@fugood/buttress-server 2.26.0-beta.12 → 2.26.0-beta.14

package/README.md

@@ -69,14 +69,14 @@ The `bricks` CLI is the tool that performs the binding and writes the local stat
 ### Bind a server to a workspace
 
 ```bash
-# Pair the local machine's buttress-server with the workspace of the current bricks-cli profile
+# Pair the local server and provision a workspace DevTools token using the current CLI profile
 bricks buttress bind
 
 # Override the auto-detected server id, give it a friendly name, or write to a custom state dir
 bricks buttress bind --server-id buttress-mac-studio --name "Studio LLM" --state-dir /etc/buttress
 
 # For headless/remote setups: emit state.json to stdout instead of writing to disk
-bricks buttress bind --print > /etc/buttress/state.json
+(umask 077; bricks buttress bind --print > /etc/buttress/state.json)
 ```
 
 The state file (`~/.bricks-cli/buttress/state.json` by default, or `$BRICKS_BUTTRESS_STATE_DIR`) stores:
@@ -84,6 +84,15 @@ The state file (`~/.bricks-cli/buttress/state.json` by default, or `$BRICKS_BUTT
 - `workspace.id` / `workspace.name` — which workspace this server belongs to
 - `workspace.serverId` — the server's stable id (defaults to `buttress-<machineId>`)
 - `workspace.issuerPublicKey` + `workspace.kid` — Ed25519 SPKI used to verify access tokens
+- `devtools` — workspace-scoped DevTools JWT (`k: "da"`), workspace id and expiry,
+  provisioned automatically by `bind` for agent device tools
+
+Normal bind/status output reports only token expiry, never the token. The CLI writes
+state with mode `0600`; `--print` includes credentials and must be treated as a secret.
+Rebinding provisions a fresh token (default lifetime: 30 days) and rotates the announce
+key. Issuance failure stops binding before changing the remote announce key. Rebinding
+or unbinding does not revoke previously issued DevTools JWTs; they remain valid until
+expiry or issuer-key rotation.
 
 **Restart `bricks-buttress` after binding** for the change to take effect — the state file is read once at startup.
 
@@ -207,6 +216,8 @@ Most ggml-llm `[generators.model]` keys can also live in `[runtime]` as defaults
 | `no_extra_bufts`             | boolean                       | Disable extra compute buffer types                                     |
 | `cpu_mask`, `cpu_strict`     | string / boolean              | CPU affinity (advanced)                                                |
 | `devices`                    | string[]                      | Restrict to specific GGML devices                                      |
+| `split_mode`                 | `"none"` / `"layer"` / `"row"` / `"tensor"` | ggml-llm GPU splitting: single GPU, layers + KV (default), weight rows, or weights + KV (experimental tensor parallelism) |
+| `tensor_split`               | number[]\|string              | ggml-llm per-GPU proportions, e.g. `[3, 1]` or `"3,1"`; unset uses the native automatic split |
 | Speculative keys             | various                       | `speculative`, `spec_type`, `spec_draft_n_max/n_min/p_min/p_split`, plus draft-model GPU/cache settings |
 
 ### `[autodiscover]`
@@ -268,6 +279,10 @@ Additional keys honored by **ggml-llm**, **ggml-stt** and **ggml-tts** (mlx-llm
 
 Loads a GGUF LLM. Runtime keys above can be overridden per-generator under `[generators.model]`; `[generators.backend]` only controls backend selection and resource planning.
 
+GPU splitting (`split_mode`, `tensor_split`) is currently configurable only in Buttress.
+TODO: Expose these options in the LLM (GGML) generator. The experimental `tensor` mode
+requires a compatible GPU backend and model architecture.
+
 **`[generators.backend]`**
 
 | Key                   | Type     | Default                                       | Notes                                                            |
@@ -283,6 +298,7 @@ Loads a GGUF LLM. Runtime keys above can be overridden per-generator under `[gen
 | ----------------------------------------------------------------------------- | ---------------- | -------------------------------------------------------------------- |
 | `n_ctx`                                                                       | number           | Context window. Auto-capped at the model's training context.         |
 | `n_gpu_layers`                                                                | number\|`"auto"` | Layers offloaded to GPU (default `"auto"`)                           |
+| `split_mode`, `tensor_split`                                                   | string / number[]\|string | Per-model overrides for the GPU split defaults in `[runtime]`       |
 | `n_batch`                                                                     | number           | Prompt batch size (default `512`)                                    |
 | `embedding`                                                                   | boolean          | Enable llama.cpp embedding mode for a dedicated embedding model (default `false`) |
 | `pooling_type`                                                                | string           | Optional embedding pooling override such as `mean`                   |
@@ -355,12 +371,21 @@ spec_draft_cache_type_v = "f16"
 
 Loads a Whisper GGML model for speech-to-text.
 
+With `@fugood/whisper.node` 1.2.0-rc.0, `snapdragon` enables experimental
+Whisper Hexagon NPU acceleration on Linux arm64 with a working Qualcomm
+HTP/FastRPC runtime. This variant does not include OpenCL. Install it with
+`--ggml-variant=snapdragon` (the default CPU package remains available for fallback).
+The platform package includes the HTP libraries; the loader defaults
+`ADSP_LIBRARY_PATH` to that package directory and `GGML_HEXAGON_NDEV` to `16`,
+without overriding explicit environment values. Hexagon always enables flash
+attention, even when `use_flash_attn` is off.
+
 **`[generators.backend]`**
 
 | Key                   | Type     | Default                       | Notes                              |
 | --------------------- | -------- | ----------------------------- | ---------------------------------- |
-| `variant`             | string   | auto                          | `cuda` / `vulkan` / `default`      |
-| `variant_preference`  | string[] | `["cuda","vulkan","default"]` | Probe order                        |
+| `variant`             | string   | auto                          | `cuda` / `vulkan` / `snapdragon` / `default` |
+| `variant_preference`  | string[] | `["cuda","vulkan","snapdragon","default"]` | Probe order                  |
 | `gpu_memory_fraction` | number   | `0.85`                        |                                    |
 | `cpu_memory_fraction` | number   | `0.5`                         |                                    |
 
@@ -385,7 +410,7 @@ Loads a Whisper GGML model for speech-to-text.
 [[generators]]
 type = "ggml-stt"
 [generators.backend]
-variant_preference = ["cuda", "vulkan", "default"]
+variant_preference = ["cuda", "vulkan", "snapdragon", "default"]
 [generators.model]
 repo_id = "BricksDisplay/whisper-ggml"
 filename = "ggml-large-v3-turbo-q8_0.bin"
@@ -914,16 +939,23 @@ fail closed unless marked `optional = true`.
 
 For example, scan with `{}` then call `devtools` with
 `{ "action": "tree", "address": "192.168.1.42", "port": 19851 }`.
-Supply `passcode` or a workspace DevTools `accessToken` if required; otherwise the
-transport uses the device's advertised Chrome Inspect credential. Buttress does **not**
-read the host's CLI login or mint tokens. Calls open/close their own connection;
+After `bricks buttress bind` and a restart, the workspace credential is injected
+server-side automatically when the device advertises the same workspace and issuer.
+The token is validated for signature, kind and expiry before use; inbound Buttress
+JWTs (`k: "ba"`) are never forwarded. Older bindings without the credential, or expired
+credentials, require rebinding and a restart. The server does not read the CLI login
+or renew tokens itself. Local previews can still use advertised inspect credentials;
+explicit `passcode` / `accessToken` arguments remain optional overrides. Calls open/close their own connection;
 avoid parallel calls or other inspectors on the same device (one CDP client at a time).
 
 **Security:** enabling this option grants network discovery and device control,
 including arbitrary JavaScript evaluation on reachable authenticated devices. It is
-not read-only or a network sandbox. Tool arguments/results (including credentials,
-console output and device data) enter model context and persisted session transcripts;
-only grant this to trusted agents/callers. Disable with `local_devices = false` and
+not read-only or a network sandbox. Automatic authentication requires a **trusted LAN**:
+workspace/issuer hints from device info are not cryptographic peer authentication.
+The stored token is never inserted into model arguments and is redacted from text
+results/errors before truncation. Explicit credentials supplied in arguments still
+enter session transcripts, as do console output and device data; prefer automatic
+authentication and only grant this to trusted agents/callers. Disable with `local_devices = false` and
 restart Buttress. Aborting closes active CDP connections; discovery completes its
 bounded scan before returning the abort result.
 

package/config/sample.toml

@@ -78,6 +78,12 @@ cache_type_k = "q8_0"
 cache_type_v = "q8_0"
 n_threads = 6
 
+# Optional ggml-llm multi-GPU defaults (also supported under [generators.model]).
+# split_mode = "layer"   # none / layer (default) / row / tensor (experimental)
+# tensor_split = [3, 1]  # per-GPU proportions; also accepts "3,1"
+# Tensor mode requires a compatible GPU backend and model architecture.
+# TODO: Expose these Buttress-only options in the LLM (GGML) generator.
+
 # Delay before releasing context when refCount reaches 0 (default: 10000ms)
 # Set to 0 for immediate release
 context_release_delay_ms = 30000

package/lib/agent/device-auth.d.ts

@@ -0,0 +1,3 @@
+import type { WorkspaceState } from '../utils/workspaceState';
+/** Select only the bound workspace's credential; never reuse an inbound Buttress (k=ba) JWT. */
+export declare const resolveDeviceAccessToken: (address: string, port: number, state?: WorkspaceState, signal?: AbortSignal) => Promise<string | undefined>;

package/lib/agent/device-tools.d.ts

@@ -1,4 +1,5 @@
 import type { AgentFunctionTool } from './tools';
 import type { AgentDefinition } from './types';
+import type { WorkspaceState } from '../utils/workspaceState';
 /** Two tools regardless of device count; no remote tool-list expansion or persistent sockets. */
-export declare const buildLocalDeviceTools: (agent: Pick<AgentDefinition, 'localDevices'>, configDir: string) => AgentFunctionTool[];
+export declare const buildLocalDeviceTools: (agent: Pick<AgentDefinition, 'localDevices'>, configDir: string, workspaceState?: WorkspaceState) => AgentFunctionTool[];