mirror of
https://github.com/awizemann/scarf.git
synced 2026-05-08 02:14:37 +00:00
feat(hermes-v12): provider catalog + auxiliary swap (Phase B)
Adds the five v0.12 inference providers to ModelCatalogService.overlayOnlyProviders so the model picker reaches them. IDs match HERMES_OVERLAYS verbatim: - gmi → GMI Cloud (api_key) - azure-foundry → Azure AI Foundry (api_key) - lmstudio → LM Studio (api_key, promoted from custom-endpoint alias) - minimax-oauth → MiniMax (OAuth, oauth_external) - tencent-tokenhub → Tencent TokenHub (api_key) Auxiliary tasks: drop the `flush_memories` row (Hermes removed it entirely in v0.12) and add `auxiliary.curator` so users can configure the model the autonomous curator's review fork uses. The Curator row is gated on HermesCapabilities.hasCuratorAux, so v0.11 hosts don't see a control that writes a key Hermes ignores. AuxiliarySettings, the YAML parser, and HealthViewModel's Tool Gateway breakdown are all updated. Side fixes: - CredentialPoolsGatingTests was missing `import ScarfCore` after ModelCatalogService moved to the package (broke the test target's compile against pure-Mac scarf). - Promoted `ModelCatalogService.overlayOnlyProviders` to public so the new `v012OverlayProvidersCarryCorrectAuthTypes` lock-in test can reach it. Tests: 14 ToolGateway tests pass; 209 ScarfCore tests pass; both Mac and iOS schemes build clean. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Alan Wizemann
@@ -258,7 +258,13 @@ public struct VoiceSettings: Sendable, Equatable {
|
||||
)
|
||||
}
|
||||
|
||||
/// Eight sub-models that share the same provider/model/base_url/api_key/timeout shape.
|
||||
/// Per-task auxiliary model overrides.
|
||||
///
|
||||
/// `flush_memories` was removed entirely in Hermes v0.12 (the underlying
|
||||
/// task no longer exists), so the corresponding field was dropped here.
|
||||
/// `curator` was added in v0.12 — Curator's review fork uses its own
|
||||
/// model so users can keep main-model spend separate from background
|
||||
/// maintenance.
|
||||
public struct AuxiliarySettings: Sendable, Equatable {
|
||||
public var vision: AuxiliaryModel
|
||||
public var webExtract: AuxiliaryModel
|
||||
@@ -267,7 +273,8 @@ public struct AuxiliarySettings: Sendable, Equatable {
|
||||
public var skillsHub: AuxiliaryModel
|
||||
public var approval: AuxiliaryModel
|
||||
public var mcp: AuxiliaryModel
|
||||
public var flushMemories: AuxiliaryModel
|
||||
/// v0.12+; pre-v0.12 Hermes installs ignore this slot.
|
||||
public var curator: AuxiliaryModel
|
||||
|
||||
|
||||
public init(
|
||||
@@ -278,7 +285,7 @@ public struct AuxiliarySettings: Sendable, Equatable {
|
||||
skillsHub: AuxiliaryModel,
|
||||
approval: AuxiliaryModel,
|
||||
mcp: AuxiliaryModel,
|
||||
flushMemories: AuxiliaryModel
|
||||
curator: AuxiliaryModel
|
||||
) {
|
||||
self.vision = vision
|
||||
self.webExtract = webExtract
|
||||
@@ -287,7 +294,7 @@ public struct AuxiliarySettings: Sendable, Equatable {
|
||||
self.skillsHub = skillsHub
|
||||
self.approval = approval
|
||||
self.mcp = mcp
|
||||
self.flushMemories = flushMemories
|
||||
self.curator = curator
|
||||
}
|
||||
public nonisolated static let empty = AuxiliarySettings(
|
||||
vision: .empty,
|
||||
@@ -297,7 +304,7 @@ public struct AuxiliarySettings: Sendable, Equatable {
|
||||
skillsHub: .empty,
|
||||
approval: .empty,
|
||||
mcp: .empty,
|
||||
flushMemories: .empty
|
||||
curator: .empty
|
||||
)
|
||||
}
|
||||
|
||||
|
||||
@@ -122,7 +122,7 @@ public extension HermesConfig {
|
||||
skillsHub: aux("skills_hub"),
|
||||
approval: aux("approval"),
|
||||
mcp: aux("mcp"),
|
||||
flushMemories: aux("flush_memories")
|
||||
curator: aux("curator")
|
||||
)
|
||||
|
||||
let security = SecuritySettings(
|
||||
|
||||
@@ -425,15 +425,17 @@ public struct ModelCatalogService: Sendable {
|
||||
|
||||
// MARK: - Hermes overlay providers
|
||||
|
||||
/// The six providers Hermes surfaces via `hermes model` that have no
|
||||
/// The 11 providers Hermes surfaces via `hermes model` that have no
|
||||
/// entry in `models_dev_cache.json` (models.dev doesn't mirror them).
|
||||
/// Mirrors the overlay-only subset of `HERMES_OVERLAYS` in
|
||||
/// `hermes-agent/hermes_cli/providers.py`. The other ~19 overlay entries
|
||||
/// `hermes-agent/hermes_cli/providers.py`. The other overlay entries
|
||||
/// already ship in the cache and only add augmentation (base-URL
|
||||
/// override, extra env vars) that Scarf doesn't currently display.
|
||||
///
|
||||
/// Keep this in sync with the Python side on Hermes version bumps.
|
||||
static let overlayOnlyProviders: [String: HermesProviderOverlay] = [
|
||||
/// Keep this in sync with the Python side on Hermes version bumps —
|
||||
/// see `ToolGatewayTests.v012OverlayProvidersCarryCorrectAuthTypes`
|
||||
/// for the auth-type lock-in.
|
||||
public static let overlayOnlyProviders: [String: HermesProviderOverlay] = [
|
||||
"nous": HermesProviderOverlay(
|
||||
displayName: "Nous Portal",
|
||||
baseURL: "https://inference-api.nousresearch.com/v1",
|
||||
@@ -476,6 +478,53 @@ public struct ModelCatalogService: Sendable {
|
||||
subscriptionGated: false,
|
||||
docURL: nil
|
||||
),
|
||||
// -- v0.12 additions ---------------------------------------------
|
||||
// Hermes v2026.4.30 added five overlay-only providers that
|
||||
// models.dev doesn't mirror. Provider IDs match HERMES_OVERLAYS
|
||||
// verbatim — drift here means the picker can't reach them.
|
||||
"gmi": HermesProviderOverlay(
|
||||
displayName: "GMI Cloud",
|
||||
baseURL: "https://api.gmi-serving.com/v1",
|
||||
authType: .apiKey,
|
||||
subscriptionGated: false,
|
||||
docURL: nil
|
||||
),
|
||||
"azure-foundry": HermesProviderOverlay(
|
||||
displayName: "Azure AI Foundry",
|
||||
// Base URL is per-tenant — Hermes resolves it from the
|
||||
// AZURE_FOUNDRY_BASE_URL env var at runtime. Leave nil so the
|
||||
// settings UI shows "Tenant URL — set via env" instead of a
|
||||
// misleading default.
|
||||
baseURL: nil,
|
||||
authType: .apiKey,
|
||||
subscriptionGated: false,
|
||||
docURL: nil
|
||||
),
|
||||
"lmstudio": HermesProviderOverlay(
|
||||
displayName: "LM Studio",
|
||||
// v0.12 promotes LM Studio from custom-endpoint alias to a
|
||||
// first-class provider. 1234 is the LM Studio default port;
|
||||
// users with a non-default port set LM_BASE_URL.
|
||||
baseURL: "http://127.0.0.1:1234/v1",
|
||||
authType: .apiKey,
|
||||
subscriptionGated: false,
|
||||
docURL: nil
|
||||
),
|
||||
"minimax-oauth": HermesProviderOverlay(
|
||||
displayName: "MiniMax (OAuth)",
|
||||
baseURL: "https://api.minimax.io/anthropic",
|
||||
authType: .oauthExternal,
|
||||
subscriptionGated: false,
|
||||
docURL: nil
|
||||
),
|
||||
"tencent-tokenhub": HermesProviderOverlay(
|
||||
displayName: "Tencent TokenHub",
|
||||
// Resolved from TOKENHUB_BASE_URL at runtime.
|
||||
baseURL: nil,
|
||||
authType: .apiKey,
|
||||
subscriptionGated: false,
|
||||
docURL: nil
|
||||
),
|
||||
]
|
||||
}
|
||||
|
||||
|
||||
@@ -129,7 +129,7 @@ struct HermesFileService: Sendable {
|
||||
skillsHub: aux("skills_hub"),
|
||||
approval: aux("approval"),
|
||||
mcp: aux("mcp"),
|
||||
flushMemories: aux("flush_memories")
|
||||
curator: aux("curator")
|
||||
)
|
||||
|
||||
let security = SecuritySettings(
|
||||
|
||||
@@ -180,7 +180,7 @@ final class HealthViewModel {
|
||||
("skills_hub", config.auxiliary.skillsHub.provider),
|
||||
("approval", config.auxiliary.approval.provider),
|
||||
("mcp", config.auxiliary.mcp.provider),
|
||||
("flush_memories", config.auxiliary.flushMemories.provider),
|
||||
("curator", config.auxiliary.curator.provider),
|
||||
].filter { $0.1 == "nous" }.map(\.0)
|
||||
if !auxOnNous.isEmpty {
|
||||
checks.append(HealthCheck(
|
||||
|
||||
@@ -9,25 +9,41 @@ import ScarfCore
|
||||
/// (subscription-routed) and `auto` (inherit main provider) — Hermes derives
|
||||
/// the gateway routing from that single field; there is no separate
|
||||
/// `use_gateway` key to write.
|
||||
///
|
||||
/// v0.12 dropped the `flush_memories` aux task (the underlying memory
|
||||
/// pipeline was rewritten upstream) and added `curator` (the autonomous
|
||||
/// skill-maintenance review fork). The Curator row only appears when
|
||||
/// `HermesCapabilities.hasCuratorAux` is set so v0.11 hosts don't see a
|
||||
/// row that writes a key Hermes ignores.
|
||||
struct AuxiliaryTab: View {
|
||||
@Bindable var viewModel: SettingsViewModel
|
||||
|
||||
@Environment(\.serverContext) private var serverContext
|
||||
@Environment(\.hermesCapabilities) private var capabilitiesStore
|
||||
@State private var subscription: NousSubscriptionState = .absent
|
||||
@State private var showNousSignIn: Bool = false
|
||||
|
||||
// Keyed by the config path name — matches `auxiliary.<task>.*` in config.yaml.
|
||||
private let tasks: [(key: String, title: LocalizedStringKey, icon: String)] = [
|
||||
// Static base list; the v0.12-only `curator` row is appended at render
|
||||
// time when the target Hermes supports it.
|
||||
private let baseTasks: [(key: String, title: LocalizedStringKey, icon: String)] = [
|
||||
("vision", "Vision", "eye"),
|
||||
("web_extract", "Web Extract", "doc.richtext"),
|
||||
("compression", "Compression", "arrow.down.right.and.arrow.up.left.circle"),
|
||||
("session_search", "Session Search", "magnifyingglass"),
|
||||
("skills_hub", "Skills Hub", "books.vertical"),
|
||||
("approval", "Approval", "checkmark.seal"),
|
||||
("mcp", "MCP", "puzzlepiece"),
|
||||
("flush_memories", "Flush Memories", "trash.slash")
|
||||
("mcp", "MCP", "puzzlepiece")
|
||||
]
|
||||
|
||||
private var tasks: [(key: String, title: LocalizedStringKey, icon: String)] {
|
||||
var t = baseTasks
|
||||
if capabilitiesStore?.capabilities.hasCuratorAux ?? false {
|
||||
t.append(("curator", "Curator", "sparkles"))
|
||||
}
|
||||
return t
|
||||
}
|
||||
|
||||
var body: some View {
|
||||
Text("Auxiliary tasks use separate, typically cheaper models. Leave Provider as `auto` to inherit the main provider.")
|
||||
.font(.caption)
|
||||
@@ -94,7 +110,7 @@ struct AuxiliaryTab: View {
|
||||
case "skills_hub": return viewModel.config.auxiliary.skillsHub
|
||||
case "approval": return viewModel.config.auxiliary.approval
|
||||
case "mcp": return viewModel.config.auxiliary.mcp
|
||||
case "flush_memories": return viewModel.config.auxiliary.flushMemories
|
||||
case "curator": return viewModel.config.auxiliary.curator
|
||||
default: return .empty
|
||||
}
|
||||
}
|
||||
|
||||
@@ -198,9 +198,9 @@ private struct ContextBoundRoot: View {
|
||||
@State private var chatViewModel: ChatViewModel
|
||||
/// Per-window snapshot of the target Hermes installation's capability
|
||||
/// flags. Drives sidebar visibility (Curator, Kanban only on v0.12+),
|
||||
/// settings rows (flush_memories aux dropped on v0.12), and version
|
||||
/// banners. Refreshes once on init; explicit `refresh()` call rerun
|
||||
/// after a `hermes update`.
|
||||
/// settings rows (curator aux added on v0.12), and version banners.
|
||||
/// Refreshes once on init; explicit `refresh()` call rerun after a
|
||||
/// `hermes update`.
|
||||
@State private var capabilities: HermesCapabilitiesStore
|
||||
|
||||
init(context: ServerContext) {
|
||||
|
||||
@@ -1,5 +1,6 @@
|
||||
import Testing
|
||||
import Foundation
|
||||
import ScarfCore
|
||||
@testable import scarf
|
||||
|
||||
/// Tests that ``CredentialPoolsOAuthGate`` steers each known provider to
|
||||
|
||||
@@ -55,11 +55,37 @@ import ScarfCore
|
||||
#expect(ids.contains("nous"), "Nous Portal must appear after overlay merge")
|
||||
#expect(ids.contains("openai-codex"), "OpenAI Codex overlay must appear")
|
||||
#expect(ids.contains("qwen-oauth"), "Qwen OAuth overlay must appear")
|
||||
// v0.12 additions — IDs must match HERMES_OVERLAYS in
|
||||
// hermes-agent/hermes_cli/providers.py exactly. Drift here
|
||||
// means the picker can't reach the new providers.
|
||||
#expect(ids.contains("gmi"), "GMI Cloud overlay must appear (v0.12)")
|
||||
#expect(ids.contains("azure-foundry"), "Azure AI Foundry overlay must appear (v0.12)")
|
||||
#expect(ids.contains("lmstudio"), "LM Studio overlay must appear (v0.12)")
|
||||
#expect(ids.contains("minimax-oauth"), "MiniMax OAuth overlay must appear (v0.12)")
|
||||
#expect(ids.contains("tencent-tokenhub"), "Tencent TokenHub overlay must appear (v0.12)")
|
||||
// Cached providers still present.
|
||||
#expect(ids.contains("anthropic"))
|
||||
#expect(ids.contains("openai"))
|
||||
}
|
||||
|
||||
@Test func v012OverlayProvidersCarryCorrectAuthTypes() throws {
|
||||
// The auth-type drives whether Settings shows an API-key field,
|
||||
// an OAuth flow, or external-process wiring. Locking the v0.12
|
||||
// additions here so a typo doesn't quietly land users in the
|
||||
// wrong setup flow.
|
||||
let overlays = ModelCatalogService.overlayOnlyProviders
|
||||
#expect(overlays["gmi"]?.authType == .apiKey)
|
||||
#expect(overlays["azure-foundry"]?.authType == .apiKey)
|
||||
#expect(overlays["lmstudio"]?.authType == .apiKey)
|
||||
#expect(overlays["minimax-oauth"]?.authType == .oauthExternal)
|
||||
#expect(overlays["tencent-tokenhub"]?.authType == .apiKey)
|
||||
// None of the v0.12 additions are subscription-gated (only Nous
|
||||
// Portal is).
|
||||
for id in ["gmi", "azure-foundry", "lmstudio", "minimax-oauth", "tencent-tokenhub"] {
|
||||
#expect(overlays[id]?.subscriptionGated == false, "\(id) shouldn't be subscription-gated")
|
||||
}
|
||||
}
|
||||
|
||||
@Test func nousPortalSortsFirst() throws {
|
||||
let path = try writeCacheFixture()
|
||||
let service = ModelCatalogService(path: path)
|
||||
|
||||
Reference in New Issue
Block a user