staging <- dev

.github/workflows/release-mcp.yml

@@ -45,6 +45,11 @@ jobs:
       - name: Build MCP package
         run: pnpm --filter @mysten-incubation/memwal-mcp build
 
+      # Gate every publish on the integration suite (auth-required -> bridge
+      # hot-handoff, login callback). Never ship the MCP package without it.
+      - name: Test MCP package
+        run: pnpm --filter @mysten-incubation/memwal-mcp test
+
       # ── main branch → changeset version + stable release (latest) ──
       - name: Apply changesets (update version & CHANGELOG)
         if: github.ref == 'refs/heads/main'

.github/workflows/test.yml

@@ -32,6 +32,34 @@ jobs:
       - name: Check compatibility contract
         run: node scripts/check-compatibility-contract.mjs
 
+  mcp-tests:
+    name: MCP / Integration (login handoff)
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v4
+
+      - name: Setup Node
+        uses: actions/setup-node@v4
+        with:
+          node-version: "22"
+          cache: pnpm
+
+      - name: Install deps
+        run: pnpm install --frozen-lockfile
+
+      # Builds the package (tsc) then runs the node:test integration suite:
+      # spawns the real stdio MCP with empty creds, writes credentials
+      # mid-process, and asserts memwal_recall is served against a mock relayer
+      # without a restart (auth-required -> bridge hot-handoff).
+      - name: MCP integration tests
+        run: pnpm --filter @mysten-incubation/memwal-mcp test
+
   chatbot-e2e:
     name: Chatbot / Playwright E2E
     runs-on: ubuntu-latest

README.md

@@ -12,8 +12,7 @@ data they can trust.
 ## For AI Agents
 
 - **Single-file guide**: Read [`SKILL.md`](SKILL.md) for a complete integration reference (install, configure, API surface, troubleshooting)
-- **LLM-friendly docs**: [`llms.txt`](https://docs.wal.app/walrus-memory/llms.txt) — structured overview following the [llmstxt.org](https://llmstxt.org) standard
-- **Full context**: [`llms-full.txt`](https://docs.wal.app/walrus-memory/llms-full.txt) — expanded version with inlined page content
+- **LLM-friendly docs**: [`llms.txt`](https://docs.wal.app/llms.txt) — structured overview following the [llmstxt.org](https://llmstxt.org) standard
 
 ## Install