing again.
Co-authored-by: Cursor <cursoragent@cursor.com>
Drop the user_confirmed_at gate so public, online, unexpired pages can enter sitemap/llms and receive SEO/GEO tags. Defaults now enable all discovery switches; stored all-off config is still preserved until admin saves.
Co-authored-by: Cursor <cursoragent@cursor.com>
Centralize page HTML quota checks with grace-write semantics across page
services and workspace tools, keep localhost MindSpace links clickable in
chat display, and expand Page Data/static-page skill plus local dev docs
for quota, delivery URLs, and native Aider/OpenHands tooling.
Co-authored-by: Cursor <cursoragent@cursor.com>
Map Page Data failure codes to Chinese remediation hints, trigger one-shot
goosed repair for remediable cases, and recover poisoned thinking sessions.
Route local DeepSeek through the no-think proxy via host.docker.internal so
tool rounds no longer hit reasoning_content 400; document gate and case-study
scenarios for event registration repair.
Co-authored-by: Cursor <cursoragent@cursor.com>
Ensure page sync runs via pageSyncService in remote mode, fall back to
workspace HTML when storage assets are missing, and prevent production
Portal from linking to loopback Plaza URLs baked in at build time.
Co-authored-by: Cursor <cursoragent@cursor.com>
john