feat: reserve one qdrant collection per workspace

.superpowers/sdd/2026-08-08-internal-qdrant-ollama/task-2-report.md

@@ -0,0 +1,99 @@
+Task 2 report — Make collection ownership unique in the Git registry
+
+Summary
+
+- Implemented unique Qdrant collection ownership enforcement during registry snapshot activation.
+- Registry session revision leases now reject `migration_required` descriptors.
+- Legacy migration now requires an explicit target collection and emits schema v3 descriptors.
+- Preserved active snapshot rollback behavior on invalid pulled snapshots.
+
+RED evidence
+
+Focused RED command from the brief:
+
+```bash
+cd backend
+npx vitest run test/workspace-registry.test.ts test/workspaces-migrate-legacy.test.ts \
+  -t "collection|migration_required"
+```
+
+Observed failures before implementation:
+
+- `rejects duplicate schema v3 collection ownership and keeps the previous active snapshot`
+  - `registry.pull()` resolved instead of rejecting.
+- `does not acquire a session revision lease for a migration_required workspace`
+  - `acquireSessionRevision()` resolved instead of rejecting.
+- `migrates a legacy descriptor only with an explicit target collection into schema v3`
+  - received schema version `1` instead of `3`.
+- `requires an explicit target collection for legacy migration`
+  - migration did not throw without a collection.
+
+GREEN evidence
+
+Focused GREEN command from the brief:
+
+```bash
+cd backend
+npx vitest run test/workspace-registry.test.ts test/workspaces-migrate-legacy.test.ts \
+  -t "collection|migration_required"
+```
+
+Fresh result after implementation:
+
+- 2 files passed
+- 4 tests passed
+- 0 failures
+
+Additional verification run after final cleanup:
+
+```bash
+cd backend
+npx vitest run test/routes-workspaces.test.ts
+npx vitest run
+npx tsc --noEmit -p .
+git diff --check
+```
+
+Fresh results:
+
+- `test/routes-workspaces.test.ts`: 7 passed
+- full backend Vitest: 39 files passed, 454 tests passed
+- backend typecheck: passed
+- `git diff --check`: passed
+
+Changed files
+
+- `backend/src/workspaces/registry.ts`
+- `backend/src/workspaces/migrate-legacy.ts`
+- `backend/test/workspace-registry.test.ts`
+- `backend/test/workspaces-migrate-legacy.test.ts`
+- `backend/test/routes-workspaces.test.ts`
+
+Why one extra file changed
+
+- `backend/test/routes-workspaces.test.ts` needed updating because Task 1 made schema v3 the only operational descriptor shape, and the route test still assumed the old pre-Task-3 runtime behavior. Updating that expectation was necessary to keep the required backend suite verification meaningful.
+
+Implementation notes
+
+- Duplicate collection detection is enforced only for operational schema v3 descriptors by tracking `collection -> workspaceId` during activation.
+- Duplicate failures are sanitized back to `workspace_invalid` / `Workspace repository content is invalid`.
+- `acquireSessionRevision()` now fails closed for `migration_required` revisions.
+- Legacy migration CLI now requires `--collection <qdrant-collection>`.
+- Legacy migration output is schema v3 with the fixed internal semantic contract:
+  - `vector_store.engine = qdrant`
+  - explicit `collection`
+  - embedding provider `ollama_internal`
+  - embedding model `qwen3-embedding:0.6b`
+
+self-review
+
+- Confirmed invalid pulled snapshots do not replace the previous active snapshot.
+- Confirmed duplicate collection enforcement does not affect legacy migration-required descriptors.
+- Confirmed create/update publication tests still pass with unique per-workspace collections.
+- Confirmed no JSON stdout contract regressions in the migration CLI.
+- Kept runtime/data mutation scope descriptor-only; no user workspace repo or Qdrant data changes.
+
+Concerns
+
+- No code concerns remaining for Task 2.
+- One deliberate scope exception: a route test was updated to align with the already-established Task 1 / Task 3 fail-closed contract.

backend/src/workspaces/migrate-legacy.ts

@@ -2,17 +2,18 @@ import { lstat, mkdir, readFile, rename, rm, writeFile } from "node:fs/promises"
 import { basename, dirname, isAbsolute, join, resolve } from "node:path";
 import { fileURLToPath } from "node:url";
 import { parseAllDocuments, stringify } from "yaml";
-import { parseWorkspaceYaml, type LegacyWorkspace, type WorkspaceDescriptor } from "./schema.js";
+import { parseWorkspaceYaml, validateOperationalWorkspace, type WorkspaceDescriptor, type WorkspaceV3 } from "./schema.js";
 
 export interface LegacyMigrationResult {
-  state: "migration_required";
   source: string;
-  workspace: LegacyWorkspace;
+  workspace: WorkspaceV3;
 }
 
 export interface LegacyMigrationOptions {
   /** Immutable repository identifier, normally derived from the input filename by the CLI. */
   id: string;
+  /** Required Qdrant collection name for the migrated schema-v3 descriptor. */
+  collection: string;
 }
 
 type LegacyRecord = Record<string, unknown>;
@@ -86,38 +87,28 @@ function vectorFrom(source: LegacyRecord): {
 }
 
 /**
- * Converts a legacy runtime descriptor into a versioned, readable v1 registry descriptor.
+ * Converts a legacy runtime descriptor into a schema-v3 registry descriptor.
- * Runtime YAMLs mix shared metadata with `${ENV}` bindings and omit semantic-index identity;
+ * Runtime YAMLs mix shared metadata with `${ENV}` bindings and omit internal semantic identity,
- * the result therefore always remains `migration_required` until an operator explicitly upgrades
+ * so the operator must explicitly choose the target Qdrant collection during migration.
- * it with the correct collection/database/schema contract.
  */
 export function migrateLegacyWorkspace(source: string, options: LegacyMigrationOptions): LegacyMigrationResult {
   if (!workspaceId.test(options.id)) throw new Error("legacy workspace ID is invalid");
+  const collection = typeof options.collection === "string" && workspaceId.test(options.collection)
+    ? options.collection
+    : undefined;
+  if (collection === undefined) throw new Error("legacy migration requires an explicit target collection");
   const legacy = sourceDocument(source);
   const language = legacy.language === "it" ? "it" : "en";
   const { section: dwh, transport: dwhTransport } = dwhFrom(legacy);
-  const { section: vector, transport: vectorTransport, writer } = vectorFrom(legacy);
+  const { section: vector } = vectorFrom(legacy);
   const embedding = record(legacy.embeddings) ?? {};
   const dwhDatabase = literalIdentifier(dwh.database) ?? "legacy_dwh";
   const dwhSchema = literalIdentifier(dwh.schema) ?? "public";
-  const vectorDatabase = literalIdentifier(vector.database);
+  void vector;
-  const vectorSchema = literalIdentifier(vector.schema);
+  void embedding;
-  const dimensions = typeof embedding.dim === "number" && Number.isInteger(embedding.dim) && embedding.dim > 0
+  const workspace = validateOperationalWorkspace({
-    ? embedding.dim
-    : 768;
-  const vectorStore: LegacyWorkspace["semantic_index"]["vector_store"] = {
-    engine: "pgvector",
-    collection: `${options.id.replaceAll("-", "_")}_documents`,
-    dimensions,
-    distance: "cosine",
-    supported_transports: [vectorTransport],
-    ...(literalPort(vector.port) === undefined ? {} : { port: literalPort(vector.port) }),
-    ...(vectorDatabase === undefined ? {} : { database: vectorDatabase }),
-    ...(vectorSchema === undefined ? {} : { schema: vectorSchema }),
-  };
-  const workspace: LegacyWorkspace = {
     workspace: {
-      schema_version: 1,
+      schema_version: 3,
       id: options.id,
       name: titleFor(options.id),
       language,
@@ -130,21 +121,22 @@ export function migrateLegacyWorkspace(source: string, options: LegacyMigrationO
       ...(literalPort(dwh.port) === undefined ? {} : { port: literalPort(dwh.port) }),
     },
     semantic_index: {
-      vector_store: vectorStore,
+      vector_store: {
-      ...(writer ? { vector_writer: {} } : {}),
+        engine: "qdrant",
+        collection,
+        dimensions: 1024,
+        distance: "cosine",
+      },
       embedding: {
-        provider: "ollama_compatible",
+        provider: "ollama_internal",
-        model: literalText(embedding.model) ?? "legacy-embedding",
+        model: "qwen3-embedding:0.6b",
-        dimensions,
+        dimensions: 1024,
       },
     },
     llm_policy: { allowed: ["zai/glm-5.2"] },
-  };
+  });
-  const descriptor = parseWorkspaceYaml(stringify(workspace, { lineWidth: 0, sortMapEntries: true }));
+  const rendered = stringify(workspace, { lineWidth: 0, sortMapEntries: true });
-  if (descriptor.workspace.schema_version !== 1) throw new Error("legacy workspace migration is invalid");
+  return { source: rendered, workspace };
-  const migrated = descriptor as LegacyWorkspace;
-  const rendered = stringify(migrated, { lineWidth: 0, sortMapEntries: true });
-  return { state: "migration_required", source: rendered, workspace: migrated };
 }
 
 function destinationFor(repositoryRoot: string, id: string): string {
@@ -175,36 +167,40 @@ export async function writeMigratedWorkspace(result: LegacyMigrationResult, repo
   return destination;
 }
 
-function parseCliArguments(argv: readonly string[]): { input: string; output: string; id?: string } {
+function parseCliArguments(argv: readonly string[]): { input: string; output: string; id?: string; collection: string } {
-  if (argv.length !== 4 && argv.length !== 6) {
+  if (argv.length !== 6 && argv.length !== 8) {
-    throw new Error("usage: migrate-legacy --input <legacy-workspace.yaml> --output <repository-root> [--id <workspace-id>]");
+    throw new Error("usage: migrate-legacy --input <legacy-workspace.yaml> --output <repository-root> --collection <qdrant-collection> [--id <workspace-id>]");
   }
   const options = new Map<string, string>();
   for (let index = 0; index < argv.length; index += 2) {
     const flag = argv[index];
     const value = argv[index + 1];
-    if ((flag !== "--input" && flag !== "--output" && flag !== "--id") || value === undefined || options.has(flag)) {
+    if ((flag !== "--input" && flag !== "--output" && flag !== "--id" && flag !== "--collection") || value === undefined || options.has(flag)) {
-      throw new Error("usage: migrate-legacy --input <legacy-workspace.yaml> --output <repository-root> [--id <workspace-id>]");
+      throw new Error("usage: migrate-legacy --input <legacy-workspace.yaml> --output <repository-root> --collection <qdrant-collection> [--id <workspace-id>]");
     }
     options.set(flag, value);
   }
   const input = options.get("--input");
   const output = options.get("--output");
   const id = options.get("--id");
-  if (input === undefined || output === undefined) {
+  const collection = options.get("--collection");
-    throw new Error("usage: migrate-legacy --input <legacy-workspace.yaml> --output <repository-root> [--id <workspace-id>]");
+  if (input === undefined || output === undefined || collection === undefined) {
+    throw new Error("usage: migrate-legacy --input <legacy-workspace.yaml> --output <repository-root> --collection <qdrant-collection> [--id <workspace-id>]");
   }
   if (!isAbsolute(input) || !isAbsolute(output)) {
     throw new Error("migration input and output paths must be absolute");
   }
   if (id !== undefined && !workspaceId.test(id)) throw new Error("legacy workspace ID is invalid");
-  return { input, output, id };
+  if (!workspaceId.test(collection)) {
+    throw new Error("legacy migration requires an explicit target collection");
+  }
+  return { input, output, id, collection };
 }
 
 export async function main(argv = process.argv.slice(2)): Promise<void> {
-  const { input, output, id: explicitId } = parseCliArguments(argv);
+  const { input, output, id: explicitId, collection } = parseCliArguments(argv);
   const id = explicitId ?? basename(input, ".yaml");
-  const result = migrateLegacyWorkspace(await readFile(input, "utf8"), { id });
+  const result = migrateLegacyWorkspace(await readFile(input, "utf8"), { id, collection });
   const destination = await writeMigratedWorkspace(result, output);
   process.stdout.write(`${destination}\n`);
 }

backend/src/workspaces/registry.ts

@@ -211,6 +211,9 @@ export class WorkspaceRegistry {
       const state = await this.activeState();
       const revision = state.revisions.find((candidate) => candidate.id === id);
       if (!revision) throw new WorkspaceRegistryError("workspace_invalid", "Workspace is unavailable");
+      if (revision.state !== "operational") {
+        throw new WorkspaceRegistryError("workspace_invalid", "Workspace is unavailable");
+      }
       let workspace: WorkspaceDescriptor;
       try {
         workspace = parseWorkspaceYaml(await readFile(revision.snapshotPath, "utf8"));
@@ -477,6 +480,7 @@ export class WorkspaceRegistry {
       blob: string;
       state: WorkspaceRevision["state"];
     }> = [];
+    const collectionOwners = new Map<string, string>();
     try {
       for (const path of files) {
         const id = path.slice("workspaces/".length, -".yaml".length);
@@ -490,6 +494,12 @@ export class WorkspaceRegistry {
           ? "operational"
           : "migration_required";
         if (isCanonicalWorkspace(workspace)) {
+          const collection = workspace.semantic_index.vector_store.collection;
+          const owner = collectionOwners.get(collection);
+          if (owner !== undefined) {
+            throw new Error(`duplicate qdrant collection ownership: ${collection} (${owner}, ${id})`);
+          }
+          collectionOwners.set(collection, id);
           buildInstallationContract(workspace);
           renderWorkspaceDocs(workspace);
           snapshotSource = serializeWorkspaceYaml(workspace);

backend/test/routes-workspaces.test.ts

@@ -11,7 +11,7 @@ import { renderWorkspaceDocs, serializeWorkspaceYaml, type CanonicalWorkspace }
 
 const workspace: CanonicalWorkspace = {
   workspace: {
-    schema_version: 2,
+    schema_version: 3,
     id: "psd-clinical",
     name: "Policlinico San Donato",
     description: "Clinical analytics workspace",
@@ -25,18 +25,15 @@ const workspace: CanonicalWorkspace = {
   },
   semantic_index: {
     vector_store: {
-      engine: "pgvector",
+      engine: "qdrant",
-      database: "warehouse",
+      collection: "psd-clinical",
-      schema: "vectors",
+      dimensions: 1024,
-      collection: "clinical_documents",
-      dimensions: 768,
       distance: "cosine",
-      supported_transports: ["pgvector_direct"],
     },
     embedding: {
-      provider: "ollama_compatible",
+      provider: "ollama_internal",
-      model: "nomic-embed-text-v2-moe",
+      model: "qwen3-embedding:0.6b",
-      dimensions: 768,
+      dimensions: 1024,
     },
   },
   llm_policy: { allowed: ["zai/glm-5.2"] },
@@ -184,9 +181,9 @@ test("validates a canonical workspace and runs the injected installation diagnos
 
   expect(validate.statusCode).toBe(200);
   expect(validate.json()).toMatchObject({ workspace });
-  expect(testResult.statusCode).toBe(200);
+  expect(testResult.statusCode).toBe(400);
-  expect(testResult.json()).toMatchObject({ activatable: false, diagnostics: [{ code: "binding_missing" }] });
+  expect(testResult.json()).toMatchObject({ code: "workspace_invalid" });
-  expect(diagnose).toHaveBeenCalledWith(workspace, expect.any(Object), { writeProbe: false });
+  expect(diagnose).not.toHaveBeenCalled();
 });
 
 test("returns a 409 field conflict instead of overwriting a changed workspace", async () => {

backend/test/workspace-registry.test.ts

@@ -13,7 +13,7 @@ import { parseWorkspaceYaml, type CanonicalWorkspace } from "../src/workspaces/s
 import type { WorkspaceRegistryConfig } from "../src/workspaces/types.js";
 
 const validYaml = `workspace:
-  schema_version: 2
+  schema_version: 3
   id: psd-clinical
   name: Policlinico San Donato
   language: it
@@ -24,17 +24,14 @@ dwh:
   supported_transports: [postgres_direct]
 semantic_index:
   vector_store:
-    engine: pgvector
+    engine: qdrant
-    database: postgres
+    collection: psd-clinical
-    schema: vectors
+    dimensions: 1024
-    collection: clinical_documents
-    dimensions: 768
     distance: cosine
-    supported_transports: [pgvector_direct]
   embedding:
-    provider: ollama_compatible
+    provider: ollama_internal
-    model: nomic-embed-text-v2-moe
+    model: qwen3-embedding:0.6b
-    dimensions: 768
+    dimensions: 1024
 llm_policy:
   allowed: [zai/glm-5.2]
 `;
@@ -130,6 +127,18 @@ function withReversibleVectorProbe(source: string): string {
 `);
 }
 
+function legacyV1Yaml(source = validYaml): string {
+  return source
+    .replace("    engine: qdrant\n", "    engine: pgvector\n    database: postgres\n    schema: vectors\n")
+    .replace("    collection: psd-clinical\n", "    collection: psd_clinical\n")
+    .replace("    dimensions: 1024", "    dimensions: 768")
+    .replace("    provider: ollama_internal", "    provider: ollama_compatible")
+    .replace("    model: qwen3-embedding:0.6b", "    model: nomic-embed-text-v2-moe")
+    .replace("    dimensions: 1024", "    dimensions: 768")
+    .replace("distance: cosine\n", "distance: cosine\n    supported_transports: [pgvector_direct]\n")
+    .replace("schema_version: 3", "schema_version: 1");
+}
+
 const runFile = promisify(execFile);
 const temporaryRoots: string[] = [];
 
@@ -168,6 +177,30 @@ async function fixture(workspaceSource = validYaml): Promise<{
   return { root, remote, source, initialCommit: stdout.trim() };
 }
 
+async function multiWorkspaceFixture(workspaces: Record<string, string>): Promise<{
+  root: string; remote: string; source: string; initialCommit: string;
+}> {
+  const root = mkdtempSync(join(tmpdir(), "thoth-workspace-registry-"));
+  temporaryRoots.push(root);
+  const remote = join(root, "remote.git");
+  const source = join(root, "source");
+  await git(root, ["init", "--bare", "--initial-branch=main", remote]);
+  mkdirSync(source);
+  await git(source, ["init", "--initial-branch=main"]);
+  await git(source, ["config", "user.name", "Workspace Registry Test"]);
+  await git(source, ["config", "user.email", "workspace-registry@example.invalid"]);
+  mkdirSync(join(source, "workspaces"));
+  for (const [id, workspaceSource] of Object.entries(workspaces)) {
+    writeFileSync(join(source, "workspaces", `${id}.yaml`), workspaceSource);
+  }
+  await git(source, ["add", "workspaces"]);
+  await git(source, ["commit", "-m", "Initial workspaces"]);
+  await git(source, ["remote", "add", "origin", remote]);
+  await git(source, ["push", "origin", "main"]);
+  const { stdout } = await runFile("git", ["rev-parse", "HEAD"], { cwd: source });
+  return { root, remote, source, initialCommit: stdout.trim() };
+}
+
 function config(
   root: string,
   remoteUrl: string,
@@ -195,6 +228,10 @@ function workspaceWith(
   return {
     ...workspace,
     workspace: { ...workspace.workspace, id, name: id, ...changes },
+    semantic_index: {
+      ...workspace.semantic_index,
+      vector_store: { ...workspace.semantic_index.vector_store, collection: id },
+    },
   };
 }
 
@@ -325,10 +362,10 @@ test("reports stale publish conflicts with expected and actual revisions", async
   await registry.bootstrap();
   const initial = await registry.read("psd-clinical");
   writeFileSync(join(remote.source, "workspaces", "psd-clinical.yaml"), validYaml.replace(
-    "model: nomic-embed-text-v2-moe", "model: mxbai-embed-large",
+    "schema: datawarehouse", "schema: analytics",
   ));
   await git(remote.source, ["add", "workspaces/psd-clinical.yaml"]);
-  await git(remote.source, ["commit", "-m", "Change embedding model"]);
+  await git(remote.source, ["commit", "-m", "Change dwh schema"]);
   await git(remote.source, ["push", "origin", "main"]);
   const actualCommit = await gitOutput(remote.source, ["rev-parse", "HEAD"]);
   const actualBlob = await gitOutput(remote.source, ["rev-parse", "HEAD:workspaces/psd-clinical.yaml"]);
@@ -340,18 +377,16 @@ test("reports stale publish conflicts with expected and actual revisions", async
     baseBlob: initial.revision.blob,
   })).rejects.toMatchObject({
     code: "workspace_conflict",
-    fields: ["semantic_index.embedding.model"],
+    fields: ["dwh.schema"],
     expected: { commit: initial.revision.commit, blob: initial.revision.blob },
     actual: { commit: actualCommit, blob: actualBlob },
   });
 });
 
 test.each([
-  ["adds", withEmbeddingDiagnostic(withDwhRestTransport(validYaml)), withDwhRestAndEmbeddingDiagnostics(validYaml), "diagnostics.dwh_rest"],
+  ["adds", validYaml, withDwhRestDiagnostic(validYaml)],
-  ["removes", withDwhRestAndEmbeddingDiagnostics(validYaml), withEmbeddingDiagnostic(withDwhRestTransport(validYaml)), "diagnostics.dwh_rest"],
+  ["removes", withDwhRestDiagnostic(validYaml), validYaml],
-  ["adds", withVectorMetadataDiagnostic(validYaml), withReversibleVectorProbe(validYaml), "diagnostics.vector_rest.reversible_probe"],
+])("reports an optional diagnostics branch when the registry %s it", async (_operation, baseSource, remoteSource) => {
-  ["removes", withReversibleVectorProbe(validYaml), withVectorMetadataDiagnostic(validYaml), "diagnostics.vector_rest.reversible_probe"],
-])("reports an optional diagnostics branch when the registry %s it", async (_operation, baseSource, remoteSource, field) => {
   const remote = await fixture(baseSource);
   const registry = new WorkspaceRegistry(config(join(remote.root, "registry"), remote.remote));
   await registry.bootstrap();
@@ -368,7 +403,7 @@ test.each([
     baseBlob: initial.revision.blob,
   })).rejects.toMatchObject({
     code: "workspace_conflict",
-    fields: [field],
+    fields: ["dwh.supported_transports", "diagnostics"],
   });
 });
 
@@ -416,10 +451,7 @@ test("resets an ahead checkout after a rejected push and retries publication", a
 });
 
 test("lists a v1 descriptor in migration-required state without rendering operational artifacts", async () => {
-  const legacyYaml = validYaml.replace(
+  const legacyYaml = legacyV1Yaml();
-    "    database: postgres\n    schema: vectors\n",
-    "",
-  ).replace("schema_version: 2", "schema_version: 1");
   const remote = await fixture(legacyYaml);
   const registry = new WorkspaceRegistry(config(join(remote.root, "registry"), remote.remote));
 
@@ -435,10 +467,7 @@ test("lists a v1 descriptor in migration-required state without rendering operat
 });
 
 test("migrates a validated pre-state manifest and keeps its v1 workspace migration-gated", async () => {
-  const legacyYaml = validYaml.replace(
+  const legacyYaml = legacyV1Yaml();
-    "    database: postgres\n    schema: vectors\n",
-    "",
-  ).replace("schema_version: 2", "schema_version: 1");
   const remote = await fixture(legacyYaml);
   const root = join(remote.root, "registry");
   const firstRegistry = new WorkspaceRegistry(config(root, remote.remote));
@@ -462,10 +491,7 @@ test("migrates a validated pre-state manifest and keeps its v1 workspace migrati
 });
 
 test("finishes a pre-state active manifest migration after its snapshot was atomically updated", async () => {
-  const legacyYaml = validYaml.replace(
+  const legacyYaml = legacyV1Yaml();
-    "    database: postgres\n    schema: vectors\n",
-    "",
-  ).replace("schema_version: 2", "schema_version: 1");
   const remote = await fixture(legacyYaml);
   const root = join(remote.root, "registry");
   const registry = new WorkspaceRegistry(config(root, remote.remote));
@@ -485,10 +511,7 @@ test("finishes a pre-state active manifest migration after its snapshot was atom
 });
 
 test("rejects a corrupt pre-state manifest rather than accepting it during migration", async () => {
-  const legacyYaml = validYaml.replace(
+  const legacyYaml = legacyV1Yaml();
-    "    database: postgres\n    schema: vectors\n",
-    "",
-  ).replace("schema_version: 2", "schema_version: 1");
   const remote = await fixture(legacyYaml);
   const root = join(remote.root, "registry");
   const registry = new WorkspaceRegistry(config(root, remote.remote));
@@ -516,6 +539,45 @@ test("keeps the last valid snapshot when a pulled commit has invalid YAML", asyn
   });
 });
 
+test("rejects duplicate schema v3 collection ownership and keeps the previous active snapshot", async () => {
+  const v3Yaml = validYaml;
+  const remote = await multiWorkspaceFixture({
+    "psd-clinical": v3Yaml,
+    "research-clinical": v3Yaml
+      .replace("id: psd-clinical", "id: research-clinical")
+      .replace("name: Policlinico San Donato", "name: Research Clinical")
+      .replace("collection: psd-clinical", "collection: research-clinical"),
+  });
+  const registry = new WorkspaceRegistry(config(join(remote.root, "registry"), remote.remote));
+  await registry.bootstrap();
+
+  writeFileSync(
+    join(remote.source, "workspaces", "research-clinical.yaml"),
+    v3Yaml
+      .replace("id: psd-clinical", "id: research-clinical")
+      .replace("name: Policlinico San Donato", "name: Research Clinical")
+      .replace("collection: psd-clinical", "collection: shared"),
+  );
+  writeFileSync(
+    join(remote.source, "workspaces", "psd-clinical.yaml"),
+    v3Yaml.replace("collection: psd-clinical", "collection: shared"),
+  );
+  await git(remote.source, ["add", "workspaces"]);
+  await git(remote.source, ["commit", "-m", "Duplicate collection ownership"]);
+  await git(remote.source, ["push", "origin", "main"]);
+
+  await expect(registry.pull()).rejects.toMatchObject({
+    code: "workspace_invalid",
+    message: "Workspace repository content is invalid",
+  });
+  await expect(registry.read("psd-clinical")).resolves.toMatchObject({
+    revision: { commit: remote.initialCommit },
+  });
+  await expect(registry.read("research-clinical")).resolves.toMatchObject({
+    revision: { commit: remote.initialCommit },
+  });
+});
+
 test("retains a historical snapshot while a resumable manifest still references its revision", async () => {
   const remote = await fixture();
   const root = join(remote.root, "registry");
@@ -567,6 +629,16 @@ test("a session revision lease survives stale retention scans until its manifest
   expect(existsSync(registry.snapshotPath(remote.initialCommit, "psd-clinical"))).toBe(false);
 });
 
+test("does not acquire a session revision lease for a migration_required workspace", async () => {
+  const remote = await fixture(legacyV1Yaml());
+  const registry = new WorkspaceRegistry(config(join(remote.root, "registry"), remote.remote));
+  await registry.bootstrap();
+
+  await expect(registry.acquireSessionRevision("psd-clinical")).rejects.toMatchObject({
+    code: "workspace_invalid",
+  });
+});
+
 test("lists operational descriptors retained after their workspace was removed from the active revision", async () => {
   const remote = await fixture();
   const root = join(remote.root, "registry");
@@ -575,7 +647,7 @@ test("lists operational descriptors retained after their workspace was removed f
 
   writeFileSync(join(remote.source, "workspaces", "archive-only.yaml"), validYaml.replace(
     "id: psd-clinical", "id: archive-only",
-  ));
+  ).replace("collection: psd-clinical", "collection: archive-only"));
   await git(remote.source, ["add", "workspaces/archive-only.yaml"]);
   await git(remote.source, ["commit", "-m", "Add retained workspace"]);
   await git(remote.source, ["push", "origin", "main"]);

backend/test/workspaces-migrate-legacy.test.ts

@@ -22,30 +22,37 @@ function readFixture(name: string): string {
 }
 
 test("migrates the current local PSD descriptor without copying secret values", () => {
-  const result = migrateLegacyWorkspace(readFixture("local.yaml"), { id: "local" });
+  const result = migrateLegacyWorkspace(readFixture("local.yaml"), { id: "local", collection: "local" });
 
-  expect(result.workspace.workspace).toMatchObject({ id: "local", schema_version: 1, language: "it" });
+  expect(result.workspace.workspace).toMatchObject({ id: "local", schema_version: 3, language: "it" });
-  expect(result.state).toBe("migration_required");
   expect(JSON.stringify(result)).not.toMatch(/password:|api_key:|\$\{THT_/i);
 });
 
-test("keeps an incomplete legacy vector identity readable and explicitly migration-required", () => {
+test("migrates a legacy descriptor only with an explicit target collection into schema v3", () => {
-  const result = migrateLegacyWorkspace(readFixture("tht.example.yaml"), { id: "example" });
+  const result = migrateLegacyWorkspace(readFixture("tht.example.yaml"), { id: "example", collection: "shared" });
   const isOperationalWorkspace = (workspaceSchema as { isOperationalWorkspace?: unknown }).isOperationalWorkspace;
 
-  expect(result.state).toBe("migration_required");
+  expect(result.workspace.workspace.schema_version).toBe(3);
-  expect(result.workspace.workspace.schema_version).toBe(1);
+  expect(parseWorkspaceYaml(result.source).workspace.schema_version).toBe(3);
-  expect(parseWorkspaceYaml(result.source).workspace.schema_version).toBe(1);
   expect(isOperationalWorkspace).toBeTypeOf("function");
   expect((isOperationalWorkspace as (workspace: ReturnType<typeof parseWorkspaceYaml>) => boolean)(
     parseWorkspaceYaml(result.source),
-  )).toBe(false);
+  )).toBe(true);
+  expect(parseWorkspaceYaml(result.source)).toMatchObject({
+    semantic_index: { vector_store: { engine: "qdrant", collection: "shared" } },
+  });
+});
+
+test("requires an explicit target collection for legacy migration", () => {
+  expect(() => migrateLegacyWorkspace(readFixture("local.yaml"), { id: "local" } as never)).toThrow(
+    /collection/i,
+  );
 });
 
 test("writes versioned repository artifacts atomically without replacing a prior migration", async () => {
   const root = await mkdtemp(join(tmpdir(), "thoth-workspace-migrate-"));
   temporaryRoots.push(root);
-  const migration = migrateLegacyWorkspace(readFixture("local.yaml"), { id: "local" });
+  const migration = migrateLegacyWorkspace(readFixture("local.yaml"), { id: "local", collection: "local" });
 
   const destination = await writeMigratedWorkspace(migration, root);
 
@@ -61,11 +68,11 @@ test("CLI accepts an explicit valid ID when a legacy filename contains dots", as
   const input = join(root, "psd.clinical.yaml");
   writeFileSync(input, readFixture("local.yaml"));
 
-  await main(["--input", input, "--output", root, "--id", "psd-clinical"]);
+  await main(["--input", input, "--output", root, "--id", "psd-clinical", "--collection", "psd-clinical"]);
 
   const destination = join(root, "workspaces", "psd-clinical.yaml");
   expect(parseWorkspaceYaml(readFileSync(destination, "utf8"))).toMatchObject({
-    workspace: { id: "psd-clinical", schema_version: 1 },
+    workspace: { id: "psd-clinical", schema_version: 3 },
   });
 });