@vellumai/assistant 0.12.5-dev.202609260125.8a6a403 → 0.12.5-staging.1

package/ARCHITECTURE.md

@@ -45,12 +45,6 @@ Resource pressure monitoring reports sustained CPU/memory pressure on platform-h
 
 **Lifecycle:** `src/daemon/resource-pressure-guard-lifecycle.ts` starts the guard at daemon boot with the first sample deferred onto a macrotask so it never blocks startup, and stops the guard (cancelling any pending deferred sample) on shutdown. The web chat banner built on this status is documented in the repo-level [`/ARCHITECTURE.md`](../ARCHITECTURE.md) "Resource Pressure Monitoring" section.
 
-### OOM Priority
-
-When the container hits its memory limit, the kernel kills the process with the highest `oom_score` (memory share plus `oom_score_adj`). `src/util/oom-priority.ts` steers that choice so the assistant is never the victim of a runaway tool: the daemon sets itself to -700 as the first thing in `runDaemon`, before any spawn. Children inherit the parent's value at fork, so every spawn path resets its child: shell children get +1000 from the Linux prefix in `buildShellInvocation` (`packages/environments`), desktop children and ACP agents and terminals get +1000 from `withOomScoreAdj`, an `sh -c ... exec` wrapper that keeps pid and process group, and workers, the embedding worker, and Qdrant get 0 from a parent-side write right after spawn (the resource monitor gets -500 so it outlives everything but the daemon and records the kill). The monitor also sweeps the daemon's descendants every few seconds (`src/monitoring/oom-inheritance-sweep.ts`) and writes 0 over any negative value it finds, logging the comm, so a spawn path that forgets the reset degrades to a few seconds of inherited protection instead of a protected runaway. Bundled plugins (the memory plugin's rerank worker, for one) rely on the sweep rather than importing the helper, since the plugin import boundary forbids reaching outside the plugin. Raising never needs a capability; lowering needs `CAP_SYS_RESOURCE`, which the platform grants and self-hosted Docker does not, where the daemon simply stays at 0. Nothing here caps a single process or acts early: it only decides who dies once the kernel has to kill.
-
-**Verification telemetry:** the resource monitor reports kernel OOM kills in the container as a `watchdog` event with `check_name` `oom_kill` (`src/monitoring/oom-kill-reporter.ts`). It reads the kernel log through `dmesg` (the `/dev/kmsg` device is not mounted in the container; `dmesg` uses the syslog syscall, which the Kata profile permits) when the cgroup `memory.events` kill counter moves, on its first scan after start (the kill that took the daemon down restarts the monitor with it), and once a minute as a fallback. The kernel log is shared by every container on that kernel and cgroup paths read `/` from inside, so attribution is by the counter: a move of N claims the N most recent new kills; the first scan and hosts without the counter claim every new kill and say so with `attribution: "kernel_log"`. The `detail` bag lists each victim's comm, `oom_score_adj` and resident size, the cgroup memory figures at the time, and `killed_daemon`, detected by the daemon's -700 score rather than a pid the restart has since replaced; where the log is unreadable a counter move still yields a victimless event. A cursor (kernel boot id plus last log timestamp) on the monitor data dir advances only after the event is queued, so a telemetry store outage delays a report rather than losing one, and resets when the kernel reboots.
-
 ### Single-Header JWT Auth Model
 
 All HTTP API requests use a single `Authorization: Bearer <jwt>` header for authentication. The JWT carries identity, permissions, and policy versioning in a unified token.
@@ -908,23 +902,6 @@ graph LR
 
 ---
 
-## Automatic Done
-
-`src/conversations/auto-archive.ts` runs an opt-in automatic Done sweep after startup recovery, on assistant config or feature-flag invalidation, and hourly. It marks eligible inactive native chats through the existing `archived_at` state. Candidate selection and the conditional update share the same persisted eligibility predicates; runtime checks run synchronously with each conditional write. Explicit reopening records `last_reopened_at` to grant a full inactivity interval.
-
-```mermaid
-flowchart LR
-    TRIGGER[Startup recovery / config or feature-flag change / hourly timer] --> PAGE[Bounded candidate page]
-    PAGE --> GUARDIAN[Read pending guardian requests]
-    GUARDIAN --> CHECK[Recheck config and runtime activity]
-    CHECK --> CAS[Conditional SQLite Done update]
-    CAS --> SYNC[Conversation list and metadata invalidation]
-```
-
-Readiness, eligibility, restart behavior, and the gateway snapshot boundary are documented in [Automatic Done](docs/automatic-done.md).
-
----
-
 ## Plugin-Declared Schedules: Declaration → Reconciler → cron_jobs
 
 Plugins contribute recurring schedules as a surface: directories under

package/docs/desktop-browser-cli.md

@@ -53,9 +53,8 @@ Non-platform assistants keep their connected-computer behavior.
 Explicit `target: "connected-computer"` or `target_client_id` selects a connected
 computer on every client. The selected target never falls back to another
 computer when unavailable. The virtual desktop uses a fixed 1440x810 display.
-Observations return full-screen screenshots and mouse actions use screen
-points. Screenshot metadata uses the transport-optimized image dimensions and
-provides per-axis conversion from image pixels to those screen points.
+Observations return full-screen screenshots and mouse actions use the same
+screen coordinates.
 Viewer resizing scales the canvas locally without changing the desktop layout.
 
 Observe first and pass the returned `observation_id` with each native action.

package/node_modules/@vellumai/ces-client/node_modules/@vellumai/service-contracts/package.json

@@ -28,8 +28,7 @@
     "./guardian-requests": "./src/guardian-requests.ts",
     "./platform-credential": "./src/platform-credential.ts",
     "./plan-credit": "./src/plan-credit.ts",
-    "./integration-categories": "./src/integration-categories.ts",
-    "./mcp-oauth": "./src/mcp-oauth.ts"
+    "./integration-categories": "./src/integration-categories.ts"
   },
   "scripts": {
     "typecheck": "bunx tsc --noEmit",

package/node_modules/@vellumai/ces-client/node_modules/@vellumai/service-contracts/src/index.ts

@@ -30,7 +30,6 @@ export * from "./ingress.js";
 export * from "./no-response.js";
 export * from "./plan-credit.js";
 export * from "./integration-categories.js";
-export * from "./mcp-oauth.js";
 export * from "./platform-credential.js";
 export * from "./remote-web-pairing.js";
 export * from "./twilio-ingress.js";

package/node_modules/@vellumai/environments/src/shell.test.ts

@@ -3,27 +3,15 @@ import { describe, expect, test } from "bun:test";
 import {
   buildShellInvocation,
   buildShellSpawnFlags,
-  CHILD_OOM_SCORE_ADJ,
   pathListDelimiter,
   prependUniquePathEntries,
 } from "./shell.js";
 
 describe("buildShellInvocation", () => {
   test("uses Bash on POSIX hosts", () => {
-    expect(buildShellInvocation("printf hello", "darwin")).toEqual({
-      command: "bash",
-      args: ["-c", "--", "printf hello"],
-    });
-  });
-
-  test("raises the shell's OOM-kill priority on Linux", () => {
     expect(buildShellInvocation("printf hello", "linux")).toEqual({
       command: "bash",
-      args: [
-        "-c",
-        "--",
-        `echo ${CHILD_OOM_SCORE_ADJ} 2>/dev/null >/proc/self/oom_score_adj; printf hello`,
-      ],
+      args: ["-c", "--", "printf hello"],
     });
   });
 

package/node_modules/@vellumai/environments/src/shell.ts

@@ -24,23 +24,6 @@ if (-not $__vellumCommandSucceeded) {
 }
 exit 0`;
 
-/**
- * `oom_score_adj` for assistant-owned tool processes. Any positive value
- * outranks a 0-scored process; 1000 makes a tool the kernel's first choice
- * whenever the container runs out of memory, so the assistant itself is never
- * the victim of a runaway command.
- */
-export const CHILD_OOM_SCORE_ADJ = 1000;
-
-/**
- * Prefix for a Bash `-c` script that raises the shell's own OOM-kill priority
- * before the command runs. Every process the command forks inherits it. Bash
- * performs a builtin's redirection itself, so `/proc/self` is the shell.
- * Redirections apply left to right, so stderr goes to /dev/null before the
- * procfs open can fail and complain about it.
- */
-const LINUX_OOM_SCORE_ADJ_PREFIX = `echo ${CHILD_OOM_SCORE_ADJ} 2>/dev/null >/proc/self/oom_score_adj; `;
-
 export function buildShellInvocation(
   command: string,
   hostPlatform: NodeJS.Platform = process.platform,
@@ -62,11 +45,7 @@ export function buildShellInvocation(
       ],
     };
   }
-  const script =
-    hostPlatform === "linux"
-      ? `${LINUX_OOM_SCORE_ADJ_PREFIX}${command}`
-      : command;
-  return { command: "bash", args: ["-c", "--", script] };
+  return { command: "bash", args: ["-c", "--", command] };
 }
 
 /**

package/node_modules/@vellumai/gateway-client/node_modules/@vellumai/service-contracts/package.json

@@ -28,8 +28,7 @@
     "./guardian-requests": "./src/guardian-requests.ts",
     "./platform-credential": "./src/platform-credential.ts",
     "./plan-credit": "./src/plan-credit.ts",
-    "./integration-categories": "./src/integration-categories.ts",
-    "./mcp-oauth": "./src/mcp-oauth.ts"
+    "./integration-categories": "./src/integration-categories.ts"
   },
   "scripts": {
     "typecheck": "bunx tsc --noEmit",

package/node_modules/@vellumai/gateway-client/node_modules/@vellumai/service-contracts/src/index.ts

@@ -30,7 +30,6 @@ export * from "./ingress.js";
 export * from "./no-response.js";
 export * from "./plan-credit.js";
 export * from "./integration-categories.js";
-export * from "./mcp-oauth.js";
 export * from "./platform-credential.js";
 export * from "./remote-web-pairing.js";
 export * from "./twilio-ingress.js";

package/node_modules/@vellumai/service-contracts/package.json

@@ -28,8 +28,7 @@
     "./guardian-requests": "./src/guardian-requests.ts",
     "./platform-credential": "./src/platform-credential.ts",
     "./plan-credit": "./src/plan-credit.ts",
-    "./integration-categories": "./src/integration-categories.ts",
-    "./mcp-oauth": "./src/mcp-oauth.ts"
+    "./integration-categories": "./src/integration-categories.ts"
   },
   "scripts": {
     "typecheck": "bunx tsc --noEmit",

package/node_modules/@vellumai/service-contracts/src/index.ts

@@ -30,7 +30,6 @@ export * from "./ingress.js";
 export * from "./no-response.js";
 export * from "./plan-credit.js";
 export * from "./integration-categories.js";
-export * from "./mcp-oauth.js";
 export * from "./platform-credential.js";
 export * from "./remote-web-pairing.js";
 export * from "./twilio-ingress.js";

package/openapi.yaml

@@ -12482,9 +12482,6 @@ paths:
                           type: number
                         updatedAt:
                           type: number
-                        revision:
-                          description: Bumped by one on every write. Absent from assistants that predate document revisions.
-                          type: number
                       required:
                         - surfaceId
                         - conversationId
@@ -12499,10 +12496,7 @@ paths:
     post:
       operationId: documents_post
       summary: Save a document
-      description:
-        Create or upsert a document (by surfaceId). With `baseRevision`, an existing document is only overwritten
-        while its revision still equals `baseRevision`; otherwise the save is rejected with 409 CONFLICT and
-        `error.details` carries the current `revision`, `title`, and `content`.
+      description: Create or upsert a document (by surfaceId).
       tags:
         - documents
       requestBody:
@@ -12527,13 +12521,6 @@ paths:
                 wordCount:
                   type: number
                   description: Word count
-                baseRevision:
-                  description:
-                    The revision this save was edited from. When set and the stored revision differs, nothing is written and
-                    the request fails with 409. Omit for an unconditional write.
-                  type: integer
-                  minimum: 0
-                  maximum: 9007199254740991
               required:
                 - surfaceId
                 - conversationId
@@ -12553,9 +12540,6 @@ paths:
                     const: true
                   surfaceId:
                     type: string
-                  revision:
-                    description: The document's revision after the save. Absent from assistants that predate document revisions.
-                    type: number
                 required:
                   - success
                   - surfaceId
@@ -12597,9 +12581,6 @@ paths:
                     type: number
                   updatedAt:
                     type: number
-                  revision:
-                    description: Bumped by one on every write. Absent from assistants that predate document revisions.
-                    type: number
                 required:
                   - success
                   - surfaceId
@@ -13026,157 +13007,6 @@ paths:
               schema:
                 type: string
                 format: binary
-  /v1/documents/{id}/revisions:
-    get:
-      operationId: documents_by_id_revisions_get
-      summary: List a document's revisions
-      description: Return the document's history snapshots, newest first, without their content.
-      tags:
-        - documents
-      parameters:
-        - name: id
-          in: path
-          required: true
-          schema:
-            type: string
-      responses:
-        "200":
-          description: Successful response
-          content:
-            application/json:
-              schema:
-                type: object
-                properties:
-                  revisions:
-                    type: array
-                    items:
-                      type: object
-                      properties:
-                        revision:
-                          type: number
-                          description: The document revision this snapshot holds
-                        author:
-                          type: string
-                          enum:
-                            - user
-                            - assistant
-                          description: Who made the edit that replaced this state
-                        createdAt:
-                          type: number
-                          description: When the snapshot was taken
-                        wordCount:
-                          type: number
-                        title:
-                          type: string
-                      required:
-                        - revision
-                        - author
-                        - createdAt
-                        - wordCount
-                        - title
-                      additionalProperties: false
-                required:
-                  - revisions
-                additionalProperties: false
-  /v1/documents/{id}/revisions/{revision}:
-    get:
-      operationId: documents_by_id_revisions_by_revision_get
-      summary: Get a document revision
-      description: Return one history snapshot with its content.
-      tags:
-        - documents
-      parameters:
-        - name: id
-          in: path
-          required: true
-          schema:
-            type: string
-        - name: revision
-          in: path
-          required: true
-          schema:
-            type: string
-      responses:
-        "200":
-          description: Successful response
-          content:
-            application/json:
-              schema:
-                type: object
-                properties:
-                  revision:
-                    type: number
-                    description: The document revision this snapshot holds
-                  author:
-                    type: string
-                    enum:
-                      - user
-                      - assistant
-                    description: Who made the edit that replaced this state
-                  createdAt:
-                    type: number
-                    description: When the snapshot was taken
-                  wordCount:
-                    type: number
-                  title:
-                    type: string
-                  content:
-                    type: string
-                required:
-                  - revision
-                  - author
-                  - createdAt
-                  - wordCount
-                  - title
-                  - content
-                additionalProperties: false
-  /v1/documents/{id}/revisions/{revision}/restore:
-    post:
-      operationId: documents_by_id_revisions_by_revision_restore_post
-      summary: Restore a document revision
-      description:
-        Write a snapshot's title and content back as a new revision. The current state is snapshotted first, so the
-        restore can itself be undone.
-      tags:
-        - documents
-      parameters:
-        - name: id
-          in: path
-          required: true
-          schema:
-            type: string
-        - name: revision
-          in: path
-          required: true
-          schema:
-            type: string
-      responses:
-        "200":
-          description: Successful response
-          content:
-            application/json:
-              schema:
-                type: object
-                properties:
-                  success:
-                    type: boolean
-                    const: true
-                  surfaceId:
-                    type: string
-                  revision:
-                    type: number
-                    description: The document's revision after the restore
-                  title:
-                    type: string
-                  content:
-                    type: string
-                required:
-                  - success
-                  - surfaceId
-                  - revision
-                  - title
-                  - content
-                additionalProperties: false
   /v1/documents/create:
     post:
       operationId: documents_create_post
@@ -13223,9 +13053,6 @@ paths:
                     type: number
                   updatedAt:
                     type: number
-                  revision:
-                    description: Bumped by one on every write. Absent from assistants that predate document revisions.
-                    type: number
                 required:
                   - success
                   - surfaceId
@@ -14462,8 +14289,6 @@ paths:
                         type: boolean
                       appPins:
                         type: boolean
-                      chatsSettings:
-                        type: boolean
                     required:
                       - memoryOptOut
                       - retryLastTurn
@@ -14698,8 +14523,6 @@ paths:
                         type: boolean
                       appPins:
                         type: boolean
-                      chatsSettings:
-                        type: boolean
                     required:
                       - memoryOptOut
                       - retryLastTurn
@@ -38731,41 +38554,6 @@ components:
     ConfigPatchRequest:
       type: object
       properties:
-        conversations:
-          anyOf:
-            - type: object
-              properties:
-                autoArchive:
-                  anyOf:
-                    - type: object
-                      properties:
-                        enabled:
-                          anyOf:
-                            - type: boolean
-                            - type: "null"
-                        afterDays:
-                          anyOf:
-                            - type: number
-                              enum:
-                                - 1
-                                - 7
-                                - 14
-                                - 30
-                            - type: "null"
-                      additionalProperties: {}
-                    - type: "null"
-              additionalProperties: {}
-            - type: "null"
-        notifications:
-          anyOf:
-            - type: object
-              properties:
-                newMessageEnabled:
-                  anyOf:
-                    - type: boolean
-                    - type: "null"
-              additionalProperties: {}
-            - type: "null"
         llm:
           type: object
           properties:
@@ -39499,42 +39287,6 @@ components:
     ConfigGetResponse:
       type: object
       properties:
-        conversations:
-          type: object
-          properties:
-            autoArchive:
-              type: object
-              properties:
-                enabled:
-                  default: false
-                  description: Automatically mark eligible inactive chats Done.
-                  type: boolean
-                afterDays:
-                  default: 7
-                  description: Elapsed days of inactivity before a chat is eligible for Done.
-                  type: number
-                  enum:
-                    - 1
-                    - 7
-                    - 14
-                    - 30
-              required:
-                - enabled
-                - afterDays
-              additionalProperties: {}
-          required:
-            - autoArchive
-          additionalProperties: {}
-        notifications:
-          type: object
-          properties:
-            newMessageEnabled:
-              default: true
-              description: Allow ordinary chat-reply alerts. Does not change unread state or action-required alerts.
-              type: boolean
-          required:
-            - newMessageEnabled
-          additionalProperties: {}
         llm:
           type: object
           properties:

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@vellumai/assistant",
-  "version": "0.12.5-dev.202609260125.8a6a403",
+  "version": "0.12.5-staging.1",
   "license": "MIT",
   "type": "module",
   "exports": {

package/src/__tests__/agent-loop.test.ts

@@ -181,82 +181,6 @@ describe("AgentLoop", () => {
     expect(calls[1].options?.config?.overrideProfile).toBe("quality-optimized");
   });
 
-  test("the auto origin follows the routed override and drops once a switch replaces it", async () => {
-    const toolCallId = "tool-1";
-    const { provider, calls } = createMockProvider([
-      toolUseResponse(toolCallId, "read_file", { path: "/tmp/test.txt" }),
-      textResponse("done"),
-    ]);
-    let overrideProfile = "quality-optimized";
-    const loop = new AgentLoop({
-      provider: provider,
-      systemPrompt: "system",
-      conversationId: "test-conversation",
-      tools: dummyTools,
-      toolExecutor: async () => {
-        // A confirmed profile session mid-turn: the user's pick, not Auto's.
-        overrideProfile = "cost-optimized";
-        return { content: "ok", isError: false };
-      },
-    });
-
-    await loop.run({
-      messages: [userMessage],
-      onEvent: collectEvents([]),
-      trust: { sourceChannel: "vellum", trustClass: "unknown" },
-      requestId: "req-1",
-      callSite: "mainAgent",
-      overrideProfile: "quality-optimized",
-      overrideProfileOrigin: "auto",
-      resolveOverrideProfile: () => overrideProfile,
-    });
-
-    expect(calls).toHaveLength(2);
-    expect(calls[0].options?.config?.overrideProfile).toBe("quality-optimized");
-    expect(calls[0].options?.config?.overrideProfileOrigin).toBe("auto");
-    expect(calls[1].options?.config?.overrideProfile).toBe("cost-optimized");
-    expect(calls[1].options?.config?.overrideProfileOrigin).toBeUndefined();
-  });
-
-  test("a pre-model-call hook keeps the auto origin unless it changes the profile", async () => {
-    let hookProfile: string | null | undefined;
-    registerPlugin({
-      manifest: { name: "profile-hook", version: "0.0.1" },
-      hooks: {
-        "pre-model-call": async (ctx) =>
-          hookProfile === undefined
-            ? ctx
-            : { ...ctx, modelProfile: hookProfile },
-      },
-    });
-    const run = async () => {
-      const { provider, calls } = createMockProvider([textResponse("ok")]);
-      const loop = new AgentLoop({
-        provider: provider,
-        systemPrompt: "system",
-        conversationId: "test-conversation",
-        tools: dummyTools,
-      });
-      await loop.run({
-        messages: [userMessage],
-        onEvent: collectEvents([]),
-        trust: { sourceChannel: "vellum", trustClass: "unknown" },
-        requestId: "req-1",
-        callSite: "mainAgent",
-        overrideProfile: "quality-optimized",
-        overrideProfileOrigin: "auto",
-      });
-      return calls[0].options?.config;
-    };
-
-    hookProfile = undefined;
-    expect((await run())?.overrideProfileOrigin).toBe("auto");
-    hookProfile = "cost-optimized";
-    expect((await run())?.overrideProfileOrigin).toBeUndefined();
-    hookProfile = null;
-    expect((await run())?.overrideProfileOrigin).toBeUndefined();
-  });
-
   test("re-resolves max input tokens before truncating tool results", async () => {
     const toolCallId = "tool-1";
     const toolOutput = "x".repeat(2_500);