## Stack Context
This stack moves memory deduplication **out of the playground UI and into the SDKs themselves**, so every integration injects a single, deduplicated, self-replacing memory block. Three PRs:
1. **`sdk-dedup/tools-ts`** (this PR) — TypeScript SDK core + integrations
2. `sdk-dedup/python` — Python SDKs
3. `sdk-dedup/playground` — playground debug view reflects the SDK-owned block
## What?
Move profile deduplication into the SDK middleware for the TypeScript tools package.
- Facts are normalized (strip leading `[YYYY-MM-DD]`, trim, collapse whitespace, casefold) and deduplicated in **`static > dynamic > search`** priority within a single request.
- The result is injected as one **owned `<supermemory>` block** that *replaces* the previous block instead of accumulating a new one each turn.
- Dedup is **mode-aware**: in query mode, search results are not dropped against a profile that isn't being injected.
- Deduplication is **request-local** — no global/browser `Set`. Safe for multiple users, concurrent requests, and Cloudflare Worker isolates.
Covers AI SDK, OpenAI (Chat + Responses), Mastra, and VoltAgent. New `shared/memory-context.ts` owns the block-replacement logic.
## Why?
The earlier "conversation-scoped deduplication" was only a playground browser `Set` — a UI debug affordance that did not change what the SDK sent to the model, and would have been unsafe as server-side global state. Real cross-source dedup belongs in the SDK, applied fresh per stateless model request.
## Testing
- `bun run test` in `packages/tools`: 145 passed (the one failing suite, `claude-memory.test.ts`, is a pre-existing broken import unrelated to this change).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
<!-- CURSOR_SUMMARY -->
---
> [!NOTE]
> **Medium Risk**
> Changes how system prompts and instructions are built across all TypeScript integrations; behavior is well-covered by unit tests but incorrect strip/replace logic could drop or duplicate context in production prompts.
>
> **Overview**
> Moves **cross-source memory deduplication** and **owned prompt injection** into `@supermemory/tools` so every integration sends one deduplicated memory block per request instead of growing context each turn.
>
> **Deduplication:** Facts are normalized via `normalizeMemoryFact` (strip `[YYYY-MM-DD]`, trim, collapse whitespace, lowercase) and deduplicated with **static → dynamic → search** priority. `deduplicateMemoriesForMode` keeps search hits in **query** mode when the profile is not injected.
>
> **Owned `<supermemory>` block:** New `shared/memory-context.ts` wraps memories in `<supermemory context="user-memories" readonly>`, strips stale blocks, and **replaces** prior SDK context while preserving caller system instructions. Applied in AI SDK (`injectMemoriesIntoParams`), OpenAI Chat/Responses middleware, Mastra input processor (`wrapMemoryContext`), and VoltAgent hooks.
>
> **Tests:** Unit coverage for block replacement (with-supermemory, OpenAI, VoltAgent), Mastra wrapper tag assertion, normalized dedup variants, and concurrent `containerTag` isolation.
>
> <sup>Reviewed by [Cursor Bugbot](https://cursor.com/bugbot) for commit 2fa2e0d85c. Bugbot is set up for automated code reviews on this repo. Configure [here](https://www.cursor.com/dashboard/bugbot).</sup>
<!-- /CURSOR_SUMMARY -->
## Summary
- Add `apps/sdk-playground` — chat UI to test TS/Python SDK integrations
- Context panel with document memories, API keys in dashboard, tools reference tab
- Python FastAPI server on port 8792; portless entry in `portless.json`
Stacked on #1436
## Test plan
- [ ] `cd apps/sdk-playground && bun run check-types`
- [ ] `bun run dev` with Supermemory + OpenAI keys in UI
- [ ] Switch SDKs and verify chat + context panel
Made with [Cursor](https://cursor.com)
## Summary
- Re-export full tool set from `@supermemory/tools/ai-sdk`
- Add unit tests for tool re-exports
Stacked on #1432
## Test plan
- [ ] `bun run test:unit` in `packages/ai-sdk`
Made with [Cursor](https://cursor.com)
## Summary
- Refresh canonical tool descriptions in `tools-shared.ts`
- Align OpenAI and AI SDK tool bindings with 7-tool surface
- Export `TOOL_DESCRIPTIONS` / `PARAMETER_DESCRIPTIONS` from package index
Stacked on #1431
## Test plan
- [ ] `bun run test:unit` in `packages/tools`
Made with [Cursor](https://cursor.com)
Cherry-picks two cleanup PRs and finishes the job. Net 262 deletions.
- #1473 (@abhay-codes07): drops a `console.log` in the fullscreen note editor that printed the whole note body on every keystroke, which PostHog session replay can capture.
- #1563 (@ishaanxgupta): removes `/api/onboarding/research` and `/api/onboarding/extract-content`. Neither has a caller anywhere in the repo, and both spent metered Exa and xAI quota. This reverts the guards added for them in #1589, which only existed to make unreachable code safe.
- On top: `EXA_API_KEY`, `XAI_API_KEY` and the `@ai-sdk/xai` dependency are removed, since deleting those routes left them with no consumer.
Co-Authored-By: abhay-codes07 <182421137+abhay-codes07@users.noreply.github.com>
Co-Authored-By: ishaanxgupta <124028055+ishaanxgupta@users.noreply.github.com>
Cherry-picks four contributor PRs for `@supermemory/tools` onto one branch, and bumps the package to 2.2.0.
- #1244 (@rajarshidattapy): `withSupermemory` accepts `options.apiKey` instead of only reading `SUPERMEMORY_API_KEY`, matching the Vercel, Mastra and Voltagent integrations. Unblocks secrets managers, edge runtimes and per-request keys.
- #1574 (@Agnik47): re-exports `PromptTemplate`, `MemoryPromptData` and `WithSupermemoryOptions` from `ai-sdk`. `./vercel` is not a published subpath, so the documented custom-template example did not compile.
- #1488 (@abhinav7x94): malformed tool-call JSON returns an error result instead of throwing out of the request.
- #1507 (@abhinav7x94): VoltAgent `onEnd` awaits the conversation save, which was fire-and-forget and could be dropped when a serverless runtime tore down.
Dropped the `middleware.test.ts` added by #1244. Note that editing `packages/tools/package.json` triggers the npm publish workflow on merge.
Co-Authored-By: rajarshidattapy <138959719+rajarshidattapy@users.noreply.github.com>
Co-Authored-By: Agnik47 <140933190+Agnik47@users.noreply.github.com>
Co-Authored-By: abhinav7x94 <204053250+abhinav7x94@users.noreply.github.com>
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
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.
Consolidate login routes, add animated tool context visualization on the hero panel, and polish loading states across onboarding, settings, and workspace bootstrap.
## Summary
- Add a "Cancel subscription" button to the billing card for Pro and Scale users, alongside the existing "Manage billing" portal link.
- Cancellation is scheduled end-of-cycle via `autumn.updateSubscription({ planId, cancelAction: "cancel_end_of_cycle" })`, so users keep paid features until period end.
- Confirmation dialog surfaces remaining days from `useTokenUsage` and the correct plan name (Pro/Scale).
- Enterprise users do not see the cancel button — contract-based, portal/sales only.
Previously, the only path off a paid plan was Autumn's external customer portal, which contradicted the in-app "cancel anytime from the Billing tab" copy in the support FAQ.
## Summary
Addresses user feedback about the graph view: content in node popovers was hard-truncated at 100 characters with no way to read the full text or navigate to the document.
### Changes
**Scrollable popover content (no more truncation)**
- Removed the `truncate(content, 100)` call -- full content is now rendered in the popover
- Made the content area scrollable with `maxHeight: 100px` and `overflowY: auto`
- Increased version timeline entry truncation from 60 to 120 characters
- Increased version timeline container height from 120px to 160px
**"View document" in shortcuts panel**
- Added an eye icon + "View document" entry to the shortcuts panel (alongside existing arrow navigation)
- Clicking opens the Document Modal for the corresponding document
- Works for both document nodes (opens that document) and memory nodes (opens the parent document)
- Popover is automatically dismissed before the modal opens (prevents z-index overlap)
- Wired `onOpenDocument` callback through the full component chain: `page.tsx` -> `GraphLayoutView` -> `MemoryGraph` wrapper -> `MemoryGraphBase` -> `NodeHoverPopover`
**Shortcuts panel visual cleanup**
- Icon badges (`KeyBadge`) retain dark background and border for a keyboard-key badge look
- All icons (including the new EyeIcon) are uniformly wrapped in `KeyBadge` for consistent styling
- Removed background, border, and border-radius from the shortcuts panel container itself
- Widened shortcuts panel from 100px to 160px so navigation labels are fully visible
**Type safety improvement**
- Changed `MemoryGraphProps.colors` from `GraphThemeColors` to `Partial<GraphThemeColors>` to match actual usage (callers pass partial overrides that get merged with defaults)
- Replaced `as Partial<typeof import(...)>` cast with proper `satisfies Partial<GraphThemeColors>` using a direct type import
### Files changed
- `packages/memory-graph/src/components/node-hover-popover.tsx` -- scrollable content, EyeIcon + "View document" in shortcuts, KeyBadge styling, NavButton icon type broadened to `React.ReactNode`, all icons wrapped in KeyBadge
- `packages/memory-graph/src/components/memory-graph.tsx` -- popover dismissal wrapper (`handleOpenDocument`), prop forwarding
- `packages/memory-graph/src/types.ts` -- `onOpenDocument` prop on `MemoryGraphProps`, `colors` changed to `Partial<GraphThemeColors>`
- `apps/web/components/graph-layout-view.tsx` -- accepts and passes `onOpenDocument`
- `apps/web/components/memory-graph/memory-graph-wrapper.tsx` -- `onOpenDocument` in wrapper props, clean type import
- `apps/web/app/(app)/page.tsx` -- `handleOpenDocumentById` handler, wired to `GraphLayoutView`
## Testing
### Unit Tests (154/154 passed)
- **Command:** `cd packages/memory-graph && bun run test`
- **Result:** 9 test files, 154 tests passed (107 existing + 47 new)
- **New test file:** `packages/memory-graph/src/__tests__/node-hover-popover.test.tsx` covering:
- Layout constants (`SHORTCUTS_W=160`, `CARD_W=280`, `TOTAL_W` formula)
- Content area scrollability (`maxHeight:100`, `overflowY:auto`, `flex:1 1 auto`, no truncation)
- KeyBadge styles (has `backgroundColor: colors.controlBg`, has `border` with `colors.controlBorder`)
- Shortcuts panel styles (no `backgroundColor`, no `border`, no `borderRadius`)
- EyeIcon SVG component (`aria-hidden`, viewBox, path, circle, stroke)
- "View document" button render guard (`onOpenDocument && documentId`)
- `documentId` derivation (memory vs document node)
- VersionTimeline truncation limit (120 chars) and container maxHeight (160)
- `onOpenDocument` prop wiring
- NavButton icon prop type (`React.ReactNode`), all icons wrapped in KeyBadge
- `handleOpenDocument` wrapper in MemoryGraph (dismiss calls, ordering, useCallback deps)
### Build
- **Command:** `bun run --filter=@supermemory/memory-graph build`
- **Result:** Build succeeds, 0 errors
### Lint/Format
- **Command:** `bunx biome ci --changed --since=origin/main --no-errors-on-unmatched`
- **Result:** 6 files checked, no fixes applied, 0 errors
### Type Checking
- `packages/memory-graph`: 0 errors
- `apps/web`: 0 new errors introduced
### CI
- "CI - Type Check, Format & Lint" workflow passes
- "Claude Code Review" workflow fails due to repo config (`non-human actor: vorflux` not in `allowed_bots` list) -- not related to code changes
---
**Session Details**
- Session: [View Session](https://supermemory.us1.vorflux.com/agent-sessions/d0c067f1-38c1-40c5-ada9-b2ffb7331d1c)
- Requested by: Mahesh Sanikommu (mahesh@supermemory.com)
- Address comments on this PR. Add `(aside)` to your comment to have me ignore it.
- Switch to infinite query with viewport-triggered pagination (loads more when user zooms out 3x past node bounds)
- Remove maxNodes cap so all data renders
- Remove background color and dot pattern from graph
- Make document-memory edges light grey
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Fixed memory graph stuck on "Loading graph data..." caused by field name mismatch after package rewrite (`memoryEntries` vs `memories`, `type` vs `documentType`)
- Improved MCP app UI to match console-v2 design: golden derives edges (#FBBF24), edge glow pass, node hover effects, dot mode at low zoom, vertical controls with keyboard shortcuts, expandable legend
- Synced local `memory-graph` edgeDerives constant with published npm version
## Test plan
- [ ] Run `cd apps/mcp && bun run dev`, connect via Claude Desktop
- [ ] Verify graph renders with golden derives edges and blue dashed extends edges
- [ ] Verify node hover/selection glow effects
- [ ] Verify Fit (Z), Center (C), zoom (+/-) keyboard shortcuts
- [ ] Verify legend expands/collapses
The subscription check on plugins, connections, and usage pages was hardcoded to only look at `api_pro`. Users on scale or enterprise plans saw "Upgrade to Pro" and couldn't create plugin keys or manage connections.
Also, the check only looked at product `status` but not the `allowed` flag from autumn — so users who downgraded but were still in their billing period got locked out early.
