@crvouga/mockingbird-service-llamacloud 3.2.0 → 3.3.0

package/CHANGELOG.md

@@ -1,15 +1,21 @@
 # Changelog — @crvouga/mockingbird-service-llamacloud
 
-## 3.2.0 (2026-10-09)
+## 3.3.0 (2026-10-12)
 
 ### Features
 
-- match Emulate provider coverage ([16b65ce](https://github.com/crvouga/mockingbird/commit/16b65ce791fa216f28ec1521df636f57771d9c5e))
+- unify the emulator control plane ([2068547](https://github.com/crvouga/mockingbird/commit/2068547bedde4e2e2331c1657c1a78120e84428b))
 
 ### Dependencies
 
 - `@crvouga/mockingbird-service-sqlite`
 
+## 3.2.0 (2026-10-08)
+
+### Features
+
+- match Emulate provider coverage ([16b65ce](https://github.com/crvouga/mockingbird/commit/16b65ce791fa216f28ec1521df636f57771d9c5e))
+
 ## 3.1.0 (2026-10-08)
 
 ### Features

package/DISCOVERY.md

@@ -35,6 +35,7 @@ repository; live parity is an additional oracle check, not a substitute for the
 
 - `GET /__admin/health`
 - `GET /__admin`
+- `GET /__admin/controls`
 - `GET /__admin/state`
 - `GET /__admin/requests`
 - `GET /__admin/metrics`
@@ -45,6 +46,28 @@ For HTTP services, use `x-mockingbird-namespace` (or the documented credential/p
 parallel tests do not share state. Admin state, journal, metrics and fault-preset endpoints are
 designed for assertions and diagnosis by consuming test suites.
 
+## Admin controls
+
+Every HTTP emulator takes the same controls, and each means the same thing at boot
+(`createRuntime({ adminControls })`, `serve --admin-controls`), over `/__admin`, and in a request header:
+
+| Control | Admin API | Request header | Interface |
+| --- | --- | --- | --- |
+| Isolate state by key | `GET /__admin/namespaces` | `x-mockingbird-namespace: <key>` | [Namespaces](https://mockingbird.chrisvouga.dev/docs/admin-controls/namespaces) |
+| Reset to boot | `POST /__admin/reset` | `x-mockingbird-reset: true` | [Reset](https://mockingbird.chrisvouga.dev/docs/admin-controls/reset) |
+| Snapshot and restore | `POST /__admin/snapshots`, `POST /__admin/snapshots/<id>/restore` | `x-mockingbird-snapshot: <name>`, `x-mockingbird-restore: <name>` | [Snapshot and restore](https://mockingbird.chrisvouga.dev/docs/admin-controls/snapshot-and-restore) |
+| Move the clock | `POST /__admin/clock` | `x-mockingbird-time: <instant>`, `x-mockingbird-advance: <duration>` | [Clock](https://mockingbird.chrisvouga.dev/docs/admin-controls/clock) |
+| Read or branch from old state | `GET /__admin/timeline` | `x-mockingbird-at: <checkpoint>`, `x-mockingbird-branch: <name>` | [Checkpoints and branches](https://mockingbird.chrisvouga.dev/docs/admin-controls/checkpoints-and-branches) |
+| Inject a fault | `POST /__admin/faults` | `x-mockingbird-fault: <fault>` | [Faults](https://mockingbird.chrisvouga.dev/docs/admin-controls/faults) |
+| Latency and timeouts | `PUT /__admin/latency` | `x-mockingbird-delay: <duration>`, `x-mockingbird-timeout: <duration>` | [Latency and timeouts](https://mockingbird.chrisvouga.dev/docs/admin-controls/latency-and-timeouts) |
+| Replay from a seed | `PUT /__admin/controls` | `x-mockingbird-seed: <seed>` | [Seed and determinism](https://mockingbird.chrisvouga.dev/docs/admin-controls/seed-and-determinism) |
+| See what happened | `GET /__admin/requests`, `GET /__admin/state` | | [Journal, metrics and state](https://mockingbird.chrisvouga.dev/docs/admin-controls/journal-and-state) |
+
+`GET /__admin` lists every route and header with a one-line summary, and `GET /__admin/controls`
+is the namespace's current control document. The same seed and the same requests give the same
+bytes. Setting controls as an emulator starts is [Boot](https://mockingbird.chrisvouga.dev/docs/admin-controls/boot); every route and header is in
+the [Reference](https://mockingbird.chrisvouga.dev/docs/admin-controls/reference). The full guide: <https://mockingbird.chrisvouga.dev/docs/admin-controls>.
+
 ## Report a mismatch or missing capability
 
 Follow the [agent reporting contract](https://github.com/crvouga/mockingbird/blob/main/docs/REPORTING_ISSUES.md). Include package version,

package/README.md

@@ -37,24 +37,32 @@ Point the app at it:
 | Python chat service (`llama_cloud_services`) | `LLAMA_CLOUD_BASE_URL=http://127.0.0.1:8805` (no `/api/v1`; verified below, no code change needed) |
 
 ```ts
+import LlamaCloud from "@llamaindex/llama-cloud"
 import { createRuntime } from "@crvouga/mockingbird-service-llamacloud"
 
-const llama = createRuntime({ pipelines: [{ name: "acme-member-kb-v1", projectName: "Default" }] })
-const admin = (path: string, body: unknown) =>
-  llama.fetch(
-    new Request(`http://llamacloud.test/__admin${path}`, {
-      method: "PUT",
-      headers: { "content-type": "application/json" },
-      body: JSON.stringify(body),
+// Test setup: a pipeline, and what retrieval answers for queries mentioning "apob".
+// Anything else is ranked by term overlap over the documents the app upserted.
+const emulator = createRuntime({ pipelines: [{ name: "acme-member-kb-v1", projectName: "Default" }] })
+await emulator.fetch(
+  new Request("http://llamacloud.test/__admin/retrieval", {
+    method: "PUT",
+    headers: { "content-type": "application/json" },
+    body: JSON.stringify({
+      match: { contains: "apob" },
+      nodes: [{ text: "ApoB counts atherogenic particles.", score: 0.92, metadata: { file_name: "apob.md" } }],
     }),
-  )
-
-// Script what the chat tool retrieves for queries mentioning "apob"…
-await admin("/retrieval", {
-  match: { contains: "apob" },
-  nodes: [{ text: "ApoB counts atherogenic particles.", score: 0.92, metadata: { file_name: "apob.md" } }],
+  }),
+)
+
+// The LlamaCloud SDK your application already uses. Only its fetch changes: requests reach
+// the emulator in this process instead of api.cloud.llamaindex.ai.
+const llama = new LlamaCloud({
+  apiKey: "llx-mockingbird",
+  fetch: (url: RequestInfo | URL, init?: RequestInit) => emulator.fetch(new Request(url, init)),
 })
-// …anything else is ranked by term overlap over the documents the app upserted.
+const [pipeline] = await llama.pipelines.list({ project_name: "Default", pipeline_name: "acme-member-kb-v1" })
+const { retrieval_nodes } = await llama.pipelines.retrieve(pipeline?.id ?? "", { query: "What does ApoB measure?" })
+console.log(retrieval_nodes[0]?.node.text, retrieval_nodes[0]?.score) // ApoB counts atherogenic particles. 0.92
 ```
 
 ### Routes