### TL;DR
Added new digest feature images and wordmark asset to the web app's public directory.
### What changed?
Five new images were added under `apps/web/public/images/digest/`:
- `feat-memory.png`
- `feat-profiles.png`
- `feat-retrieval.png`
- `feat-router.png`
- `wordmark.png`
### How to test?
Verify the images are accessible via their public URLs (e.g., `/images/digest/feat-memory.png`) and render correctly wherever they are referenced in the application.
### Why make this change?
These assets are needed to support the digest feature showcase, providing visual representations of the memory, profiles, retrieval, and router features, along with a wordmark for branding purposes.
## What
Makes the MCP `recall` tool render memories the same way the console **playground** does — rich, scored, relation-aware output — instead of the old flat `### Memory N (NN% match)` list.
The playground pattern is: **profile + rich search → one shared `formatMemories` formatter → feed to the model.** This ports that into the standalone MCP server.
## Changes
- **`format.ts`** (new) — `formatMemories`, lifted verbatim from the playground. Pure, dependency-free. Renders similarity scores, `agg`/`chunk` markers, related-memory arrows (`←` parent, `→` child, `~` related), attached document summaries, and temporal context.
- **`client.ts`** — `SupermemoryClient.search()` now accepts the playground's retrieval knobs (`searchMode` / `rerank` / `rewriteQuery` / `include`) and **stops discarding** the rich response fields (`context`, `documents`, `metadata`, `isAggregated`) during normalization, so the formatter has data to render.
- **`server.ts` (`handleRecall`)** — retrieves memories via `client.search()` with `include: { documents, relatedMemories }`, prepends the user's profile (stable + recent facts) when `includeProfile` is set, and renders memories through `formatMemories`. Output capped at `MAX_RECALL_CHARS`.
## How it maps to the playground
| | Playground | MCP (this PR) |
|---|---|---|
| Profile | injected into its own system prompt (`<user_profile>`) | prepended to `recall` output / exposed via `context` prompt |
| Search | hook pre-fetch **or** tool | `recall` tool (model-triggered) |
| Formatting | `formatMemories` → system prompt | `formatMemories` → `recall` tool result |
## Testing
Verified end-to-end against a local API + local MCP, driven through a real MCP client (and Claude Desktop). `recall` now returns scored memory blocks with `Source:`/`Document:` lines and the legend header. Existing e2e assertions remain compatible (they match `## User Profile` / `## Relevant Memories`, both still emitted).
## Notes
- One deliberate gap vs the playground: `aggregate` — the public SDK's `search.memories` doesn't expose it, so no `agg` synthesis blocks. Relation arrows render only when memories actually have linked parent/child versions.
- Possible follow-up: expose the `include` / `rerank` knobs as `recall` tool params so callers can toggle relations/documents per call.
- Pre-existing `tsc` errors in `server.ts` (dual `@modelcontextprotocol/sdk` install) and `mcp-app.ts` (DOM `lib`) are unrelated to this change — confirmed by an unchanged baseline.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---
**Session Details**
- Session: [View Session](https://supermemory.us1.vorflux.com/agent-sessions/c616fc88-a725-4961-9011-712bce740601)
- Requested by: Sreeram Sreedhar (sreeram@supermemory.com)
- Address comments on this PR. Add `(aside)` to your comment to have me ignore it.
- "What to remember" per-space context — set on create, on edit, and in the space profile; steers what gets extracted into memory
- Edit-space modal (name + context) from the space pill's hover edit, mirroring the create modal
- Manual saves respect a space's configured context instead of overwriting it with the generic default
Fixes the onboarding source-integrations step: corrects connector behavior, applies plan-based gating to connectors, and polishes the integrations layout.
Remove intermediate use-case screen; Apple Shortcuts card now opens a single modal with use cases + Add/Search memory shortcut buttons inline. Fixes res.data.key bug (was reading res.key which is always undefined). Buttons stack on mobile, no 'I'm good' dismiss.
## What changed
- Added a chat-shell-level file drag/drop target so the whole chat area, including the conversation area, shows the drop overlay and accepts files.
- Disabled the nested composer drop handler for the sidebar chat path to avoid duplicate drops or stuck overlay state.
- Kept the existing local input drop behavior available for other ChatInput usages like the home composer.
Adds .html/.htm extensions and text/html MIME to the upload accept list and validation. The API already ingests HTML files; the uploader was rejecting them client-side.
## Summary
- Replace the memory graph page's top-left package loading indicator with a centered Supermemory loader overlay.
- Reuse the same SuperLoader animation in the graph preview card loading state.
## Summary
- Allow logged-out users to load `/?view=integrations` without being redirected to login
- Add a public integrations header with Supermemory branding and a login CTA
- Run IntegrationsView in guest mode with static catalog data only, no auth-backed queries
- Route all guest Connect actions to login with a redirect back to the integrations page
<img width="1680" height="1009" alt="Screenshot 2026-06-09 at 9 32 33 PM" src="https://github.com/user-attachments/assets/111f2b1a-5224-47ae-8152-e1349c05842d" />
feat(web): make brain onboarding the default flow
- Replace legacy /onboarding with the brain onboarding flow (about -> sources -> ingest -> team -> done)
- Add brain-home dashboard page and widgets
- Old /onboarding/brain deep links fall through to the existing catch-all redirect
feat(onboarding): create org on About step + remove skip
- Create the organization on About -> Continue (authClient.organization.create + setActiveOrg), persisting brain context to org metadata; fixes the no-org redirect loop
- Show a Creating… state on Continue while the org is created
- Remove Skip on the required About step (name + workspace)
- Add generateOrgSlug/generateUsername helpers to onboarding-brain/types
- New Poke card in Apps & extensions linking to the Poke recipe (OAuth, no manual MCP URL)
- Featured hero slot for Poke with palm-tree emoji
- Reusable NEW badge, applied to Poke
- Poke logo asset, full-cover icon styling
- POKE_RECIPE_URL constant; guard Chrome-only analytics on external cards
- Added attachment draft/shared types and validation in components/chat/attachments.ts.
- Added paperclip upload UI to Nova composer with file chips, size/status, remove, retry, and per-file Save / Chat only toggle.
- Wired uploads before send to /chat/attachments, then sends returned attachment references in chat message metadata.
- Preserved attachment metadata when loading threads and rendered attachment chips on user messages.
- Added attachment support from the home composer into the full chat view.
- Extended chat analytics with attachment counts.
<img width="1905" height="900" alt="image" src="https://github.com/user-attachments/assets/631001b8-7c68-4015-b36b-06c69cdad271" />
<img width="1323" height="1600" alt="image" src="https://github.com/user-attachments/assets/77eee08a-b235-41eb-b03a-f406b81b46e7" />
- ensured responsiveness
End-to-end tests that drive the deployed Supermemory MCP over streamable HTTP (no mocks):
- discovery: handshake + tool/resource/prompt listing
- identity: whoAmI, listProjects
- memory: save -> recall round-trip, profile variants, forget, container scoping
- graph/resources/prompts: memory-graph, fetch-graph-data, profile/projects, context
- oauth: protected-resource discovery chain, dynamic client registration,
token-endpoint negatives, and a real refresh -> access token round-trip
- auth: GET / info, 401 on missing/invalid token
- root-scope: x-sm-project scoping behavior
Tests skip without SUPERMEMORY_API_KEY (OAuth tier-D skips without its token env vars) so CI is safe without secrets. Adds vitest + a test:e2e script and documents running + the OAuth flow in apps/mcp/README.md.
Clicking stop before the first assistant token streamed in deleted the triggering user message: handleStop unconditionally sliced off a trailing user message, which during the submitted/thinking window is the user's own
message. Drop the slice so the message is preserved on stop.
ENG-732
Add a Current/All scope toggle (default Current) to Nova chat history; All fetches threads across every space via the new scope=all API, with each row showing its space.
Redesign the sheet to fit the design system: cap width at 380px (was ~92vw), add search, date-grouped rows (Today/Yesterday/Previous 7/30 days/Older), hover-reveal delete, header space chip, solid primary New Conversation button.
Part of ENG-728: https://linear.app/supermemory/issue/ENG-728
Integrate the new api_max plan ($100/mo, $130 credits) from mono into the
nova web app, mirroring mono's tier semantics.
- Tiers: insert api_max between pro and scale; resolves to Pro-tier for feature gates while displaying as "Max" (queries.ts, use-token-usage)
- Billing page: Max plan card ("Most Popular"), checkout/cancel/invoice wiring; advanced carousel page now 3-up (Max/Scale/Enterprise); Scale builds on Max
- Carousel auto-opens to the page holding the current plan (Max/Scale/ Enterprise users land on their own card instead of Free+Pro)
- Cancel: retention dialog listing what you lose + type-to-confirm "CANCEL"
- Cancel/resume state mirroring console: detect canceledAt-scheduled cancellation, show "Cancelling" pill + "Cancels on {date} · N days left", and a "Resume plan" (uncancel) action
