chore: update version to 0.21.3 for v0.21.3 release

Signed-off-by: Drew Cain <groksrc@gmail.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

.github/workflows/release.yml

@@ -63,25 +63,63 @@ jobs:
     # Only run for stable releases (not dev, beta, or rc versions)
     if: ${{ !contains(github.ref_name, 'dev') && !contains(github.ref_name, 'b') && !contains(github.ref_name, 'rc') }}
     permissions:
-      contents: write
-      actions: read
+      contents: read
     steps:
+      # Inline bump replaces mislav/bump-homebrew-formula-action@v4.x.
+      # The action does a HEAD request to api.github.com /repos/.../tarball/<ref>
+      # with the bearer token and expects a 302 redirect. GitHub now returns
+      # 303 on that endpoint when authenticated, which the action treats as a
+      # fatal error. Re-implementing the bump as plain git+sed keeps the same
+      # contract (update url + sha256, commit, push) with no third-party action.
       - name: Update Homebrew formula
-        uses: mislav/bump-homebrew-formula-action@v4.1
-        with:
-          # Formula name in homebrew-basic-memory repo
-          formula-name: basic-memory
-          # The tap repository
-          homebrew-tap: basicmachines-co/homebrew-basic-memory
-          # Base branch of the tap repository
-          base-branch: main
-          # Download URL will be automatically constructed from the tag
-          download-url: https://github.com/basicmachines-co/basic-memory/archive/refs/tags/${{ github.ref_name }}.tar.gz
-          # Commit message for the formula update
-          commit-message: |
-            {{formulaName}} {{version}}
-
-            Created by https://github.com/basicmachines-co/basic-memory/actions/runs/${{ github.run_id }}
         env:
-          # Personal Access Token with repo scope for homebrew-basic-memory repo
-          COMMITTER_TOKEN: ${{ secrets.HOMEBREW_TOKEN }}
+          HOMEBREW_TOKEN: ${{ secrets.HOMEBREW_TOKEN }}
+          REF: ${{ github.ref_name }}
+          REPO: ${{ github.repository }}
+          RUN_URL: https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}
+        run: |
+          set -euo pipefail
+
+          VERSION="${REF#v}"
+          ARCHIVE_URL="https://github.com/${REPO}/archive/refs/tags/${REF}.tar.gz"
+
+          echo "::group::Compute tarball sha256"
+          SHA256="$(curl --fail --silent --location "$ARCHIVE_URL" | sha256sum | awk '{print $1}')"
+          test -n "$SHA256"
+          echo "sha256: $SHA256"
+          echo "::endgroup::"
+
+          echo "::group::Clone tap"
+          git clone \
+            --depth 1 \
+            "https://x-access-token:${HOMEBREW_TOKEN}@github.com/basicmachines-co/homebrew-basic-memory.git" \
+            tap
+          cd tap
+          git config user.name "github-actions[bot]"
+          git config user.email "41898282+github-actions[bot]@users.noreply.github.com"
+          echo "::endgroup::"
+
+          echo "::group::Patch Formula/basic-memory.rb"
+          # Pipe-delimited sed because the URL contains slashes. The Formula
+          # only has one `url` and one `sha256` directive, so a first-match
+          # replacement is unambiguous. POSIX character classes ([[:space:]])
+          # keep this portable across BSD and GNU sed.
+          sed -i -E \
+            -e "s|^([[:space:]]*url[[:space:]]+)\"[^\"]+\"|\1\"${ARCHIVE_URL}\"|" \
+            -e "s|^([[:space:]]*sha256[[:space:]]+)\"[^\"]+\"|\1\"${SHA256}\"|" \
+            Formula/basic-memory.rb
+          git --no-pager diff Formula/basic-memory.rb
+          echo "::endgroup::"
+
+          if git diff --quiet Formula/basic-memory.rb; then
+            echo "Formula already at ${REF}; nothing to do."
+            exit 0
+          fi
+
+          echo "::group::Commit & push"
+          git add Formula/basic-memory.rb
+          git commit -m "basic-memory ${VERSION}
+
+          Created by ${RUN_URL}"
+          git push origin HEAD:main
+          echo "::endgroup::"

AGENTS.md

@@ -56,6 +56,15 @@ Run `just test-smoke` when you specifically need the MCP smoke flow.
 
 If testmon is “cold,” the first run may be long. Subsequent runs get much faster.
 
+### PR CI Gate
+
+Before opening or updating a PR, run the checks that mirror the common required CI failures:
+
+- Run `just typecheck` in addition to targeted `ruff` and `pytest` commands when tests were added or changed.
+- Sign commits with `git commit -s` so DCO passes. If a PR branch already has unsigned commits, rewrite the branch with signed-off commits before asking for review.
+- Use a semantic PR title accepted by `.github/workflows/pr-title.yml`: `type(scope): summary`.
+- Use one of the allowed scopes: `core`, `cli`, `api`, `mcp`, `sync`, `ui`, `deps`, `installer`.
+
 ### Test Structure
 
 - `tests/` - Unit tests for individual components (mocked, fast)

CHANGELOG.md

@@ -1,6 +1,24 @@
 # CHANGELOG
 
-## Unreleased
+## v0.21.1 (2026-05-16)
+
+CI-only release. No user-facing changes.
+
+### Maintenance
+
+- **#833**: Replace `mislav/bump-homebrew-formula-action` with an inline
+  bash bump step. The action's `resolveRedirect` HEAD-requests
+  `api.github.com /repos/.../tarball/<ref>` expecting HTTP 302; GitHub now
+  returns 303 for authenticated requests on that endpoint, which broke
+  the v0.21.0 Homebrew job and required a manual tap bump. The inline
+  step does the same work (curl + sha256sum, clone tap, sed-bump
+  `url`/`sha256`, commit, push) without any third-party dependency.
+
+## v0.21.0 (2026-05-16)
+
+Workspace-aware everywhere: every MCP tool and CLI command now routes through the
+same workspace/project model, the search and sync paths are noticeably faster,
+and a handful of long-standing parsing and routing footguns are gone.
 
 ### Breaking Changes
 
@@ -11,6 +29,109 @@
   - Bare list wikilinks like `- [[Target]]` now index as `links_to`.
   - Prose list items like `- some other thing [[Target]]` now index as `links_to`.
   - To preserve existing multi-word relation types on re-sync, quote them before upgrading.
+  - See **#824**.
+
+### Features
+
+- **#816**: `bm orphan` CLI command surfaces entities whose underlying markdown
+  files are gone, with a flag to clean them out.
+- **#789**: Create projects directly by cloud workspace slug from MCP
+  (`create_memory_project(workspace=...)`).
+- **#757**: Discover projects across every accessible cloud workspace in MCP's
+  project list — no more per-workspace blind spots.
+- **#766**: MCP tools accept training-data-friendly parameter aliases
+  (`q`/`search`/`text` for `query`, etc.) so models reach for them naturally.
+- **#776**: `bm db reset` refuses to run while a `basic-memory mcp` process is
+  alive, so resets can no longer corrupt an open session.
+- **#791**: Search responses include result totals so pagers can stop guessing.
+- **#719**: Cloud `note_content` tenant schema primitive lands on the backend.
+- **#715**: `bm project add` accepts a `--visibility` flag for cloud projects.
+
+### Bug Fixes
+
+#### Search and recent activity
+- **#832**: SQLite project deletion now sweeps `search_index`,
+  `search_vector_chunks`, and `search_vector_embeddings`, so a project that
+  reuses a recycled auto-increment id can't inherit the previous tenant's content.
+- **#812**: `recent_activity` orders and filters by `updated_at`, so edits bubble
+  to the top instead of staying pinned to creation time.
+- **#807**: Multi-project `search_notes` is opt-in (`search_all_projects=True`);
+  default search stays scoped to the resolved project.
+- **#785**: `recent_activity` caps responses and emits an explicit truncation
+  footer instead of silently dropping rows.
+- **#713**: Eliminated an N+1 query in search result hydration.
+
+#### Workspace / project routing
+- **#822**: `bm project list` now includes projects from every workspace, not
+  just the current one.
+- **#813**, **#808**, **#806**, **#803**, **#801**, **#795**, **#790**, **#778**,
+  **#777**, **#722**, **#712**, **#704**: Workspace-qualified permalink routing
+  is centralized and applied consistently across `edit_note`, `delete_project`,
+  `build_context`, `memory://` URLs, factory-mode project listing, cloud uploads,
+  and the API client.
+
+#### Sync
+- **#827**: `rclone bisync` filters from `.bmignore` are preserved across syncs.
+- **#815**: Watch service ignores hidden paths relative to the watched project,
+  not just relative to `$HOME`.
+- **#814**: `scan` subprocesses no longer go through the shell, avoiding quoting
+  issues with paths that contain special characters.
+- **#759**: Watch service stays inside `--project` scope.
+- **#746**: Canonical markdown is preserved during single-file sync.
+
+#### Parsing
+- **#796**: Picoschema modifier descriptions (`field?(modifier, description)`)
+  parse correctly.
+- **#769**: Obsidian callout blocks are skipped by the observation parser
+  instead of being mis-extracted as observations.
+
+#### CLI
+- **#775**: `bm project set-cloud` / `set-local` cleans up local DB state for
+  the affected project.
+- **#773**: `bm cloud logout` clears `default_workspace`.
+- **#780**: `bm cloud setup` hint points at `bm cloud sync-setup`.
+- **#734**: `bm project info` shows cloud index freshness.
+- **#718**: Private cloud projects display under their `display_name` instead of
+  raw UUID.
+- **#768**: `read_note` / `view_note` drop no-op pagination params from their
+  signatures.
+
+#### Stability
+- **#774**: `sqlite-vec` failures during init degrade gracefully to keyword-only
+  search instead of crashing startup.
+- **#733**: Delete-vector and cloud-sync cleanup is now consistent.
+- **#702**: Race conditions in concurrent `delete_entity` are resolved.
+- **#724**: `external_id` is preserved when entities are re-upserted during a
+  re-index.
+- **#728**: Vector init no longer issues runtime `ALTER TABLE`.
+- **#744**: `BASIC_MEMORY_CONFIG_DIR` is honored across remaining call sites.
+- **#743**: FastEmbed cache lives under the data dir instead of `/tmp`.
+- **#752**: Cloud projects report `source=cloud` in factory mode.
+- Stripped null bytes from markdown content before DB insert.
+- Allowed long `relation_type` values in API responses.
+
+#### Installer
+- **#772**: Docker-compose config volume mounts under the `appuser` home.
+- **#695**: Bumped `brew outdated` timeout from 15s to 60s.
+
+### Performance
+
+- **#828**: CLI startup no longer pulls in the local ASGI FastAPI app when it
+  isn't needed.
+- **#751**, **#726**, **#717**, **#714**, single-file/batch indexing: marked
+  speedups on the sync hot path; unchanged markdown is skipped entirely.
+- **#731**, **#723**: Vector sync is faster on both backends, with tuned
+  fastembed defaults.
+
+### Maintenance
+
+- **#825**: Dependency refresh + security hardening.
+- Updated to `fastmcp` 3.3.1.
+- **#754**: Removed in-house telemetry wrappers in favor of direct `logfire`
+  usage.
+- **#736**: `ty` is now the default typechecker.
+- **#771**, **#770**, **#716**: New regression guards for vector-row cleanup,
+  long relation types, and recent activity hydration.
 
 ## v0.20.3 (2026-03-26)
 

README.md

@@ -123,8 +123,7 @@ phone.
   time — same files, same format, same wikilinks. Cancel anytime, your data
   stays yours.
 
-Built on WorkOS AuthKit, Neon Postgres, and Tigris S3. Platform source at
-[basic-memory-cloud](https://github.com/basicmachines-co/basic-memory-cloud).
+Built on WorkOS AuthKit, Neon Postgres, and Tigris S3.
 
 ### Pricing
 

docker-compose-postgres.yml

@@ -1,5 +1,8 @@
 # Docker Compose configuration for Basic Memory with PostgreSQL
-# Use this for local development and testing with Postgres backend
+# Use this for local development and testing with Postgres backend.
+#
+# The Postgres backend requires the pgvector extension (semantic search).
+# This image bundles pgvector; plain postgres:17 will not work for vector search.
 #
 # Usage:
 #   docker-compose -f docker-compose-postgres.yml up -d
@@ -7,7 +10,7 @@
 
 services:
   postgres:
-    image: postgres:17
+    image: pgvector/pgvector:pg17
     container_name: basic-memory-postgres
     environment:
       # Local development/test credentials - NOT for production

docs/semantic-search.md

@@ -263,6 +263,7 @@ The sqlite-vec extension is loaded per-connection. Vector tables are created laz
 ### Postgres (cloud)
 
 - **Vector storage**: [pgvector](https://github.com/pgvector/pgvector) with HNSW indexing
+- **Local Docker**: use `docker-compose-postgres.yml` (`pgvector/pgvector:pg17`). Plain `postgres:17` lacks the extension; run `CREATE EXTENSION IF NOT EXISTS vector;` on any external instance before first migration.
 - **Chunk metadata table**: Created via Alembic migration (`search_vector_chunks` with `BIGSERIAL` primary key)
 - **Embedding table**: `search_vector_embeddings` created at runtime (dimension-dependent, same pattern as SQLite)
 - **Index**: HNSW index on the embedding column for fast approximate nearest-neighbour queries

server.json

@@ -6,12 +6,12 @@
     "url": "https://github.com/basicmachines-co/basic-memory.git",
     "source": "github"
   },
-  "version": "0.20.3",
+  "version": "0.21.3",
   "packages": [
     {
       "registryType": "pypi",
       "identifier": "basic-memory",
-      "version": "0.20.3",
+      "version": "0.21.3",
       "runtimeHint": "uvx",
       "runtimeArguments": [
         {"type": "positional", "value": "basic-memory"},

src/basic_memory/__init__.py

@@ -1,7 +1,7 @@
 """basic-memory - Local-first knowledge management combining Zettelkasten with knowledge graphs"""
 
 # Package version - updated by release automation
-__version__ = "0.20.3"
+__version__ = "0.21.3"
 
 # API version for FastAPI - independent of package version
 __api_version__ = "v0"

src/basic_memory/alembic/versions/n7i8j9k0l1m2_cleanup_sqlite_search_orphans.py

@@ -0,0 +1,86 @@
+"""Remove orphaned search rows whose project was already deleted.
+
+Revision ID: n7i8j9k0l1m2
+Revises: m6h7i8j9k0l1
+Create Date: 2026-05-15 18:30:00.000000
+
+"""
+
+from typing import Sequence, Union
+
+from alembic import op
+from sqlalchemy import inspect
+
+
+revision: str = "n7i8j9k0l1m2"
+down_revision: Union[str, None] = "m6h7i8j9k0l1"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def _table_exists(connection, table_name: str) -> bool:
+    """Inspector-based table check, dialect agnostic.
+
+    Trigger: SQLite creates search_index as an FTS5 virtual table at runtime
+    via SearchRepository.init_search_index, not through Alembic, so fresh
+    installs hit this migration before the table exists.
+    Why: a blind DELETE against a missing table fails the whole upgrade.
+    Outcome: callers skip the sweep when the table isn't present yet — the
+    runtime-created table on a fresh DB has no orphans to clean.
+    """
+    return table_name in inspect(connection).get_table_names()
+
+
+def upgrade() -> None:
+    """Purge orphaned search rows left over from prior project deletions.
+
+    Trigger: project deletion on SQLite never removed the derived FTS rows,
+    because the FTS5 virtual table can't carry a foreign key. The leak shows
+    up in two shapes:
+    1. project_id no longer exists in `project` (deleted project, id never
+       reused).
+    2. project_id still exists but `entity_id` no longer exists in `entity`
+       — auto-increment handed the id to a brand-new project and the FTS
+       rows from the deleted predecessor masquerade as the new tenant's data.
+    Why: search_index.project_id is the only scope predicate the search
+    repository applies, so leftover rows surface under the wrong project on
+    every search.
+    Outcome: a one-time sweep deletes both shapes, from the FTS index and
+    from search_vector_chunks. Postgres already cascaded on FK delete, so
+    these statements are no-ops there.
+    """
+    connection = op.get_bind()
+
+    if _table_exists(connection, "search_index"):
+        op.execute(
+            """
+            DELETE FROM search_index
+            WHERE project_id NOT IN (SELECT id FROM project)
+            """
+        )
+        op.execute(
+            """
+            DELETE FROM search_index
+            WHERE entity_id IS NOT NULL
+              AND entity_id NOT IN (SELECT id FROM entity)
+            """
+        )
+
+    if _table_exists(connection, "search_vector_chunks"):
+        op.execute(
+            """
+            DELETE FROM search_vector_chunks
+            WHERE project_id NOT IN (SELECT id FROM project)
+            """
+        )
+        op.execute(
+            """
+            DELETE FROM search_vector_chunks
+            WHERE entity_id NOT IN (SELECT id FROM entity)
+            """
+        )
+
+
+def downgrade() -> None:
+    """No-op: orphan rows cannot be reconstructed."""
+    pass

src/basic_memory/config.py

@@ -21,7 +21,7 @@ from basic_memory.utils import setup_logging, generate_permalink
 
 DATABASE_NAME = "memory.db"
 APP_DATABASE_NAME = "memory.db"  # Using the same name but in the app directory
-DATA_DIR_NAME = ".basic-memory"
+DATA_DIR_NAME = "basic-memory"
 CONFIG_FILE_NAME = "config.json"
 WATCH_STATUS_JSON = "watch-status.json"
 CONFIG_DIR_MODE = 0o700
@@ -69,15 +69,18 @@ def resolve_data_dir() -> Path:
 
     Single source of truth for the per-user state directory. Honors
     ``BASIC_MEMORY_CONFIG_DIR`` so each process/worktree can isolate config
-    and database state; otherwise falls back to ``<user home>/.basic-memory``.
+    and database state; otherwise falls back to ``<user home>/.basic-memory``,
+    and then to ``XDG_CONFIG_HOME``.
 
     Cross-platform: ``Path.home()`` reads ``$HOME`` on POSIX and
     ``%USERPROFILE%`` on Windows, so there's no need to check ``$HOME``
     explicitly here.
     """
-    if config_dir := os.getenv("BASIC_MEMORY_CONFIG_DIR"):
-        return Path(config_dir)
-    return Path.home() / DATA_DIR_NAME
+    if basic_memory_dir := os.getenv("BASIC_MEMORY_CONFIG_DIR"):
+        return Path(basic_memory_dir)
+    if xdg_config := os.getenv("XDG_CONFIG_HOME"):
+        return Path(xdg_config) / DATA_DIR_NAME
+    return Path.home() / ("." + DATA_DIR_NAME)
 
 
 def default_fastembed_cache_dir() -> str:

src/basic_memory/mcp/project_context.py

@@ -142,6 +142,15 @@ async def _set_cached_active_project(
         await context.set_state("default_project_name", active_project.name)
 
 
+async def _clear_cached_active_project(context: Optional[Context]) -> None:
+    """Clear cached project metadata that may no longer match the active route."""
+    if not context:
+        return
+
+    await context.set_state("active_project", None)
+    await context.set_state("default_project_name", None)
+
+
 async def _get_cached_active_workspace(context: Optional[Context]) -> Optional[WorkspaceInfo]:
     """Return the cached active workspace from context when available."""
     if not context:
@@ -167,8 +176,7 @@ async def _set_cached_active_workspace(
         # Why: project names are only unique inside one workspace, so a cached
         #   ProjectItem from the previous tenant can point at the wrong project
         # Outcome: force the next validation call to resolve within the new tenant
-        await context.set_state("active_project", None)
-        await context.set_state("default_project_name", None)
+        await _clear_cached_active_project(context)
 
     await context.set_state("active_workspace", active_workspace.model_dump())
 
@@ -525,6 +533,28 @@ def _cloud_workspace_discovery_available(config: BasicMemoryConfig) -> bool:
     )
 
 
+def _workspace_identifier_discovery_available(
+    identifier: str,
+    config: BasicMemoryConfig,
+) -> bool:
+    """Return True when an identifier is allowed to consult workspace discovery."""
+    if _cloud_workspace_discovery_available(config):
+        return True
+
+    from basic_memory.mcp.async_client import (
+        _explicit_routing,
+        _force_local_mode,
+    )
+
+    if _explicit_routing() and _force_local_mode():
+        return False
+
+    return (
+        has_cloud_credentials(config)
+        and _split_workspace_identifier_segments(identifier) is not None
+    )
+
+
 async def resolve_workspace_qualified_memory_url(
     identifier: str,
     context: Optional[Context] = None,
@@ -1326,11 +1356,23 @@ async def detect_project_from_identifier_prefix(
         # Outcome: keep unqualified search/title input on the active/default project route.
         return None
 
-    if _cloud_workspace_discovery_available(config):
-        workspace_resolution = await resolve_workspace_qualified_identifier(
-            identifier,
-            context=context,
+    if _workspace_identifier_discovery_available(identifier, config):
+        workspace_discovery_fallback_errors = (
+            "not found",
+            "no accessible workspaces",
+            "unable to discover",
         )
+        try:
+            workspace_resolution = await resolve_workspace_qualified_identifier(
+                identifier,
+                context=context,
+            )
+        except ValueError as exc:
+            message = str(exc).lower()
+            if any(error in message for error in workspace_discovery_fallback_errors):
+                return None
+            raise
+
         if workspace_resolution is not None:
             return workspace_resolution.project_identifier
 
@@ -1345,7 +1387,7 @@ async def detect_project_from_identifier_prefix(
             )
         except ValueError as exc:
             message = str(exc).lower()
-            if "not found" in message or "no accessible workspaces" in message:
+            if any(error in message for error in workspace_discovery_fallback_errors):
                 return None
             raise
 
@@ -1462,9 +1504,54 @@ async def get_project_client(
     if project_id and not cloud_available:
         project_mode = ProjectMode.LOCAL
 
+    # Trigger: project_id is a local external_id in a mixed local+cloud setup.
+    # Why: UUIDs are not local config keys, so get_project_mode() treats them as
+    #   cloud projects. A local-first probe avoids making local UUIDs depend on
+    #   healthy cloud workspace discovery.
+    # Outcome: resolve the effective UUID against local ASGI first; if it is not
+    #   local, preserve the existing cloud workspace lookup path.
+    if (
+        project_id
+        and config.projects
+        and not factory_mode
+        and not explicit_cloud_routing
+        and project_mode == ProjectMode.CLOUD
+    ):
+        try:
+            canonical_project_id = str(UUID(resolved_project))
+        except ValueError:
+            pass
+        else:
+            with logfire.span(
+                "routing.local_project_id_probe",
+                project_id=canonical_project_id,
+            ):
+                async with get_client() as client:
+                    try:
+                        active_project = await get_active_project(
+                            client,
+                            canonical_project_id,
+                            context,
+                        )
+                    except ToolError as exc:
+                        if "not found" not in str(exc).lower():
+                            raise
+                    else:
+                        route_mode = "local_asgi"
+                        await _clear_cached_active_workspace_for_local_route(context)
+                        with logfire.span(
+                            "routing.client_session",
+                            project_name=active_project.name,
+                            route_mode=route_mode,
+                        ):
+                            logger.debug("Using local ASGI routing for project_id")
+                            yield client, active_project
+                        return
+
     if factory_mode or project_mode == ProjectMode.CLOUD or explicit_cloud_routing:
         route_mode = "factory" if factory_mode else "cloud_proxy"
         active_ws: WorkspaceInfo | None = None
+        resolved_entry: WorkspaceProjectEntry | None = None
         workspace_id: str
         project_for_api = _unqualified_project_identifier(resolved_project)
 
@@ -1487,6 +1574,13 @@ async def get_project_client(
 
         if active_ws is not None:
             await _set_cached_active_workspace(context, active_ws)
+        if resolved_entry is not None:
+            cached_project = await _get_cached_active_project(context)
+            if (
+                cached_project is not None
+                and cached_project.external_id != resolved_entry.project.external_id
+            ):
+                await _clear_cached_active_project(context)
         with logfire.span(
             "routing.client_session",
             project_name=project_for_api,

src/basic_memory/mcp/tools/edit_note.py

@@ -9,7 +9,7 @@ from pydantic import AliasChoices, Field
 
 from basic_memory.config import ConfigManager
 from basic_memory.mcp.project_context import (
-    _cloud_workspace_discovery_available,
+    _workspace_identifier_discovery_available,
     detect_project_from_memory_url_prefix,
     get_project_client,
     add_project_metadata,
@@ -368,8 +368,9 @@ async def edit_note(
                 config,
                 context=context,
             )
-        elif _cloud_workspace_discovery_available(
-            config
+        elif _workspace_identifier_discovery_available(
+            identifier,
+            config,
         ) and is_workspace_qualified_plain_identifier(identifier):
             detected = await detect_project_from_workspace_identifier_prefix(
                 identifier,

src/basic_memory/mcp/tools/search.py

@@ -10,8 +10,13 @@ from loguru import logger
 from fastmcp import Context
 from pydantic import AliasChoices, BeforeValidator, Field
 
-from basic_memory.config import ConfigManager
+from basic_memory.config import ConfigManager, has_cloud_credentials
 from basic_memory.utils import build_canonical_permalink, coerce_dict, coerce_list
+from basic_memory.mcp.async_client import (
+    _explicit_routing,
+    _force_local_mode,
+    is_factory_mode,
+)
 from basic_memory.mcp.container import get_container
 from basic_memory.mcp.project_context import (
     detect_project_from_identifier_prefix,
@@ -479,12 +484,30 @@ async def _search_all_projects(
     total = 0
     any_project_has_more = False
 
+    # Trigger: caller asked for an account-wide search.
+    # Why: project_id (external UUID) routes through the cloud v2 API path,
+    #      which 401s on local installs because there's no JWT to present.
+    #      Project names route through the local-ASGI path and work for both
+    #      backends — cloud disambiguates names via the workspace/project
+    #      qualified_name already baked into project_ref["project"].
+    # Outcome: forward project_id only when the same signals get_project_client
+    #          uses to pick a cloud route are present. Mirrors the cloud_available
+    #          composite in project_context.get_project_client (single source of
+    #          truth for "can we route to cloud?").
+    config = ConfigManager().config
+    use_cloud_routing = (
+        is_factory_mode()
+        or (_explicit_routing() and not _force_local_mode())
+        or has_cloud_credentials(config)
+    )
+
     for project_ref in project_refs:
+        recursive_project_id = project_ref["project_id"] if use_cloud_routing else None
         try:
             results = await search_notes(
                 query=query,
                 project=project_ref["project"],
-                project_id=project_ref["project_id"],
+                project_id=recursive_project_id,
                 page=1,
                 page_size=per_project_page_size,
                 search_type=search_type,

src/basic_memory/repository/project_repository.py

@@ -4,7 +4,9 @@ from pathlib import Path
 from typing import Optional, Sequence, Union
 
 
-from sqlalchemy import text
+from loguru import logger
+from sqlalchemy import inspect as sa_inspect, select, text
+from sqlalchemy.exc import NoResultFound, OperationalError
 from sqlalchemy.ext.asyncio import AsyncSession, async_sessionmaker
 
 from basic_memory import db
@@ -12,6 +14,65 @@ from basic_memory.models.project import Project
 from basic_memory.repository.repository import Repository
 
 
+async def _load_sqlite_vec_on_session(session) -> bool:
+    """Ensure the sqlite-vec extension is loaded on this session's connection.
+
+    Returns True when vec0 is available after the call. Returns False when the
+    extension can't be loaded on this Python build (e.g., python.org macOS or
+    Windows interpreters without `enable_load_extension`) — every connection in
+    the pool shares the same interpreter, so a False here also means no
+    embedding row could ever have been written, and skipping the embeddings
+    purge is safe.
+
+    Mirrors SQLiteSearchRepository._ensure_sqlite_vec_loaded but as a free
+    function: we don't have a SearchRepository instance during project delete,
+    and the per-connection nature of extension loading means a pooled connection
+    routed to this session might not have vec loaded even when other
+    connections wrote embeddings.
+    """
+    try:
+        await session.execute(text("SELECT vec_version()"))
+        return True
+    except OperationalError:
+        pass
+
+    try:
+        import sqlite_vec  # type: ignore[import-not-found]
+    except ImportError:
+        logger.debug("sqlite-vec package not installed; skipping vec purge")
+        return False
+
+    async_connection = await session.connection()
+    raw_connection = await async_connection.get_raw_connection()
+    driver_connection = raw_connection.driver_connection
+
+    if not hasattr(driver_connection, "enable_load_extension"):
+        # Trigger: CPython build without sqlite extension support (#711).
+        # Why: load_extension is unavailable, so no connection in this pool
+        #      can host vec0. No embeddings exist anywhere.
+        # Outcome: skip the embeddings purge entirely.
+        logger.debug(
+            "Skipping search_vector_embeddings purge: this Python build does "
+            "not support SQLite extension loading"
+        )
+        return False
+
+    try:
+        await driver_connection.enable_load_extension(True)
+        await driver_connection.load_extension(sqlite_vec.loadable_path())
+        await driver_connection.enable_load_extension(False)
+        await session.execute(text("SELECT vec_version()"))
+    except Exception as exc:
+        logger.warning(
+            "Failed to load sqlite-vec for project delete cleanup; "
+            "skipping embeddings purge: {}",
+            exc,
+        )
+        return False
+
+    return True
+
+
 class ProjectRepository(Repository[Project]):
     """Repository for Project model.
 
@@ -121,6 +182,83 @@ class ProjectRepository(Repository[Project]):
                 return target_project
             return None  # pragma: no cover
 
+    async def delete(self, entity_id: int) -> bool:
+        """Delete a project and its derived search rows in one transaction.
+
+        The cascade picture differs by backend:
+
+        - search_index → project: Postgres has ON DELETE CASCADE FK; SQLite
+          stores search_index as an FTS5 virtual table and can't carry FKs,
+          so it needs explicit cleanup.
+        - search_vector_chunks → project: neither backend has an FK here, so
+          both need an explicit DELETE.
+        - search_vector_embeddings → search_vector_chunks: Postgres has an FK
+          (chunk_id REFERENCES … ON DELETE CASCADE); SQLite stores embeddings
+          in a vec0 virtual table keyed by rowid with no cascade. On SQLite
+          the embeddings must be purged before the chunk rows, otherwise
+          `_run_vector_query` keeps returning stale vectors that crowd out
+          live results.
+
+        Each derived table is created lazily (search_index by
+        SearchRepository.init_search_index, the vector tables once semantic
+        search initializes), so any of them may be absent on minimal test DBs.
+        Inspect the connection once and skip whichever is missing.
+        """
+        logger.debug(f"Deleting Project and search rows for project_id: {entity_id}")
+        async with db.scoped_session(self.session_maker) as session:
+            try:
+                result = await session.execute(
+                    select(self.Model).filter(self.primary_key == entity_id)
+                )
+                project = result.scalars().one()
+            except NoResultFound:
+                logger.debug(f"No Project found to delete: {entity_id}")
+                return False
+
+            dialect_name = session.bind.dialect.name if session.bind else "sqlite"
+            is_sqlite = dialect_name == "sqlite"
+
+            existing_tables = await session.run_sync(
+                lambda sync_session: set(sa_inspect(sync_session.connection()).get_table_names())
+            )
+
+            # search_index: SQLite has no FK on the FTS5 virtual table; Postgres
+            # cascades from the project FK, so the explicit DELETE is redundant.
+            if is_sqlite and "search_index" in existing_tables:
+                await session.execute(
+                    text("DELETE FROM search_index WHERE project_id = :project_id"),
+                    {"project_id": entity_id},
+                )
+
+            # search_vector_chunks: no FK to project on either backend, so both
+            # backends need this. SQLite must purge vec0 embeddings first
+            # (rowid pseudocolumn — Postgres uses chunk_id and would 500 here);
+            # Postgres' chunk_id FK CASCADE handles its embeddings cleanup when
+            # we delete the chunk rows below.
+            if "search_vector_chunks" in existing_tables:
+                if is_sqlite and "search_vector_embeddings" in existing_tables:
+                    # Extension loading is per-connection. We must load vec0 on
+                    # *this* session before the DELETE; otherwise a different
+                    # pooled connection might have written embeddings that we'd
+                    # silently leave behind.
+                    if await _load_sqlite_vec_on_session(session):
+                        await session.execute(
+                            text(
+                                "DELETE FROM search_vector_embeddings WHERE rowid IN ("
+                                "SELECT id FROM search_vector_chunks "
+                                "WHERE project_id = :project_id)"
+                            ),
+                            {"project_id": entity_id},
+                        )
+                await session.execute(
+                    text("DELETE FROM search_vector_chunks WHERE project_id = :project_id"),
+                    {"project_id": entity_id},
+                )
+
+            await session.delete(project)
+            logger.debug(f"Deleted Project and search rows for project_id: {entity_id}")
+            return True
+
     async def update_path(self, project_id: int, new_path: str) -> Optional[Project]:
         """Update project path.
 

src/basic_memory/services/link_resolver.py

@@ -40,17 +40,29 @@ async def detect_project_from_workspace_identifier_prefix(
         return None
 
     from basic_memory.mcp.project_context import (
-        _cloud_workspace_discovery_available,
+        _workspace_identifier_discovery_available,
         resolve_workspace_qualified_identifier,
     )
 
-    if not _cloud_workspace_discovery_available(config):
+    if not _workspace_identifier_discovery_available(identifier, config):
         return None
 
-    workspace_resolution = await resolve_workspace_qualified_identifier(
-        identifier,
-        context=context,
+    workspace_discovery_fallback_errors = (
+        "not found",
+        "no accessible workspaces",
+        "unable to discover",
     )
+    try:
+        workspace_resolution = await resolve_workspace_qualified_identifier(
+            identifier,
+            context=context,
+        )
+    except ValueError as exc:
+        message = str(exc).lower()
+        if any(error in message for error in workspace_discovery_fallback_errors):
+            return None
+        raise
+
     if workspace_resolution is None:
         return None
     return workspace_resolution.project_identifier

test-int/conftest.py

@@ -149,6 +149,22 @@ def clean_routing_env(monkeypatch) -> None:
     monkeypatch.delenv("BASIC_MEMORY_EXPLICIT_ROUTING", raising=False)
 
 
+@pytest.fixture(autouse=True)
+def isolate_data_dir_env(monkeypatch) -> None:
+    """Keep host data-dir env vars from leaking into integration tests.
+
+    Why: GitHub Actions Ubuntu runners set ``XDG_CONFIG_HOME=/home/runner/.config``,
+    and ``resolve_data_dir()`` honors it ahead of ``Path.home() / ".basic-memory"``.
+    Without clearing it, the MCP tool process reads config.json from the host XDG
+    path instead of the tmp dir the ``config_manager`` fixture wrote to — so
+    ``test-project`` is missing from ``config.projects``, ``get_project_mode``
+    falls through to its CLOUD default (#837), and every tool call fails with
+    "Cloud routing requested but no credentials found."
+    """
+    monkeypatch.delenv("BASIC_MEMORY_CONFIG_DIR", raising=False)
+    monkeypatch.delenv("XDG_CONFIG_HOME", raising=False)
+
+
 POSTGRES_EPHEMERAL_TABLES = [
     "search_vector_embeddings",
     "search_vector_chunks",

tests/README.md

@@ -10,6 +10,8 @@ pytest
 
 # Run tests against Postgres only (requires docker-compose)
 docker-compose -f docker-compose-postgres.yml up -d
+BASIC_MEMORY_TEST_POSTGRES=1 \
+POSTGRES_TEST_URL=postgresql+asyncpg://basic_memory_user:dev_password@localhost:5433/basic_memory \
 pytest -m postgres
 
 # Run tests against BOTH backends
@@ -54,7 +56,7 @@ database_url = None  # Uses default SQLite path
 
 # Postgres config
 database_backend = DatabaseBackend.POSTGRES
-database_url = "postgresql+asyncpg://basic_memory_user:dev_password@localhost:5433/basic_memory_test"
+database_url = "postgresql+asyncpg://basic_memory_user:dev_password@localhost:5433/basic_memory"
 ```
 
 ## Running Postgres Tests
@@ -66,18 +68,22 @@ docker-compose -f docker-compose-postgres.yml up -d
 ```
 
 This starts:
-- Postgres 17 on port **5433** (not 5432 to avoid conflicts)
-- Test database: `basic_memory_test`
+- Postgres 17 with **pgvector** (`pgvector/pgvector:pg17`) on port **5433** (not 5432 to avoid conflicts)
+- Database: `basic_memory`
 - Credentials: `basic_memory_user` / `dev_password`
 
 ### 2. Run Postgres Tests
 
 ```bash
 # Run only Postgres tests
+BASIC_MEMORY_TEST_POSTGRES=1 \
+POSTGRES_TEST_URL=postgresql+asyncpg://basic_memory_user:dev_password@localhost:5433/basic_memory \
 pytest -m postgres
 
 # Run specific test with Postgres
-pytest tests/test_entity_repository.py::test_create -m postgres
+BASIC_MEMORY_TEST_POSTGRES=1 \
+POSTGRES_TEST_URL=postgresql+asyncpg://basic_memory_user:dev_password@localhost:5433/basic_memory \
+pytest tests/repository/test_entity_repository.py::test_create -m postgres
 
 # Skip Postgres tests (default behavior)
 pytest -m "not postgres"
@@ -121,7 +127,7 @@ jobs:
     # Postgres service container
     services:
       postgres:
-        image: postgres:17
+        image: pgvector/pgvector:pg17
         env:
           POSTGRES_DB: basic_memory_test
           POSTGRES_USER: basic_memory_user
@@ -169,4 +175,4 @@ docker-compose -f docker-compose-postgres.yml exec postgres pg_isready -U basic_
 
 - [ ] Add `--run-all-backends` CLI flag to run both backends in sequence
 - [ ] Implement test fixtures for backend-specific features (e.g., Postgres full-text search vs SQLite FTS5)
-- [ ] Add performance comparison benchmarks between backends
+- [ ] Add performance comparison benchmarks between backends

tests/conftest.py

@@ -225,6 +225,19 @@ def isolate_routing_env(monkeypatch) -> None:
     monkeypatch.delenv("BASIC_MEMORY_EXPLICIT_ROUTING", raising=False)
 
 
+@pytest.fixture(autouse=True)
+def isolate_data_dir_env(monkeypatch) -> None:
+    """Keep host data-dir env vars from leaking into tests.
+
+    Why: GitHub Actions Ubuntu runners set ``XDG_CONFIG_HOME=/home/runner/.config``,
+    and ``resolve_data_dir()`` honors it ahead of ``Path.home() / ".basic-memory"``.
+    Without clearing it, tests that monkeypatch HOME still see the host XDG path
+    and assertions against the tmp home directory fail.
+    """
+    monkeypatch.delenv("BASIC_MEMORY_CONFIG_DIR", raising=False)
+    monkeypatch.delenv("XDG_CONFIG_HOME", raising=False)
+
+
 @pytest_asyncio.fixture(autouse=True)
 async def cleanup_global_db_after_test() -> AsyncGenerator[None, None]:
     """Close any module-level DB engine created outside fixture ownership."""

tests/mcp/conftest.py

@@ -20,6 +20,31 @@ def mcp() -> FastMCP:
     return cast(Any, mcp_server)
 
 
+class ContextState:
+    """Minimal FastMCP context-state stub for MCP tests."""
+
+    def __init__(self):
+        self._state: dict[str, object] = {}
+
+    async def get_state(self, key: str):
+        return self._state.get(key)
+
+    async def set_state(self, key: str, value: object, **kwargs) -> None:
+        self._state[key] = value
+
+    async def info(self, message: str) -> None:
+        self._state["info_message"] = message
+
+
+def ctx(context: ContextState) -> Any:
+    return cast(Any, context)
+
+
+@pytest.fixture
+def context_state() -> ContextState:
+    return ContextState()
+
+
 @pytest.fixture(scope="function")
 def app(
     app_config, project_config, engine_factory, config_manager

tests/mcp/test_project_context.py

@@ -11,25 +11,7 @@ from typing import Any, AsyncIterator, cast
 
 import pytest
 
-
-class _ContextState:
-    """Minimal FastMCP context-state stub for unit tests."""
-
-    def __init__(self):
-        self._state: dict[str, object] = {}
-
-    async def get_state(self, key: str):
-        return self._state.get(key)
-
-    async def set_state(self, key: str, value: object, **kwargs) -> None:
-        self._state[key] = value
-
-    async def info(self, message: str) -> None:
-        self._state["info_message"] = message
-
-
-def _ctx(context: _ContextState) -> Any:
-    return cast(Any, context)
+from tests.mcp.conftest import ContextState, ctx
 
 
 def _workspace(
@@ -218,7 +200,7 @@ async def test_env_constraint_overrides_default(config_manager, config_home, mon
 async def test_workspace_auto_selects_single_and_caches(monkeypatch):
     from basic_memory.mcp.project_context import resolve_workspace_parameter
 
-    context = _ContextState()
+    context = ContextState()
     only_workspace = _workspace(
         tenant_id="11111111-1111-1111-1111-111111111111",
         workspace_type="personal",
@@ -236,7 +218,7 @@ async def test_workspace_auto_selects_single_and_caches(monkeypatch):
         fake_get_available_workspaces,
     )
 
-    resolved = await resolve_workspace_parameter(context=_ctx(context))
+    resolved = await resolve_workspace_parameter(context=ctx(context))
     assert resolved.tenant_id == only_workspace.tenant_id
     assert await context.get_state("active_workspace") == only_workspace.model_dump()
 
@@ -272,7 +254,7 @@ async def test_workspace_requires_user_choice_when_multiple(monkeypatch):
     )
 
     with pytest.raises(ValueError, match="Multiple workspaces are available"):
-        await resolve_workspace_parameter(context=_ctx(_ContextState()))
+        await resolve_workspace_parameter(context=ctx(ContextState()))
 
 
 @pytest.mark.asyncio
@@ -416,7 +398,7 @@ async def test_workspace_type_selection_ignores_cached_workspace_for_ambiguity(m
             role="owner",
         ),
     ]
-    context = _ContextState()
+    context = ContextState()
     await context.set_state("active_workspace", cached_workspace.model_dump())
     fetches = 0
 
@@ -431,7 +413,7 @@ async def test_workspace_type_selection_ignores_cached_workspace_for_ambiguity(m
     )
 
     with pytest.raises(ValueError) as exc_info:
-        await resolve_workspace_parameter(workspace="organization", context=_ctx(context))
+        await resolve_workspace_parameter(workspace="organization", context=ctx(context))
 
     message = str(exc_info.value)
     assert fetches == 1
@@ -452,7 +434,7 @@ async def test_workspace_uses_cached_workspace_without_fetch(monkeypatch):
         role="owner",
         is_default=True,
     )
-    context = _ContextState()
+    context = ContextState()
     await context.set_state("active_workspace", cached_workspace.model_dump())
 
     async def fail_if_called(context=None):  # pragma: no cover
@@ -463,7 +445,7 @@ async def test_workspace_uses_cached_workspace_without_fetch(monkeypatch):
         fail_if_called,
     )
 
-    resolved = await resolve_workspace_parameter(context=_ctx(context))
+    resolved = await resolve_workspace_parameter(context=ctx(context))
     assert resolved.tenant_id == cached_workspace.tenant_id
 
 
@@ -476,7 +458,7 @@ async def test_workspace_project_index_caches_and_invalidates(monkeypatch):
         invalidate_workspace_project_index,
     )
 
-    context = _ContextState()
+    context = ContextState()
     personal = _workspace(
         tenant_id="personal-tenant",
         workspace_type="personal",
@@ -513,8 +495,8 @@ async def test_workspace_project_index_caches_and_invalidates(monkeypatch):
         fake_fetch_workspace_project_entries,
     )
 
-    first = await _ensure_workspace_project_index(context=_ctx(context))
-    second = await _ensure_workspace_project_index(context=_ctx(context))
+    first = await _ensure_workspace_project_index(context=ctx(context))
+    second = await _ensure_workspace_project_index(context=ctx(context))
 
     assert [entry.qualified_name for entry in first.entries] == [
         "personal/personal-notes",
@@ -523,8 +505,8 @@ async def test_workspace_project_index_caches_and_invalidates(monkeypatch):
     assert second.entries == first.entries
     assert calls == ["personal", "acme"]
 
-    await invalidate_workspace_project_index(_ctx(context))
-    await _ensure_workspace_project_index(context=_ctx(context))
+    await invalidate_workspace_project_index(ctx(context))
+    await _ensure_workspace_project_index(context=ctx(context))
     assert calls == ["personal", "acme", "personal", "acme"]
 
 
@@ -539,7 +521,7 @@ async def test_workspace_project_index_keeps_successes_when_workspace_fetch_fail
         resolve_workspace_project_identifier,
     )
 
-    context = _ContextState()
+    context = ContextState()
     personal = _workspace(
         tenant_id="personal-tenant",
         workspace_type="personal",
@@ -572,21 +554,21 @@ async def test_workspace_project_index_keeps_successes_when_workspace_fetch_fail
         fake_fetch_workspace_project_entries,
     )
 
-    index = await _ensure_workspace_project_index(context=_ctx(context))
+    index = await _ensure_workspace_project_index(context=ctx(context))
 
     assert [entry.qualified_name for entry in index.entries] == ["personal/meeting-notes"]
     assert [workspace.slug for workspace in index.failed_workspaces] == ["acme"]
 
     resolved = await resolve_workspace_project_identifier(
         "personal/meeting-notes",
-        context=_ctx(context),
+        context=ctx(context),
     )
     assert resolved.project.external_id == "personal-meeting-notes"
 
     with pytest.raises(ValueError, match="Use 'personal/meeting-notes'"):
         await resolve_workspace_project_identifier(
             "meeting-notes",
-            context=_ctx(context),
+            context=ctx(context),
         )
 
 
@@ -800,8 +782,11 @@ async def test_detect_project_from_memory_url_prefix_skips_workspace_discovery_f
 
     monkeypatch.setattr(project_context, "_ensure_workspace_project_index", fail_if_called)
 
+    # Keep this at two path segments. Three-segment memory URLs are valid
+    # workspace-qualified candidates in mixed local+cloud mode, so they should
+    # attempt workspace discovery even when a local project is configured.
     resolved = await detect_project_from_memory_url_prefix(
-        "memory://notes/foo/bar",
+        "memory://notes/foo",
         BasicMemoryConfig(
             projects={"main": ProjectEntry(path="/tmp/main")},
             cloud_api_key="bmc_test123",
@@ -811,6 +796,101 @@ async def test_detect_project_from_memory_url_prefix_skips_workspace_discovery_f
     assert resolved is None
 
 
+@pytest.mark.asyncio
+@pytest.mark.parametrize(
+    ("identifier", "message"),
+    [
+        ("memory://docs/topic/note", "No accessible workspaces found for this account."),
+        (
+            "docs/topic/note",
+            "Unable to discover projects in any accessible workspace. Failed workspaces: personal",
+        ),
+    ],
+)
+async def test_detect_project_from_identifier_prefix_ignores_workspace_discovery_failures(
+    monkeypatch,
+    identifier,
+    message,
+):
+    import basic_memory.mcp.project_context as project_context
+    from basic_memory.config import BasicMemoryConfig, ProjectEntry
+    from basic_memory.mcp.project_context import detect_project_from_identifier_prefix
+
+    async def fail_workspace_index(context=None):
+        raise ValueError(message)
+
+    monkeypatch.setattr(project_context, "_ensure_workspace_project_index", fail_workspace_index)
+
+    resolved = await detect_project_from_identifier_prefix(
+        identifier,
+        BasicMemoryConfig(
+            projects={"main": ProjectEntry(path="/tmp/main")},
+            cloud_api_key="bmc_test123",
+        ),
+    )
+
+    assert resolved is None
+
+
+@pytest.mark.asyncio
+async def test_detect_project_from_identifier_prefix_resolves_workspace_with_local_config(
+    monkeypatch,
+):
+    import basic_memory.mcp.project_context as project_context
+    from basic_memory.config import BasicMemoryConfig, ProjectEntry
+    from basic_memory.mcp.project_context import (
+        WorkspaceProjectEntry,
+        _build_workspace_project_index,
+        detect_project_from_identifier_prefix,
+    )
+
+    personal = _workspace(
+        tenant_id="personal-tenant",
+        workspace_type="personal",
+        slug="personal",
+        name="Personal",
+        role="owner",
+        is_default=True,
+    )
+    index = _build_workspace_project_index(
+        (personal,),
+        (
+            WorkspaceProjectEntry(
+                workspace=personal,
+                project=_project("main", id=1, external_id="personal-main-id"),
+            ),
+        ),
+    )
+
+    async def fake_index(context=None):
+        return index
+
+    monkeypatch.setattr(project_context, "_ensure_workspace_project_index", fake_index)
+    monkeypatch.setattr("basic_memory.mcp.async_client.is_factory_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._explicit_routing", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._force_local_mode", lambda: False)
+
+    config = BasicMemoryConfig(
+        projects={"hermes-memory": ProjectEntry(path="/tmp/hermes-memory")},
+        cloud_api_key="bmc_test123",
+    )
+
+    assert (
+        await detect_project_from_identifier_prefix(
+            "memory://personal/main/main-to-do-list",
+            config,
+        )
+        == "personal/main"
+    )
+    assert (
+        await detect_project_from_identifier_prefix(
+            "personal/main/main-to-do-list",
+            config,
+        )
+        == "personal/main"
+    )
+
+
 @pytest.mark.asyncio
 async def test_resolve_workspace_qualified_memory_url_ignores_workspace_project_miss(
     monkeypatch,
@@ -1045,7 +1125,7 @@ async def test_resolve_workspace_project_identifier_uses_active_workspace_for_du
         resolve_workspace_project_identifier,
     )
 
-    context = _ContextState()
+    context = ContextState()
     personal = _workspace(
         tenant_id="personal-tenant",
         workspace_type="personal",
@@ -1081,7 +1161,7 @@ async def test_resolve_workspace_project_identifier_uses_active_workspace_for_du
 
     resolved = await resolve_workspace_project_identifier(
         "meeting-notes",
-        context=_ctx(context),
+        context=ctx(context),
     )
     assert resolved.workspace.slug == "acme"
     assert resolved.project.external_id == "acme-project-id"
@@ -1232,6 +1312,318 @@ async def test_get_project_client_with_project_id_routes_locally_without_cloud(
     assert captured["validated_project"] == canonical_uuid
 
 
+@pytest.mark.asyncio
+async def test_get_project_client_with_local_project_id_routes_locally_with_cloud_credentials(
+    config_manager, monkeypatch
+):
+    """A local-only project_id should resolve locally before cloud workspace discovery."""
+    import basic_memory.mcp.project_context as project_context
+    from basic_memory.config import ProjectEntry
+
+    config = config_manager.load_config()
+    config.projects["hermes-memory"] = ProjectEntry(
+        path=str(config_manager.config_dir.parent / "hermes-memory")
+    )
+    config.cloud_api_key = "bmc_test123"
+    config_manager.save_config(config)
+
+    async def fail_index(context=None):  # pragma: no cover
+        raise AssertionError("local project_id should not require cloud discovery")
+
+    captured: dict[str, object] = {}
+
+    @asynccontextmanager
+    async def fake_get_client(**kwargs) -> AsyncIterator[object]:
+        captured["get_client_kwargs"] = kwargs
+        yield object()
+
+    async def fake_get_active_project(client, project, context=None, headers=None):
+        captured["validated_project"] = project
+        return _project("Hermes Memory", id=99, external_id=project)
+
+    monkeypatch.setattr(project_context, "_ensure_workspace_project_index", fail_index)
+    monkeypatch.setattr(project_context, "has_cloud_credentials", lambda _config: True)
+    monkeypatch.setattr("basic_memory.mcp.async_client.get_client", fake_get_client)
+    monkeypatch.setattr("basic_memory.mcp.async_client.is_factory_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._explicit_routing", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._force_local_mode", lambda: False)
+    monkeypatch.setattr(project_context, "get_active_project", fake_get_active_project)
+
+    local_uuid = "55555555-5555-5555-5555-555555555555"
+    async with project_context.get_project_client(project_id=local_uuid) as (_, active):
+        assert active.external_id == local_uuid
+
+    assert captured["get_client_kwargs"] == {}
+    assert captured["validated_project"] == local_uuid
+
+
+@pytest.mark.asyncio
+async def test_get_project_client_with_local_project_id_clears_cached_workspace(
+    config_manager, monkeypatch
+):
+    """Local project_id routing must not inherit a previous cloud workspace context."""
+    import basic_memory.mcp.project_context as project_context
+    from basic_memory.config import ProjectEntry
+
+    config = config_manager.load_config()
+    config.projects["hermes-memory"] = ProjectEntry(
+        path=str(config_manager.config_dir.parent / "hermes-memory")
+    )
+    config.cloud_api_key = "bmc_test123"
+    config_manager.save_config(config)
+
+    personal = _workspace(
+        tenant_id="personal-tenant",
+        workspace_type="personal",
+        slug="personal",
+        name="Personal",
+        role="owner",
+        is_default=True,
+    )
+    context = ContextState()
+    await context.set_state("active_workspace", personal.model_dump())
+
+    async def fail_index(context=None):  # pragma: no cover
+        raise AssertionError("local project_id should not require cloud discovery")
+
+    captured: dict[str, object] = {}
+
+    @asynccontextmanager
+    async def fake_get_client(**kwargs) -> AsyncIterator[object]:
+        captured["get_client_kwargs"] = kwargs
+        yield object()
+
+    async def fake_get_active_project(client, project, context=None, headers=None):
+        captured["validated_project"] = project
+        return _project("Hermes Memory", id=99, external_id=project)
+
+    monkeypatch.setattr(project_context, "_ensure_workspace_project_index", fail_index)
+    monkeypatch.setattr(project_context, "has_cloud_credentials", lambda _config: True)
+    monkeypatch.setattr("basic_memory.mcp.async_client.get_client", fake_get_client)
+    monkeypatch.setattr("basic_memory.mcp.async_client.is_factory_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._explicit_routing", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._force_local_mode", lambda: False)
+    monkeypatch.setattr(project_context, "get_active_project", fake_get_active_project)
+
+    local_uuid = "55555555-5555-5555-5555-555555555555"
+    async with project_context.get_project_client(
+        project_id=local_uuid,
+        context=ctx(context),
+    ) as (_, active):
+        assert active.external_id == local_uuid
+
+    assert captured["get_client_kwargs"] == {}
+    assert captured["validated_project"] == local_uuid
+    assert await context.get_state("active_workspace") is None
+
+
+@pytest.mark.asyncio
+async def test_get_project_client_with_project_id_respects_env_constraint(
+    config_manager, monkeypatch
+):
+    """BASIC_MEMORY_MCP_PROJECT must remain authoritative when project_id is supplied."""
+    import basic_memory.mcp.project_context as project_context
+    from basic_memory.config import ProjectEntry
+
+    config = config_manager.load_config()
+    config.projects["env-project"] = ProjectEntry(
+        path=str(config_manager.config_dir.parent / "env-project")
+    )
+    config.projects["other-project"] = ProjectEntry(
+        path=str(config_manager.config_dir.parent / "other-project")
+    )
+    config.cloud_api_key = "bmc_test123"
+    config_manager.save_config(config)
+
+    monkeypatch.setenv("BASIC_MEMORY_MCP_PROJECT", "env-project")
+
+    async def fail_index(context=None):  # pragma: no cover
+        raise AssertionError("env-constrained local project should not use cloud discovery")
+
+    captured: dict[str, object] = {}
+
+    @asynccontextmanager
+    async def fake_get_client(**kwargs) -> AsyncIterator[object]:
+        captured["get_client_kwargs"] = kwargs
+        yield object()
+
+    async def fake_get_active_project(client, project, context=None, headers=None):
+        captured["validated_project"] = project
+        return _project(str(project), id=99, external_id="env-project-id")
+
+    monkeypatch.setattr(project_context, "_ensure_workspace_project_index", fail_index)
+    monkeypatch.setattr(project_context, "has_cloud_credentials", lambda _config: True)
+    monkeypatch.setattr("basic_memory.mcp.async_client.get_client", fake_get_client)
+    monkeypatch.setattr("basic_memory.mcp.async_client.is_factory_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._explicit_routing", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._force_local_mode", lambda: False)
+    monkeypatch.setattr(project_context, "get_active_project", fake_get_active_project)
+
+    requested_uuid = "55555555-5555-5555-5555-555555555555"
+    async with project_context.get_project_client(project_id=requested_uuid) as (_, active):
+        assert active.name == "env-project"
+
+    assert captured["get_client_kwargs"] == {}
+    assert captured["validated_project"] == "env-project"
+
+
+@pytest.mark.asyncio
+async def test_get_project_client_with_cloud_project_id_routes_to_workspace_with_local_config(
+    config_manager, monkeypatch
+):
+    """Cloud project_id routing falls through to the workspace index after a local miss."""
+    import basic_memory.mcp.project_context as project_context
+    from basic_memory.config import ProjectEntry
+    from basic_memory.mcp.project_context import (
+        WorkspaceProjectEntry,
+        _build_workspace_project_index,
+    )
+    from mcp.server.fastmcp.exceptions import ToolError
+
+    config = config_manager.load_config()
+    config.projects["hermes-memory"] = ProjectEntry(
+        path=str(config_manager.config_dir.parent / "hermes-memory")
+    )
+    config.cloud_api_key = "bmc_test123"
+    config_manager.save_config(config)
+
+    cloud_uuid = "22222222-2222-2222-2222-222222222222"
+    personal = _workspace(
+        tenant_id="personal-tenant",
+        workspace_type="personal",
+        slug="personal",
+        name="Personal",
+        role="owner",
+        is_default=True,
+    )
+    cloud_project = _project("main", id=2, external_id=cloud_uuid)
+    index = _build_workspace_project_index(
+        (personal,),
+        (WorkspaceProjectEntry(workspace=personal, project=cloud_project),),
+    )
+
+    async def fake_index(context=None):
+        return index
+
+    get_client_calls: list[dict[str, str | None]] = []
+    validated_projects: list[str] = []
+
+    @asynccontextmanager
+    async def fake_get_client(project_name=None, workspace=None):
+        get_client_calls.append({"project_name": project_name, "workspace": workspace})
+        yield object()
+
+    async def fake_get_active_project(client, project, context=None, headers=None):
+        validated_projects.append(project)
+        if project == cloud_uuid:
+            raise ToolError("project not found")
+        return _project(project, id=cloud_project.id, external_id=cloud_project.external_id)
+
+    monkeypatch.setattr(project_context, "_ensure_workspace_project_index", fake_index)
+    monkeypatch.setattr(project_context, "has_cloud_credentials", lambda _config: True)
+    monkeypatch.setattr("basic_memory.mcp.async_client.get_client", fake_get_client)
+    monkeypatch.setattr("basic_memory.mcp.async_client.is_factory_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._explicit_routing", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._force_local_mode", lambda: False)
+    monkeypatch.setattr(project_context, "get_active_project", fake_get_active_project)
+
+    async with project_context.get_project_client(project_id=cloud_uuid) as (_, active):
+        assert active.external_id == cloud_uuid
+
+    assert get_client_calls == [
+        {"project_name": None, "workspace": None},
+        {"project_name": "main", "workspace": "personal-tenant"},
+    ]
+    assert validated_projects == [cloud_uuid, "main"]
+
+
+@pytest.mark.asyncio
+async def test_get_project_client_clears_stale_cached_project_for_workspace_route(
+    config_manager, monkeypatch
+):
+    """Workspace routes must not reuse a same-name project with a different UUID."""
+    import basic_memory.mcp.project_context as project_context
+    from basic_memory.mcp.project_context import WorkspaceProjectEntry
+    from basic_memory.schemas.project_info import ProjectItem
+
+    config = config_manager.load_config()
+    config.cloud_api_key = "bmc_test123"
+    config_manager.save_config(config)
+
+    personal = _workspace(
+        tenant_id="personal-tenant",
+        workspace_type="personal",
+        slug="personal",
+        name="Personal",
+        role="owner",
+        is_default=True,
+    )
+    expected_uuid = "22222222-2222-2222-2222-222222222222"
+    expected_project = _project("main", id=2, external_id=expected_uuid)
+    stale_project = ProjectItem(
+        id=99,
+        external_id="33333333-3333-3333-3333-333333333333",
+        name="main",
+        path="/tmp/stale-main",
+        is_default=False,
+    )
+    context = ContextState()
+    await context.set_state("active_workspace", personal.model_dump())
+    await context.set_state("active_project", stale_project.model_dump())
+
+    async def fake_resolve_workspace_project_identifier(project_name, context=None):
+        assert project_name == "personal/main"
+        return WorkspaceProjectEntry(workspace=personal, project=expected_project)
+
+    @asynccontextmanager
+    async def fake_get_client(project_name=None, workspace=None):
+        assert project_name == "main"
+        assert workspace == "personal-tenant"
+        yield object()
+
+    class FakeResponse:
+        def json(self):
+            return {
+                "external_id": expected_uuid,
+                "project_id": expected_project.id,
+                "name": expected_project.name,
+                "permalink": expected_project.permalink,
+                "path": expected_project.path,
+                "is_active": True,
+                "is_default": False,
+                "resolution_method": "permalink",
+            }
+
+    calls = {"count": 0}
+
+    async def fake_call_post(*args, **kwargs):
+        calls["count"] += 1
+        return FakeResponse()
+
+    monkeypatch.setattr(
+        project_context,
+        "resolve_workspace_project_identifier",
+        fake_resolve_workspace_project_identifier,
+    )
+    monkeypatch.setattr(project_context, "has_cloud_credentials", lambda _config: True)
+    monkeypatch.setattr("basic_memory.mcp.async_client.get_client", fake_get_client)
+    monkeypatch.setattr("basic_memory.mcp.async_client.is_factory_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._explicit_routing", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._force_local_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.tools.utils.call_post", fake_call_post)
+
+    async with project_context.get_project_client(
+        project="personal/main",
+        context=ctx(context),
+    ) as (_, active):
+        assert active.external_id == expected_uuid
+
+    cached_project = await context.get_state("active_project")
+    assert isinstance(cached_project, dict)
+    assert cached_project["external_id"] == expected_uuid
+    assert calls["count"] == 1
+
+
 @pytest.mark.asyncio
 async def test_get_project_client_prefers_project_id_over_project_name(monkeypatch):
     """When both project and project_id are passed, the UUID takes precedence."""
@@ -1270,7 +1662,7 @@ async def test_resolve_project_parameter_uses_cached_active_project_before_api_d
     config.default_project = None
     config_manager.save_config(config)
 
-    context = _ContextState()
+    context = ContextState()
     cached_project = ProjectItem(
         id=1,
         external_id="11111111-1111-1111-1111-111111111111",
@@ -1288,7 +1680,7 @@ async def test_resolve_project_parameter_uses_cached_active_project_before_api_d
         fail_if_called,
     )
 
-    resolved = await resolve_project_parameter(project=None, context=_ctx(context))
+    resolved = await resolve_project_parameter(project=None, context=ctx(context))
     assert resolved == cached_project.name
 
 
@@ -1302,7 +1694,7 @@ async def test_resolve_project_parameter_caches_api_default_project_name(
     config.default_project = None
     config_manager.save_config(config)
 
-    context = _ContextState()
+    context = ContextState()
     api_calls = {"count": 0}
 
     async def fake_default_lookup():
@@ -1314,8 +1706,8 @@ async def test_resolve_project_parameter_caches_api_default_project_name(
         fake_default_lookup,
     )
 
-    first = await resolve_project_parameter(project=None, context=_ctx(context))
-    second = await resolve_project_parameter(project=None, context=_ctx(context))
+    first = await resolve_project_parameter(project=None, context=ctx(context))
+    second = await resolve_project_parameter(project=None, context=ctx(context))
 
     assert first == "cloud-default"
     assert second == "cloud-default"
@@ -1327,7 +1719,7 @@ async def test_get_active_project_uses_cached_project_before_resolution(monkeypa
     from basic_memory.mcp.project_context import get_active_project
     from basic_memory.schemas.project_info import ProjectItem
 
-    context = _ContextState()
+    context = ContextState()
     cached_project = ProjectItem(
         id=1,
         external_id="11111111-1111-1111-1111-111111111111",
@@ -1345,7 +1737,7 @@ async def test_get_active_project_uses_cached_project_before_resolution(monkeypa
         fail_if_called,
     )
 
-    resolved = await get_active_project(client=cast(Any, None), context=_ctx(context))
+    resolved = await get_active_project(client=cast(Any, None), context=ctx(context))
     assert resolved == cached_project
 
 
@@ -1354,7 +1746,7 @@ async def test_get_active_project_uses_cached_project_for_explicit_permalink(mon
     from basic_memory.mcp.project_context import get_active_project
     from basic_memory.schemas.project_info import ProjectItem
 
-    context = _ContextState()
+    context = ContextState()
     cached_project = ProjectItem(
         id=1,
         external_id="11111111-1111-1111-1111-111111111111",
@@ -1375,7 +1767,7 @@ async def test_get_active_project_uses_cached_project_for_explicit_permalink(mon
     )
 
     resolved = await get_active_project(
-        client=cast(Any, None), project="my-research", context=_ctx(context)
+        client=cast(Any, None), project="my-research", context=ctx(context)
     )
     assert resolved == cached_project
 
@@ -1391,7 +1783,7 @@ async def test_resolve_project_and_path_uses_cached_project_for_memory_url_prefi
     config.permalinks_include_project = False
     config_manager.save_config(config)
 
-    context = _ContextState()
+    context = ContextState()
     cached_project = ProjectItem(
         id=1,
         external_id="11111111-1111-1111-1111-111111111111",
@@ -1416,7 +1808,7 @@ async def test_resolve_project_and_path_uses_cached_project_for_memory_url_prefi
     active_project, resolved_path, is_memory_url = await resolve_project_and_path(
         client=cast(Any, None),
         identifier="memory://my-research/notes/roadmap.md",
-        context=_ctx(context),
+        context=ctx(context),
     )
 
     assert active_project == cached_project
@@ -1438,7 +1830,7 @@ async def test_resolve_project_and_path_keeps_workspace_qualified_canonical_path
     config.permalinks_include_project = True
     config_manager.save_config(config)
 
-    context = _ContextState()
+    context = ContextState()
     cached_project = ProjectItem(
         id=1,
         external_id="11111111-1111-1111-1111-111111111111",
@@ -1464,7 +1856,7 @@ async def test_resolve_project_and_path_keeps_workspace_qualified_canonical_path
     active_project, resolved_path, is_memory_url = await resolve_project_and_path(
         client=cast(Any, None),
         identifier="memory://team-paul/main/notes/foo",
-        context=_ctx(context),
+        context=ctx(context),
     )
 
     assert active_project == cached_project
@@ -1486,7 +1878,7 @@ async def test_resolve_project_and_path_preserves_personal_workspace_prefix(
     config.permalinks_include_project = True
     config_manager.save_config(config)
 
-    context = _ContextState()
+    context = ContextState()
     cached_project = ProjectItem(
         id=1,
         external_id="11111111-1111-1111-1111-111111111111",
@@ -1513,7 +1905,7 @@ async def test_resolve_project_and_path_preserves_personal_workspace_prefix(
     active_project, resolved_path, is_memory_url = await resolve_project_and_path(
         client=cast(Any, None),
         identifier="memory://personal/main/notes/foo",
-        context=_ctx(context),
+        context=ctx(context),
     )
 
     assert active_project == cached_project
@@ -1532,7 +1924,7 @@ async def test_resolve_project_and_path_preserves_existing_project_prefixed_memo
     config.permalinks_include_project = True
     config_manager.save_config(config)
 
-    context = _ContextState()
+    context = ContextState()
     cached_project = ProjectItem(
         id=1,
         external_id="11111111-1111-1111-1111-111111111111",
@@ -1545,7 +1937,7 @@ async def test_resolve_project_and_path_preserves_existing_project_prefixed_memo
     active_project, resolved_path, is_memory_url = await resolve_project_and_path(
         client=cast(Any, None),
         identifier="memory://main/notes/foo",
-        context=_ctx(context),
+        context=ctx(context),
     )
 
     assert active_project == cached_project
@@ -1568,7 +1960,7 @@ async def test_resolve_project_and_path_uses_cached_workspace_for_active_route(
     config.permalinks_include_project = True
     config_manager.save_config(config)
 
-    context = _ContextState()
+    context = ContextState()
     cached_project = ProjectItem(
         id=1,
         external_id="11111111-1111-1111-1111-111111111111",
@@ -1599,7 +1991,7 @@ async def test_resolve_project_and_path_uses_cached_workspace_for_active_route(
         client=cast(Any, None),
         identifier="memory://notes/foo",
         project="main",
-        context=_ctx(context),
+        context=ctx(context),
     )
 
     assert active_project == cached_project
@@ -1610,7 +2002,7 @@ async def test_resolve_project_and_path_uses_cached_workspace_for_active_route(
         client=cast(Any, None),
         identifier="memory://main",
         project="main",
-        context=_ctx(context),
+        context=ctx(context),
     )
 
     assert active_project == cached_project
@@ -1650,7 +2042,7 @@ async def test_resolve_project_and_path_uses_workspace_context_for_project_root(
             client=cast(Any, None),
             identifier="memory://main",
             project="main",
-            context=_ctx(_ContextState()),
+            context=ctx(ContextState()),
         )
 
     assert active_project == active
@@ -1671,7 +2063,7 @@ async def test_resolve_project_and_path_uses_cached_workspace_for_cached_project
     config.permalinks_include_project = True
     config_manager.save_config(config)
 
-    context = _ContextState()
+    context = ContextState()
     cached_project = ProjectItem(
         id=1,
         external_id="11111111-1111-1111-1111-111111111111",
@@ -1705,7 +2097,7 @@ async def test_resolve_project_and_path_uses_cached_workspace_for_cached_project
     active_project, resolved_path, is_memory_url = await resolve_project_and_path(
         client=cast(Any, None),
         identifier="memory://main/notes/foo",
-        context=_ctx(context),
+        context=ctx(context),
     )
 
     assert active_project == cached_project
@@ -1725,7 +2117,7 @@ async def test_resolve_project_and_path_uses_cached_workspace_for_resolved_proje
     config.permalinks_include_project = True
     config_manager.save_config(config)
 
-    context = _ContextState()
+    context = ContextState()
     team_workspace = _workspace(
         tenant_id="team-tenant",
         workspace_type="organization",
@@ -1764,7 +2156,7 @@ async def test_resolve_project_and_path_uses_cached_workspace_for_resolved_proje
     active_project, resolved_path, is_memory_url = await resolve_project_and_path(
         client=cast(Any, None),
         identifier="memory://research/notes/foo",
-        context=_ctx(context),
+        context=ctx(context),
     )
 
     assert active_project.name == "Research"
@@ -1895,7 +2287,7 @@ class TestGetProjectClientRoutingOrder:
         )
         config_manager.save_config(config)
 
-        context = _ContextState()
+        context = ContextState()
         stale_workspace = _workspace(
             tenant_id="team-tenant",
             workspace_type="organization",
@@ -1942,13 +2334,13 @@ class TestGetProjectClientRoutingOrder:
 
         async with get_project_client(
             project="local-proj",
-            context=_ctx(context),
+            context=ctx(context),
         ) as (client, active_project):
             _, resolved_path, is_memory_url = await resolve_project_and_path(
                 client=cast(Any, client),
                 identifier="memory://notes/foo",
                 project="local-proj",
-                context=_ctx(context),
+                context=ctx(context),
             )
 
         assert active_project == active
@@ -1990,7 +2382,7 @@ class TestGetProjectClientRoutingOrder:
             name="Team Paul",
             role="editor",
         )
-        context = _ContextState()
+        context = ContextState()
         await context.set_state("active_workspace", workspace.model_dump())
 
         async def fail_resolve_workspace_parameter(workspace=None, context=None):
@@ -2034,7 +2426,7 @@ class TestGetProjectClientRoutingOrder:
 
         async with get_project_client(
             project="cloud-proj",
-            context=_ctx(context),
+            context=ctx(context),
         ) as (_client, active_project):
             assert active_project.external_id == "cloud-project-id"
 
@@ -2154,7 +2546,7 @@ class TestGetProjectClientRoutingOrder:
             name="Team Paul",
             role="editor",
         )
-        context = _ContextState()
+        context = ContextState()
         await context.set_state("available_workspaces", ["ignored", workspace.model_dump()])
 
         async def fail_workspace_provider():  # pragma: no cover
@@ -2193,7 +2585,7 @@ class TestGetProjectClientRoutingOrder:
 
         async with get_project_client(
             project="cloud-proj",
-            context=_ctx(context),
+            context=ctx(context),
         ) as (_client, active_project):
             assert active_project.external_id == "cloud-project-id"
 
@@ -2335,7 +2727,7 @@ class TestGetProjectClientRoutingOrder:
         config.cloud_api_key = "bmc_test123"
         config_manager.save_config(config)
 
-        context = _ContextState()
+        context = ContextState()
         stale_workspace = _workspace(
             tenant_id="other-tenant-id",
             workspace_type="organization",
@@ -2376,7 +2768,7 @@ class TestGetProjectClientRoutingOrder:
 
         async with get_project_client(
             project="cloud-proj",
-            context=_ctx(context),
+            context=ctx(context),
         ) as (_client, active_project):
             assert active_project.external_id == "cloud-project-id"
 

tests/mcp/test_project_context_telemetry.py

@@ -4,30 +4,17 @@ from __future__ import annotations
 
 import importlib
 from contextlib import contextmanager
-from typing import Any, cast
 
 import logfire
 import pytest
 
 from basic_memory.config import ProjectEntry
 from basic_memory.schemas.cloud import WorkspaceInfo
+from tests.mcp.conftest import ContextState, ctx
 
 project_context = importlib.import_module("basic_memory.mcp.project_context")
 
 
-class _ContextState:
-    """Minimal FastMCP context-state stub for unit tests."""
-
-    def __init__(self) -> None:
-        self._state: dict[str, object] = {}
-
-    async def get_state(self, key: str):
-        return self._state.get(key)
-
-    async def set_state(self, key: str, value: object, **kwargs) -> None:
-        self._state[key] = value
-
-
 def _capture_spans():
     spans: list[tuple[str, dict]] = []
 
@@ -42,7 +29,7 @@ def _capture_spans():
 @pytest.mark.asyncio
 async def test_resolve_workspace_parameter_emits_routing_span(monkeypatch) -> None:
     spans, fake_span = _capture_spans()
-    context = _ContextState()
+    context = ContextState()
     workspace = WorkspaceInfo(
         tenant_id="11111111-1111-1111-1111-111111111111",
         workspace_type="personal",
@@ -58,7 +45,7 @@ async def test_resolve_workspace_parameter_emits_routing_span(monkeypatch) -> No
     monkeypatch.setattr(logfire, "span", fake_span)
     monkeypatch.setattr(project_context, "get_available_workspaces", fake_get_available_workspaces)
 
-    resolved = await project_context.resolve_workspace_parameter(context=cast(Any, context))
+    resolved = await project_context.resolve_workspace_parameter(context=ctx(context))
 
     assert resolved.tenant_id == workspace.tenant_id
     assert spans == [

tests/mcp/test_tool_edit_note.py

@@ -914,8 +914,8 @@ async def test_edit_note_workspace_qualified_plain_permalink_requires_explicit_r
 
     monkeypatch.setattr(
         edit_note_module,
-        "_cloud_workspace_discovery_available",
-        lambda config: True,
+        "_workspace_identifier_discovery_available",
+        lambda identifier, config: True,
     )
     monkeypatch.setattr(
         edit_note_module,
@@ -958,8 +958,8 @@ async def test_edit_note_workspace_qualified_plain_permalink_json_error(
 
     monkeypatch.setattr(
         edit_note_module,
-        "_cloud_workspace_discovery_available",
-        lambda config: True,
+        "_workspace_identifier_discovery_available",
+        lambda identifier, config: True,
     )
     monkeypatch.setattr(
         edit_note_module,
@@ -1001,8 +1001,8 @@ async def test_edit_note_ambiguous_namespace_identifier_returns_guidance(
 
     monkeypatch.setattr(
         edit_note_module,
-        "_cloud_workspace_discovery_available",
-        lambda config: True,
+        "_workspace_identifier_discovery_available",
+        lambda identifier, config: True,
     )
     monkeypatch.setattr(
         edit_note_module,
@@ -1054,6 +1054,83 @@ async def test_edit_note_workspace_project_args_compose_explicit_route(monkeypat
     assert captured_routes == [("docs/setup", None)]
 
 
+@pytest.mark.asyncio
+async def test_edit_note_plain_workspace_route_returns_guidance_with_local_config(
+    monkeypatch,
+    config_manager,
+    test_project,
+):
+    """Mixed local+cloud configs should still stop ambiguous plain write routes."""
+    from contextlib import asynccontextmanager
+    import importlib
+
+    import basic_memory.mcp.project_context as project_context
+    from basic_memory.config import ProjectEntry
+    from basic_memory.mcp.project_context import (
+        WorkspaceProjectEntry,
+        _build_workspace_project_index,
+    )
+    from basic_memory.schemas.cloud import WorkspaceInfo
+    from basic_memory.schemas.project_info import ProjectItem
+
+    edit_note_module = importlib.import_module("basic_memory.mcp.tools.edit_note")
+    config = config_manager.load_config()
+    config.projects["hermes-memory"] = ProjectEntry(
+        path=str(config_manager.config_dir.parent / "hermes-memory")
+    )
+    config.cloud_api_key = "bmc_test123"
+    config_manager.save_config(config)
+
+    personal = WorkspaceInfo(
+        tenant_id="personal-tenant",
+        workspace_type="personal",
+        slug="personal",
+        name="Personal",
+        role="owner",
+        is_default=True,
+    )
+    index = _build_workspace_project_index(
+        (personal,),
+        (
+            WorkspaceProjectEntry(
+                workspace=personal,
+                project=ProjectItem(
+                    id=1,
+                    external_id="11111111-1111-1111-1111-111111111111",
+                    name="main",
+                    path="/tmp/main",
+                    is_default=False,
+                ),
+            ),
+        ),
+    )
+
+    async def fake_index(context=None):
+        return index
+
+    @asynccontextmanager
+    async def fail_if_called(*args, **kwargs):
+        raise AssertionError("ambiguous plain identifiers should not select a project client")
+        yield
+
+    monkeypatch.setattr(project_context, "_ensure_workspace_project_index", fake_index)
+    monkeypatch.setattr("basic_memory.mcp.async_client.is_factory_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._explicit_routing", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._force_local_mode", lambda: False)
+    monkeypatch.setattr(edit_note_module, "get_project_client", fail_if_called)
+
+    result = await edit_note(
+        identifier="personal/main/team/plain-edit-note",
+        operation="append",
+        content="\nAppended via plain workspace-qualified permalink.",
+        project=None,
+    )
+
+    assert isinstance(result, str)
+    assert "# Edit Failed - Ambiguous Identifier" in result
+    assert 'project="personal/main"' in result
+
+
 @pytest.mark.asyncio
 async def test_edit_note_three_segment_plain_path_stays_local_without_workspace_discovery(
     monkeypatch,
@@ -1077,8 +1154,8 @@ async def test_edit_note_three_segment_plain_path_stays_local_without_workspace_
 
     monkeypatch.setattr(
         edit_note_module,
-        "_cloud_workspace_discovery_available",
-        lambda config: False,
+        "_workspace_identifier_discovery_available",
+        lambda identifier, config: False,
     )
     monkeypatch.setattr(
         edit_note_module,

tests/mcp/test_tool_read_content.py

@@ -6,6 +6,10 @@ We keep these tests focused on path boundary/security checks, and rely on
 
 from __future__ import annotations
 
+from contextlib import asynccontextmanager
+from pathlib import Path
+from types import SimpleNamespace
+
 import pytest
 from mcp.server.fastmcp.exceptions import ToolError
 
@@ -157,6 +161,109 @@ async def test_read_content_allows_safe_path_integration(client, test_project):
     assert "safe note" in result["text"]
 
 
+@pytest.mark.asyncio
+async def test_read_content_workspace_memory_url_routes_with_local_config(
+    monkeypatch,
+    config_manager,
+):
+    """Workspace-qualified memory URLs should route even when local projects exist."""
+    import importlib
+
+    import basic_memory.mcp.project_context as project_context
+    from basic_memory.config import ProjectEntry
+    from basic_memory.mcp.project_context import (
+        WorkspaceProjectEntry,
+        _build_workspace_project_index,
+    )
+    from basic_memory.schemas.cloud import WorkspaceInfo
+    from basic_memory.schemas.project_info import ProjectItem
+
+    read_content_module = importlib.import_module("basic_memory.mcp.tools.read_content")
+    config = config_manager.load_config()
+    config.projects["hermes-memory"] = ProjectEntry(
+        path=str(config_manager.config_dir.parent / "hermes-memory")
+    )
+    config.cloud_api_key = "bmc_test123"
+    config_manager.save_config(config)
+
+    personal = WorkspaceInfo(
+        tenant_id="personal-tenant",
+        workspace_type="personal",
+        slug="personal",
+        name="Personal",
+        role="owner",
+        is_default=True,
+    )
+    project = ProjectItem(
+        id=1,
+        external_id="11111111-1111-1111-1111-111111111111",
+        name="main",
+        path="/tmp/main",
+        is_default=False,
+    )
+    index = _build_workspace_project_index(
+        (personal,),
+        (WorkspaceProjectEntry(workspace=personal, project=project),),
+    )
+
+    async def fake_index(context=None):
+        return index
+
+    @asynccontextmanager
+    async def fake_get_project_client(project=None, context=None, project_id=None):
+        assert project == "personal/main"
+        assert project_id is None
+        yield (
+            object(),
+            SimpleNamespace(
+                name="main",
+                external_id="11111111-1111-1111-1111-111111111111",
+                home=Path("/tmp/main"),
+            ),
+        )
+
+    async def fake_resolve_project_and_path(client, identifier, project=None, context=None):
+        assert identifier == "memory://personal/main/docs/report"
+        assert project == "main"
+        return None, "personal/main/docs/report", True
+
+    async def fake_resolve_entity_id(client, project_id, url):
+        assert project_id == "11111111-1111-1111-1111-111111111111"
+        assert url == "personal/main/docs/report"
+        return "entity-1"
+
+    class FakeResponse:
+        headers = {"content-type": "text/markdown", "content-length": "17"}
+        text = "# Routed Content"
+        content = b"# Routed Content"
+
+    async def fake_call_get(client, path, **kwargs):
+        assert path == "/v2/projects/11111111-1111-1111-1111-111111111111/resource/entity-1"
+        return FakeResponse()
+
+    monkeypatch.setattr(project_context, "_ensure_workspace_project_index", fake_index)
+    monkeypatch.setattr("basic_memory.mcp.async_client.is_factory_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._explicit_routing", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._force_local_mode", lambda: False)
+    monkeypatch.setattr(read_content_module, "get_project_client", fake_get_project_client)
+    monkeypatch.setattr(
+        read_content_module,
+        "resolve_project_and_path",
+        fake_resolve_project_and_path,
+    )
+    monkeypatch.setattr(read_content_module, "resolve_entity_id", fake_resolve_entity_id)
+    monkeypatch.setattr(read_content_module, "call_get", fake_call_get)
+
+    result = await read_content(path="memory://personal/main/docs/report")
+
+    assert result == {
+        "type": "text",
+        "text": "# Routed Content",
+        "content_type": "text/markdown",
+        "encoding": "utf-8",
+    }
+
+
 @pytest.mark.asyncio
 async def test_read_content_empty_path_does_not_trigger_security_error(client, test_project):
     try:

tests/mcp/test_tool_read_note.py

@@ -1,5 +1,7 @@
 """Tests for note tools that exercise the full stack with SQLite."""
 
+from contextlib import asynccontextmanager
+from types import SimpleNamespace
 from textwrap import dedent
 
 import pytest
@@ -7,6 +9,7 @@ import pytest
 from basic_memory.mcp.tools import write_note, read_note
 from basic_memory.mcp.tools.read_note import _parse_opening_frontmatter
 from basic_memory.utils import normalize_newlines
+from tests.mcp.conftest import ContextState, ctx
 
 
 def test_parse_opening_frontmatter_handles_crlf():
@@ -166,6 +169,130 @@ async def test_read_note_title_fallback_requires_exact_title_match(monkeypatch,
     assert "existing note content" not in result
 
 
+@pytest.mark.asyncio
+async def test_read_note_explicit_workspace_project_ignores_stale_cached_project(
+    monkeypatch,
+    config_manager,
+):
+    """Explicit workspace routing should use the resolved cloud UUID, not stale cache."""
+    import importlib
+
+    import basic_memory.mcp.project_context as project_context
+    from basic_memory.mcp.project_context import WorkspaceProjectEntry
+    from basic_memory.schemas.project_info import ProjectItem
+
+    read_note_module = importlib.import_module("basic_memory.mcp.tools.read_note")
+    clients_mod = importlib.import_module("basic_memory.mcp.clients")
+
+    config = config_manager.load_config()
+    config.cloud_api_key = "bmc_test123"
+    config_manager.save_config(config)
+
+    personal = project_context.WorkspaceInfo(
+        tenant_id="personal-tenant",
+        workspace_type="personal",
+        slug="personal",
+        name="Personal",
+        role="owner",
+        is_default=True,
+    )
+    expected_uuid = "22222222-2222-2222-2222-222222222222"
+    expected_project = ProjectItem(
+        id=2,
+        external_id=expected_uuid,
+        name="main",
+        path="/tmp/main",
+        is_default=False,
+    )
+    stale_project = ProjectItem(
+        id=99,
+        external_id="33333333-3333-3333-3333-333333333333",
+        name="main",
+        path="/tmp/stale-main",
+        is_default=False,
+    )
+    context = ContextState()
+    await context.set_state("active_workspace", personal.model_dump())
+    await context.set_state("active_project", stale_project.model_dump())
+
+    async def fake_resolve_workspace_project_identifier(project_name, context=None):
+        assert project_name == "personal/main"
+        return WorkspaceProjectEntry(workspace=personal, project=expected_project)
+
+    @asynccontextmanager
+    async def fake_get_client(project_name=None, workspace=None):
+        assert project_name == "main"
+        assert workspace == "personal-tenant"
+        yield object()
+
+    class FakeProjectResolveResponse:
+        def json(self):
+            return {
+                "external_id": expected_uuid,
+                "project_id": expected_project.id,
+                "name": expected_project.name,
+                "permalink": expected_project.permalink,
+                "path": expected_project.path,
+                "is_active": True,
+                "is_default": False,
+                "resolution_method": "permalink",
+            }
+
+    async def fake_call_post(*args, **kwargs):
+        return FakeProjectResolveResponse()
+
+    class FakeKnowledgeClient:
+        def __init__(self, client, project_id):
+            assert project_id == expected_uuid
+
+        async def resolve_entity(self, identifier: str, *, strict: bool = False) -> str:
+            assert identifier == "personal/main/todo"
+            return "entity-1"
+
+        async def get_entity(self, entity_id: str):
+            assert entity_id == "entity-1"
+            return SimpleNamespace(
+                title="TODO",
+                permalink="personal/main/todo",
+                file_path="TODO.md",
+            )
+
+    class FakeResourceClient:
+        def __init__(self, client, project_id):
+            assert project_id == expected_uuid
+
+        async def read(self, entity_id: str):
+            assert entity_id == "entity-1"
+            return SimpleNamespace(
+                status_code=200,
+                text="---\ntitle: TODO\n---\n\n# TODO - Priorities & Tasks\n",
+            )
+
+    monkeypatch.setattr(
+        project_context,
+        "resolve_workspace_project_identifier",
+        fake_resolve_workspace_project_identifier,
+    )
+    monkeypatch.setattr("basic_memory.mcp.async_client.get_client", fake_get_client)
+    monkeypatch.setattr("basic_memory.mcp.async_client.is_factory_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._explicit_routing", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._force_local_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.tools.utils.call_post", fake_call_post)
+    monkeypatch.setattr(clients_mod, "KnowledgeClient", FakeKnowledgeClient)
+    monkeypatch.setattr(clients_mod, "ResourceClient", FakeResourceClient)
+
+    result = await read_note_module.read_note(
+        "memory://todo",
+        project="personal/main",
+        output_format="json",
+        context=ctx(context),
+    )
+
+    assert isinstance(result, dict)
+    assert result["permalink"] == "personal/main/todo"
+    assert result["content"].strip() == "# TODO - Priorities & Tasks"
+
+
 @pytest.mark.asyncio
 async def test_note_unicode_content(app, test_project):
     """Test handling of unicode content in"""

tests/mcp/test_tool_search.py

@@ -3,6 +3,8 @@
 import pytest
 from contextlib import asynccontextmanager
 from datetime import datetime, timedelta
+from types import SimpleNamespace
+from typing import cast
 
 from basic_memory.mcp.tools import write_note
 from basic_memory.mcp.tools.search import (
@@ -171,6 +173,109 @@ async def test_search_memory_url_with_project_prefix(client, test_project):
         pytest.fail(f"Search failed with error: {response}")
 
 
+@pytest.mark.asyncio
+async def test_search_workspace_memory_url_routes_with_local_config(monkeypatch, config_manager):
+    """Workspace-qualified memory URL searches should self-route in mixed mode."""
+    import importlib
+
+    import basic_memory.mcp.project_context as project_context
+    from basic_memory.config import ProjectEntry
+    from basic_memory.mcp.project_context import (
+        WorkspaceProjectEntry,
+        _build_workspace_project_index,
+    )
+    from basic_memory.schemas.cloud import WorkspaceInfo
+    from basic_memory.schemas.project_info import ProjectItem
+    from basic_memory.schemas.search import SearchItemType, SearchResult
+
+    search_mod = importlib.import_module("basic_memory.mcp.tools.search")
+    clients_mod = importlib.import_module("basic_memory.mcp.clients")
+    config = config_manager.load_config()
+    config.projects["hermes-memory"] = ProjectEntry(
+        path=str(config_manager.config_dir.parent / "hermes-memory")
+    )
+    config.cloud_api_key = "bmc_test123"
+    config_manager.save_config(config)
+
+    personal = WorkspaceInfo(
+        tenant_id="personal-tenant",
+        workspace_type="personal",
+        slug="personal",
+        name="Personal",
+        role="owner",
+        is_default=True,
+    )
+    project_item = ProjectItem(
+        id=1,
+        external_id="11111111-1111-1111-1111-111111111111",
+        name="main",
+        path="/tmp/main",
+        is_default=False,
+    )
+    index = _build_workspace_project_index(
+        (personal,),
+        (WorkspaceProjectEntry(workspace=personal, project=project_item),),
+    )
+
+    async def fake_index(context=None):
+        return index
+
+    captured: dict[str, object] = {}
+
+    @asynccontextmanager
+    async def fake_get_project_client(project=None, context=None, project_id=None):
+        captured["project"] = project
+        captured["project_id"] = project_id
+        yield object(), SimpleNamespace(name="main", external_id=project_item.external_id)
+
+    async def fake_resolve_project_and_path(client, identifier, project=None, context=None):
+        assert identifier == "memory://personal/main/tests/search-note"
+        assert project == "main"
+        return None, "personal/main/tests/search-note", True
+
+    class FakeSearchClient:
+        def __init__(self, client, project_id):
+            captured["search_project_id"] = project_id
+
+        async def search(self, payload, *, page, page_size):
+            captured["payload"] = payload
+            return SearchResponse(
+                results=[
+                    SearchResult(
+                        title="Search Note",
+                        type=SearchItemType.ENTITY,
+                        score=1.0,
+                        permalink="personal/main/tests/search-note",
+                        file_path="tests/Search Note.md",
+                    )
+                ],
+                current_page=page,
+                page_size=page_size,
+                total=1,
+            )
+
+    monkeypatch.setattr(project_context, "_ensure_workspace_project_index", fake_index)
+    monkeypatch.setattr("basic_memory.mcp.async_client.is_factory_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._explicit_routing", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._force_local_mode", lambda: False)
+    monkeypatch.setattr(search_mod, "get_project_client", fake_get_project_client)
+    monkeypatch.setattr(search_mod, "resolve_project_and_path", fake_resolve_project_and_path)
+    monkeypatch.setattr(clients_mod, "SearchClient", FakeSearchClient)
+
+    response = await search_notes(
+        query="memory://personal/main/tests/search-note",
+        output_format="json",
+    )
+
+    assert captured["project"] == "personal/main"
+    assert captured["project_id"] is None
+    assert captured["search_project_id"] == "11111111-1111-1111-1111-111111111111"
+    payload = cast(dict[str, object], captured["payload"])
+    assert payload["permalink"] == "personal/main/tests/search-note"
+    assert isinstance(response, dict)
+    assert response["results"][0]["permalink"] == "personal/main/tests/search-note"
+
+
 @pytest.mark.asyncio
 async def test_search_pagination(client, test_project):
     """Test basic search functionality."""

tests/mcp/test_tool_workspace_management.py

@@ -1,12 +1,11 @@
 """Tests for workspace MCP tools."""
 
-from typing import Any, cast
-
 import pytest
 
 from basic_memory.mcp.project_context import get_available_workspaces, set_workspace_provider
 from basic_memory.mcp.tools.workspaces import list_workspaces
 from basic_memory.schemas.cloud import WorkspaceInfo
+from tests.mcp.conftest import ContextState, ctx
 
 
 def _workspace(
@@ -28,17 +27,6 @@ def _workspace(
     )
 
 
-class _ContextState:
-    def __init__(self):
-        self._state: dict[str, object] = {}
-
-    async def get_state(self, key: str):
-        return self._state.get(key)
-
-    async def set_state(self, key: str, value: object, **kwargs) -> None:
-        self._state[key] = value
-
-
 @pytest.mark.asyncio
 async def test_list_workspaces_formats_workspace_rows(monkeypatch):
     async def fake_get_available_workspaces(context=None):
@@ -145,7 +133,7 @@ async def test_list_workspaces_oauth_error_bubbles_up(monkeypatch):
 
 @pytest.mark.asyncio
 async def test_list_workspaces_uses_context_cache_path(monkeypatch):
-    context = _ContextState()
+    context = ContextState()
     call_count = {"fetches": 0}
     workspace = _workspace(
         tenant_id="33333333-3333-3333-3333-333333333333",
@@ -169,8 +157,8 @@ async def test_list_workspaces_uses_context_cache_path(monkeypatch):
         fake_get_available_workspaces,
     )
 
-    first = await list_workspaces(context=cast(Any, context))
-    second = await list_workspaces(context=cast(Any, context))
+    first = await list_workspaces(context=ctx(context))
+    second = await list_workspaces(context=ctx(context))
 
     assert "# Available Workspaces (1)" in first
     assert "# Available Workspaces (1)" in second
@@ -254,14 +242,14 @@ async def test_get_available_workspaces_provider_caches_in_context():
         return [workspace]
 
     set_workspace_provider(counting_provider)
-    context = _ContextState()
+    context = ContextState()
 
     # First call: provider is invoked, result cached
-    first = await get_available_workspaces(context=cast(Any, context))
+    first = await get_available_workspaces(context=ctx(context))
     assert len(first) == 1
     assert call_count["provider"] == 1
 
     # Second call: served from context cache, provider not called again
-    second = await get_available_workspaces(context=cast(Any, context))
+    second = await get_available_workspaces(context=ctx(context))
     assert len(second) == 1
     assert call_count["provider"] == 1

tests/mcp/tools/test_search_notes_multi_project.py

@@ -8,8 +8,38 @@ import pytest
 from basic_memory.schemas.search import SearchItemType, SearchResponse, SearchResult
 
 
+def _stub_routing_mode(monkeypatch, *, cloud: bool) -> None:
+    """Pin the three cloud-route signals search.py reads.
+
+    `_search_all_projects` only forwards project_id (external UUID) when a
+    cloud route is available. The composite mirrors get_project_client:
+    factory mode OR explicit --cloud OR has_cloud_credentials. Tests stub
+    all three so a dev box with OAuth tokens on disk can't bleed into the
+    local-mode case.
+    """
+    search_mod = importlib.import_module("basic_memory.mcp.tools.search")
+    monkeypatch.setattr(search_mod, "is_factory_mode", lambda: False)
+    monkeypatch.setattr(search_mod, "_explicit_routing", lambda: cloud)
+    monkeypatch.setattr(search_mod, "_force_local_mode", lambda: False)
+    monkeypatch.setattr(search_mod, "has_cloud_credentials", lambda config: cloud)
+
+
+@pytest.fixture
+def cloud_routing(monkeypatch):
+    """Force the cloud-routing path for multi-project search tests."""
+    _stub_routing_mode(monkeypatch, cloud=True)
+
+
+@pytest.fixture
+def local_routing(monkeypatch):
+    """Force the local-routing path for multi-project search tests."""
+    _stub_routing_mode(monkeypatch, cloud=False)
+
+
 @pytest.mark.asyncio
-async def test_search_notes_search_all_projects_qualifies_result_permalinks(monkeypatch):
+async def test_search_notes_search_all_projects_qualifies_result_permalinks(
+    monkeypatch, cloud_routing
+):
     """Multi-project search belongs to search_notes and keeps result ids routable."""
     clients_mod = importlib.import_module("basic_memory.mcp.clients")
     search_mod = importlib.import_module("basic_memory.mcp.tools.search")
@@ -169,7 +199,7 @@ async def test_search_notes_search_all_projects_with_no_refs_returns_empty_all_p
 
 @pytest.mark.asyncio
 async def test_search_notes_search_all_projects_continues_after_project_failure(
-    monkeypatch,
+    monkeypatch, cloud_routing
 ):
     """One failing project should not discard successful all-project search results."""
     clients_mod = importlib.import_module("basic_memory.mcp.clients")
@@ -254,3 +284,85 @@ async def test_search_notes_search_all_projects_continues_after_project_failure(
     assert result["total"] == 1
     assert any("team-paul/main" in warning for warning in warnings)
     assert any("team index unavailable" in warning for warning in warnings)
+
+
+@pytest.mark.asyncio
+async def test_search_notes_search_all_projects_local_omits_project_id(
+    monkeypatch, local_routing
+):
+    """Without a cloud route, fan-out must address each project by name only.
+
+    project_id (external UUID) routes through the cloud v2 API path, which
+    returns 401 on local installs because there's no JWT to present. Local
+    fan-out has to fall back to the name-routed path so each per-project
+    search actually returns results instead of silently failing.
+    """
+    clients_mod = importlib.import_module("basic_memory.mcp.clients")
+    search_mod = importlib.import_module("basic_memory.mcp.tools.search")
+
+    project_refs = [
+        {
+            "project": "alpha",
+            "project_id": "11111111-1111-1111-1111-111111111111",
+        },
+        {
+            "project": "beta",
+            "project_id": "22222222-2222-2222-2222-222222222222",
+        },
+    ]
+    searched_projects: list[tuple[str | None, str | None]] = []
+
+    async def fake_load_search_project_refs(context=None):
+        return project_refs
+
+    class StubProject:
+        def __init__(self, name: str | None, external_id: str | None):
+            self.name = name or "main"
+            self.external_id = external_id or "local-main"
+
+    @asynccontextmanager
+    async def fake_get_project_client(project=None, context=None, project_id=None):
+        searched_projects.append((project, project_id))
+        yield object(), StubProject(project, project_id)
+
+    async def fake_resolve_project_and_path(client, identifier, project=None, context=None):
+        return StubProject(project, None), identifier, False
+
+    class MockSearchClient:
+        def __init__(self, client, project_id):
+            self.project_id = project_id
+
+        async def search(self, payload, page, page_size):
+            return SearchResponse(
+                results=[
+                    SearchResult(
+                        title=f"Note in {self.project_id or 'local'}",
+                        permalink="notes/example",
+                        content="",
+                        type=SearchItemType.ENTITY,
+                        score=0.5,
+                        file_path="/notes/example.md",
+                    )
+                ],
+                current_page=page,
+                page_size=page_size,
+                total=1,
+            )
+
+    monkeypatch.setattr(search_mod, "_load_search_project_refs", fake_load_search_project_refs)
+    monkeypatch.setattr(search_mod, "get_project_client", fake_get_project_client)
+    monkeypatch.setattr(search_mod, "resolve_project_and_path", fake_resolve_project_and_path)
+    monkeypatch.setattr(clients_mod, "SearchClient", MockSearchClient)
+
+    result = await search_mod.search_notes(
+        query="anything",
+        search_all_projects=True,
+        output_format="json",
+    )
+
+    assert isinstance(result, dict)
+    assert searched_projects == [("alpha", None), ("beta", None)], (
+        "Local fan-out must omit project_id so the recursive search_notes calls "
+        "take the name-routed path."
+    )
+    assert result["total"] == 2

tests/services/test_link_resolver.py

@@ -139,11 +139,11 @@ async def test_workspace_identifier_project_detection_requires_workspace_shape(
     """Two-segment project-relative paths should not trigger workspace discovery."""
     from basic_memory.services import link_resolver
 
-    def fail_if_called(config):
+    def fail_if_called(identifier, config):
         raise AssertionError("plain project-relative paths should skip cloud discovery")
 
     monkeypatch.setattr(
-        "basic_memory.mcp.project_context._cloud_workspace_discovery_available",
+        "basic_memory.mcp.project_context._workspace_identifier_discovery_available",
         fail_if_called,
     )
 
@@ -164,8 +164,8 @@ async def test_workspace_identifier_project_detection_skips_without_discovery(
     from basic_memory.services import link_resolver
 
     monkeypatch.setattr(
-        "basic_memory.mcp.project_context._cloud_workspace_discovery_available",
-        lambda config: False,
+        "basic_memory.mcp.project_context._workspace_identifier_discovery_available",
+        lambda identifier, config: False,
     )
 
     detected = await link_resolver.detect_project_from_workspace_identifier_prefix(
@@ -189,8 +189,8 @@ async def test_workspace_identifier_project_detection_returns_project(
         return SimpleNamespace(project_identifier="team-acme/research")
 
     monkeypatch.setattr(
-        "basic_memory.mcp.project_context._cloud_workspace_discovery_available",
-        lambda config: True,
+        "basic_memory.mcp.project_context._workspace_identifier_discovery_available",
+        lambda identifier, config: True,
     )
     monkeypatch.setattr(
         "basic_memory.mcp.project_context.resolve_workspace_qualified_identifier",
@@ -205,6 +205,104 @@ async def test_workspace_identifier_project_detection_returns_project(
     assert detected == "team-acme/research"
 
 
+@pytest.mark.asyncio
+@pytest.mark.parametrize(
+    "message",
+    [
+        "Workspace 'team-acme' was not found.",
+        "No accessible workspaces found for this account.",
+        "Unable to discover projects in any accessible workspace. Failed workspaces: team-acme",
+    ],
+)
+async def test_workspace_identifier_project_detection_ignores_discovery_failures(
+    monkeypatch,
+    config_manager,
+    message,
+):
+    """Workspace detection should fall back when cloud discovery is unavailable."""
+    from basic_memory.services import link_resolver
+
+    async def fail_workspace_identifier(identifier, context=None):
+        raise ValueError(message)
+
+    monkeypatch.setattr(
+        "basic_memory.mcp.project_context._workspace_identifier_discovery_available",
+        lambda identifier, config: True,
+    )
+    monkeypatch.setattr(
+        "basic_memory.mcp.project_context.resolve_workspace_qualified_identifier",
+        fail_workspace_identifier,
+    )
+
+    detected = await link_resolver.detect_project_from_workspace_identifier_prefix(
+        "team-acme/research/note",
+        config_manager.config,
+    )
+
+    assert detected is None
+
+
+@pytest.mark.asyncio
+async def test_workspace_identifier_project_detection_allows_mixed_local_cloud_config(
+    monkeypatch,
+    config_manager,
+):
+    """Plain workspace routes should be discoverable even with local projects configured."""
+    from basic_memory.config import ProjectEntry
+    from basic_memory.mcp.project_context import (
+        WorkspaceProjectEntry,
+        _build_workspace_project_index,
+    )
+    from basic_memory.schemas.cloud import WorkspaceInfo
+    from basic_memory.schemas.project_info import ProjectItem
+    from basic_memory.services import link_resolver
+
+    config = config_manager.load_config()
+    config.projects["hermes-memory"] = ProjectEntry(
+        path=str(config_manager.config_dir.parent / "hermes-memory")
+    )
+    config.cloud_api_key = "bmc_test123"
+    config_manager.save_config(config)
+
+    workspace = WorkspaceInfo(
+        tenant_id="personal-tenant",
+        workspace_type="personal",
+        slug="personal",
+        name="Personal",
+        role="owner",
+        is_default=True,
+    )
+    project = ProjectItem(
+        id=1,
+        external_id="11111111-1111-1111-1111-111111111111",
+        name="main",
+        path="/tmp/main",
+        is_default=False,
+    )
+    index = _build_workspace_project_index(
+        (workspace,),
+        (WorkspaceProjectEntry(workspace=workspace, project=project),),
+    )
+
+    async def fake_index(context=None):
+        return index
+
+    monkeypatch.setattr(
+        "basic_memory.mcp.project_context._ensure_workspace_project_index",
+        fake_index,
+    )
+    monkeypatch.setattr("basic_memory.mcp.async_client.is_factory_mode", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._explicit_routing", lambda: False)
+    monkeypatch.setattr("basic_memory.mcp.async_client._force_local_mode", lambda: False)
+
+    detected = await link_resolver.detect_project_from_workspace_identifier_prefix(
+        "personal/main/todo",
+        config_manager.config,
+    )
+
+    assert detected == "personal/main"
+
+
 @pytest.mark.asyncio
 async def test_exact_permalink_match(link_resolver, test_entities, project_prefix):
     """Test resolving a link that exactly matches a permalink."""

tests/services/test_project_removal_bug.py

@@ -6,7 +6,9 @@ from datetime import timezone, datetime
 from pathlib import Path
 
 import pytest
+from sqlalchemy import text
 
+from basic_memory import db
 from basic_memory.services.project_service import ProjectService
 
 
@@ -136,3 +138,215 @@ async def test_remove_project_with_related_entities(project_service: ProjectServ
                     project = await project_service.get_project(test_project_name)
                     if project:
                         await project_service.repository.delete(project.id)
+
+
+async def _table_exists(session_maker, table: str) -> bool:
+    """Return True if the named table is present on the current connection."""
+    from sqlalchemy import inspect as sa_inspect
+
+    async with db.scoped_session(session_maker) as session:
+        return await session.run_sync(
+            lambda sync_session: table in sa_inspect(sync_session.connection()).get_table_names()
+        )
+
+
+@pytest.mark.asyncio
+async def test_remove_project_purges_search_rows(project_service: ProjectService):
+    """Project deletion must sweep the derived search tables.
+
+    SQLite stores search_index as an FTS5 virtual table, which cannot carry a
+    foreign key, so without an explicit purge the FTS rows survive the project
+    and leak into the next project that reuses the same auto-increment id.
+    Postgres has the cascade FK, but we expect the same end-state on either
+    backend. This test fails on the pre-fix code: search_index still holds the
+    project's rows after remove_project completes.
+    """
+    test_project_name = f"test-search-cleanup-{os.urandom(4).hex()}"
+    with tempfile.TemporaryDirectory() as temp_dir:
+        test_project_path = str(Path(temp_dir) / "test-search-cleanup")
+        os.makedirs(test_project_path, exist_ok=True)
+
+        await project_service.add_project(test_project_name, test_project_path)
+        project = await project_service.get_project(test_project_name)
+        assert project is not None
+        project_id = project.id
+
+        # Seed both derived tables directly. The bug is in the cleanup path,
+        # not the indexer, so a synthetic row is enough to prove the sweep.
+        async with db.scoped_session(project_service.repository.session_maker) as session:
+            await session.execute(
+                text(
+                    "INSERT INTO search_index "
+                    "(id, title, content_stems, content_snippet, permalink, "
+                    " file_path, type, project_id) "
+                    "VALUES (:id, :title, :stems, :snippet, :permalink, "
+                    " :file_path, :type, :project_id)"
+                ),
+                {
+                    "id": 999_001,
+                    "title": "leak canary",
+                    "stems": "leak canary",
+                    "snippet": "leak canary",
+                    "permalink": f"leak-canary-{project_id}",
+                    "file_path": "leak-canary.md",
+                    "type": "entity",
+                    "project_id": project_id,
+                },
+            )
+            await session.execute(
+                text(
+                    "INSERT INTO search_vector_chunks "
+                    "(entity_id, project_id, chunk_key, chunk_text, source_hash, "
+                    " entity_fingerprint, embedding_model) "
+                    "VALUES (:entity_id, :project_id, :chunk_key, :chunk_text, "
+                    " :source_hash, :entity_fingerprint, :embedding_model)"
+                ),
+                {
+                    "entity_id": 999_001,
+                    "project_id": project_id,
+                    "chunk_key": "canary",
+                    "chunk_text": "leak canary",
+                    "source_hash": "abc",
+                    "entity_fingerprint": "",
+                    "embedding_model": "",
+                },
+            )
+
+        async with db.scoped_session(project_service.repository.session_maker) as session:
+            pre_index = (
+                await session.execute(
+                    text("SELECT COUNT(*) FROM search_index WHERE project_id = :pid"),
+                    {"pid": project_id},
+                )
+            ).scalar_one()
+            pre_chunks = (
+                await session.execute(
+                    text("SELECT COUNT(*) FROM search_vector_chunks WHERE project_id = :pid"),
+                    {"pid": project_id},
+                )
+            ).scalar_one()
+        assert pre_index >= 1, "seed row should exist before removal"
+        assert pre_chunks >= 1, "seed chunk should exist before removal"
+
+        await project_service.remove_project(test_project_name)
+
+        async with db.scoped_session(project_service.repository.session_maker) as session:
+            post_index = (
+                await session.execute(
+                    text("SELECT COUNT(*) FROM search_index WHERE project_id = :pid"),
+                    {"pid": project_id},
+                )
+            ).scalar_one()
+            post_chunks = (
+                await session.execute(
+                    text("SELECT COUNT(*) FROM search_vector_chunks WHERE project_id = :pid"),
+                    {"pid": project_id},
+                )
+            ).scalar_one()
+
+        assert post_index == 0, (
+            f"search_index still has {post_index} rows for deleted project_id={project_id} "
+            "— project deletion did not sweep the FTS table."
+        )
+        assert post_chunks == 0, (
+            f"search_vector_chunks still has {post_chunks} rows for deleted "
+            f"project_id={project_id}."
+        )
+
+
+@pytest.mark.asyncio
+async def test_delete_returns_false_for_missing_project_id(project_service: ProjectService):
+    """ProjectRepository.delete must return False when the project id is gone.
+
+    The override loses the base Repository.delete contract if the NoResultFound
+    branch isn't covered — a silent True would mislead callers into thinking
+    a non-existent project was removed.
+    """
+    result = await project_service.repository.delete(9_999_999)
+    assert result is False
+
+
+@pytest.mark.asyncio
+async def test_remove_project_purges_vector_embeddings(project_service: ProjectService):
+    """Project deletion must also drop sqlite-vec embeddings keyed by chunk rowid.
+
+    sqlite-vec stores vectors in a vec0 virtual table that has no cascade
+    behavior. If embeddings linger after the chunks they reference are gone,
+    `_run_vector_query` pulls them as top-k candidates and crowds out live
+    results. The test only runs when the embeddings table is present, which
+    matches the install path that exercises semantic search.
+    """
+    test_project_name = f"test-vec-cleanup-{os.urandom(4).hex()}"
+    session_maker = project_service.repository.session_maker
+
+    # The embeddings table only exists once semantic search has initialized.
+    # Skipping when it's absent keeps this test honest on minimal CI DBs.
+    if not await _table_exists(session_maker, "search_vector_embeddings"):
+        pytest.skip("search_vector_embeddings is not present on this connection")
+
+    with tempfile.TemporaryDirectory() as temp_dir:
+        test_project_path = str(Path(temp_dir) / "test-vec-cleanup")
+        os.makedirs(test_project_path, exist_ok=True)
+
+        await project_service.add_project(test_project_name, test_project_path)
+        project = await project_service.get_project(test_project_name)
+        assert project is not None
+        project_id = project.id
+
+        async with db.scoped_session(session_maker) as session:
+            await session.execute(
+                text(
+                    "INSERT INTO search_vector_chunks "
+                    "(id, entity_id, project_id, chunk_key, chunk_text, source_hash, "
+                    " entity_fingerprint, embedding_model) "
+                    "VALUES (:id, :entity_id, :project_id, :chunk_key, :chunk_text, "
+                    " :source_hash, :entity_fingerprint, :embedding_model)"
+                ),
+                {
+                    "id": 999_201,
+                    "entity_id": 999_201,
+                    "project_id": project_id,
+                    "chunk_key": "vec-canary",
+                    "chunk_text": "vec canary",
+                    "source_hash": "abc",
+                    "entity_fingerprint": "",
+                    "embedding_model": "",
+                },
+            )
+            # vec0 requires a vector matching the configured dimensions, but the
+            # delete path filters by rowid; a non-existing dimension would block
+            # this seed step. Skip the insert if the embeddings DDL hasn't run.
+            try:
+                await session.execute(
+                    text(
+                        "INSERT INTO search_vector_embeddings (rowid, embedding) "
+                        "VALUES (:rowid, :embedding)"
+                    ),
+                    {"rowid": 999_201, "embedding": "[" + ",".join(["0.0"] * 384) + "]"},
+                )
+            except Exception:
+                pytest.skip("search_vector_embeddings rejected the synthetic seed row")
+
+        async with db.scoped_session(session_maker) as session:
+            pre = (
+                await session.execute(
+                    text("SELECT COUNT(*) FROM search_vector_embeddings WHERE rowid = :rowid"),
+                    {"rowid": 999_201},
+                )
+            ).scalar_one()
+        assert pre >= 1, "seed embedding should exist before removal"
+
+        await project_service.remove_project(test_project_name)
+
+        async with db.scoped_session(session_maker) as session:
+            post = (
+                await session.execute(
+                    text("SELECT COUNT(*) FROM search_vector_embeddings WHERE rowid = :rowid"),
+                    {"rowid": 999_201},
+                )
+            ).scalar_one()
+
+        assert post == 0, (
+            f"search_vector_embeddings still has {post} rows for rowid 999_201 "
+            "— project deletion did not sweep the sqlite-vec embeddings table."
+        )

tests/test_config.py

@@ -308,8 +308,9 @@ class TestDataDirHelpers:
     """Module-level helpers that resolve the Basic Memory data directory."""
 
     def test_resolve_data_dir_defaults_to_home_dot_basic_memory(self, config_home, monkeypatch):
-        """Without BASIC_MEMORY_CONFIG_DIR, resolver returns ~/.basic-memory."""
+        """Without BASIC_MEMORY_CONFIG_DIR and XDG_CONFIG_HOME, resolver returns ~/.basic-memory."""
         monkeypatch.delenv("BASIC_MEMORY_CONFIG_DIR", raising=False)
+        monkeypatch.delenv("XDG_CONFIG_HOME", raising=False)
 
         assert resolve_data_dir() == config_home / ".basic-memory"
 
@@ -320,6 +321,23 @@ class TestDataDirHelpers:
 
         assert resolve_data_dir() == custom
 
+    def test_resolve_data_dir_honors_xdg_config_home(self, tmp_path, monkeypatch):
+        """XDG_CONFIG_HOME is honored when BASIC_MEMORY_CONFIG_DIR is not set."""
+        monkeypatch.delenv("BASIC_MEMORY_CONFIG_DIR", raising=False)
+        xdg_config = tmp_path / "xdg-config"
+        monkeypatch.setenv("XDG_CONFIG_HOME", str(xdg_config))
+
+        assert resolve_data_dir() == xdg_config / "basic-memory"
+
+    def test_basic_memory_config_dir_takes_precedence_over_xdg(self, tmp_path, monkeypatch):
+        """BASIC_MEMORY_CONFIG_DIR takes precedence over XDG_CONFIG_HOME."""
+        xdg_config = tmp_path / "xdg-config"
+        custom = tmp_path / "custom-config"
+        monkeypatch.setenv("XDG_CONFIG_HOME", str(xdg_config))
+        monkeypatch.setenv("BASIC_MEMORY_CONFIG_DIR", str(custom))
+
+        assert resolve_data_dir() == custom
+
     def test_default_fastembed_cache_dir_uses_data_dir(self, config_home, monkeypatch):
         """Default cache path is a subdir of the Basic Memory data dir."""
         monkeypatch.delenv("BASIC_MEMORY_CONFIG_DIR", raising=False)