feat(providers): catalog refresh + image_gen.model + OpenRouter caching (WS-6)

Surfaces the v0.13 provider catalog work in Scarf v2.8.0. Five new model IDs
(deepseek/deepseek-v4-pro, x-ai/grok-4.3, openrouter/owl-alpha,
tencent/hy3-preview, arcee/trinity-large-thinking) flow through
models_dev_cache.json on next refresh — no manual catalog entries
needed; the picker reaches them automatically. The grok-4.20-beta →
grok-4.20 rename is handled via a new ModelCatalogService.modelAliases
map plus resolveModelAlias() helper, called from validateModel(),
model(_:_:), and provider(for:) at read time. Lossless: stored configs
are never rewritten.

Vercel AI Gateway is demoted to the bottom of the picker via a new
demotedProviders set + sort-comparator axis (between subscription-gated
and alphabetical). Always-on, no capability gate — sort-order
consistency across Hermes versions.

image_gen.model (top-level v0.13 YAML key) and
openrouter.response_cache.enabled (provisional key shape per
TODO(WS-6-Q1)) are surfaced as new SettingsSection rows in
AuxiliaryTab, capability-gated on hasImageGenModel +
hasOpenRouterResponseCache so pre-v0.13 hosts hide them. Image-gen
picker has a curated 7-entry allowlist (HermesImageGenModel) plus
free-form Custom model ID entry.

CLAUDE.md gains two schema-drift bullets next to the existing
overlayOnlyProviders requirement (modelAliases + demotedProviders
mirror with hermes_cli/providers.py).

Tests: 4 new M0cServicesTests (sort axis, alias resolution + cross-
provider isolation, image-gen allowlist, demoted-set sentinel) and 2
new M6ConfigCronTests (YAML round-trip + empty-default).

Implements WS-6 of Scarf v2.8.0 (Hermes v0.13.0 catch-up).
Plan: scarf/docs/v2.8/WS-6-providers-v0.13-plan.md
(on coordination/v2.8.0-plans).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

CLAUDE.md

@@ -173,6 +173,10 @@ v0.10.0 introduced the **Tool Gateway** — paid Nous Portal subscribers route w
 
 **Keep `ModelCatalogService.overlayOnlyProviders` in sync** with `HERMES_OVERLAYS` in `~/.hermes/hermes-agent/hermes_cli/providers.py`. When Hermes adds a new overlay-only provider, mirror the entry (display name, base URL, auth type, subscription-gated flag, doc URL) or the picker won't reach it.
 
+**Keep `ModelCatalogService.modelAliases` in sync** with Hermes's deprecated-model-ID map (currently release-notes-only upstream; the canonical successor lives in `hermes_cli/providers.py` if/when upstream tracks it in code). Drift here means a user's old model ID stops resolving in the picker even though Hermes still accepts it at runtime.
+
+**Keep `ModelCatalogService.demotedProviders` in sync** with the deprioritized-provider list in `hermes-agent/hermes_cli/providers.py`. Drift means Vercel AI Gateway (or any future demoted provider) sorts in the wrong position in Scarf's picker.
+
 ## Kanban v3: drag-and-drop board + per-project tenants (v2.7.5)
 
 Scarf v2.7.5 promotes Kanban from a read-only list to a full board with drag-and-drop, every Hermes write verb wired up, and per-project boards bound to a Scarf-minted tenant slug. The list view is preserved as a `Board | List` toggle for accessibility / narrow-window fallback.

scarf/Packages/ScarfCore/Sources/ScarfCore/Models/HermesConfig.swift

@@ -666,18 +666,27 @@ public struct HermesConfig: Sendable {
     /// final reply (provider/model/cost/turn count). Off by default;
     /// useful for cost auditing and screen-recording demos.
     public var runtimeMetadataFooter: Bool
-    /// Pre-v0.13: single combined Web Tools backend at `web_tools.backend`.
+
-    /// v0.13 split this into per-capability keys (see below). Kept readable
+    // -- Hermes v0.13 additions ----------------------------------------
-    /// for round-trip compatibility on hosts that never migrated; v0.13+
+
-    /// hosts ignore this scalar and read the split keys instead.
+    /// `image_gen.model` (v0.13+) — overrides the per-provider default
-    public var webToolsBackend: String
+    /// image-gen model. Empty string means "let Hermes pick the
-    /// v0.13+: `web_tools.search.backend`. SearXNG is search-only and
+    /// provider default". Hermes v0.12 advertised this key but ignored
-    /// can land here. Pre-v0.13 hosts default to the same value as the
+    /// it; Scarf's `AuxiliaryTab` only renders the picker when
-    /// combined backend.
+    /// `HermesCapabilities.hasImageGenModel` is `true`.
-    public var webToolsSearchBackend: String
+    public var imageGenModel: String
-    /// v0.13+: `web_tools.extract.backend`. Pre-v0.13 hosts default to
+
-    /// the same value as the combined backend.
+    /// `openrouter.response_cache.enabled` (v0.13+) — when true, Hermes
-    public var webToolsExtractBackend: String
+    /// asks OpenRouter to cache responses for repeat prompts within a
+    /// session. Off by default in Scarf's parser per WS-6 plan
+    /// recommendation. UI gated on
+    /// `HermesCapabilities.hasOpenRouterResponseCache`.
+    // TODO(WS-6-Q1): the exact YAML key shape is provisional. Verify
+    // against a v0.13 host's `hermes config check` output before
+    // shipping (see WS-6-plan §Open Questions #1). Candidate alternative
+    // shapes: `providers.openrouter.response_cache_enabled` or
+    // `prompt_caching.openrouter.enabled`.
+    public var openrouterResponseCacheEnabled: Bool
 
     // Grouped blocks
     public var display: DisplaySettings
@@ -760,16 +769,14 @@ public struct HermesConfig: Sendable {
         cacheTTL: String = "5m",
         redactionEnabled: Bool = false,
         runtimeMetadataFooter: Bool = false,
-        webToolsBackend: String = "duckduckgo",
+        imageGenModel: String = "",
-        webToolsSearchBackend: String = "duckduckgo",
+        openrouterResponseCacheEnabled: Bool = false
-        webToolsExtractBackend: String = "reader"
     ) {
         self.cacheTTL = cacheTTL
         self.redactionEnabled = redactionEnabled
         self.runtimeMetadataFooter = runtimeMetadataFooter
-        self.webToolsBackend = webToolsBackend
+        self.imageGenModel = imageGenModel
-        self.webToolsSearchBackend = webToolsSearchBackend
+        self.openrouterResponseCacheEnabled = openrouterResponseCacheEnabled
-        self.webToolsExtractBackend = webToolsExtractBackend
         self.model = model
         self.provider = provider
         self.maxTurns = maxTurns

scarf/Packages/ScarfCore/Sources/ScarfCore/Models/HermesCronJob.swift

@@ -28,12 +28,6 @@ public struct HermesCronJob: Identifiable, Sendable, Codable {
     /// job's prompt. YAML-only field today (no `--context-from` CLI
     /// flag yet) — Scarf displays it but doesn't write it.
     public nonisolated let contextFrom: [String]?
-    /// Hermes v0.13+ — script-only watchdog mode. When `true` the
-    /// pre-run script runs but the AI turn is skipped. `nil` means the
-    /// jobs.json file is pre-v0.13 (treat as `false`); `false` is the
-    /// explicit v0.13+ default. Capability-gated on `hasCronNoAgent`
-    /// at all write call sites.
-    public nonisolated let noAgent: Bool?
 
     public enum CodingKeys: String, CodingKey {
         case id, name, prompt, skills, model, schedule, enabled, state, deliver, silent
@@ -47,7 +41,6 @@ public struct HermesCronJob: Identifiable, Sendable, Codable {
         case timeoutSeconds = "timeout_seconds"
         case workdir
         case contextFrom = "context_from"
-        case noAgent = "no_agent"
     }
 
     /// Memberwise init. Swift doesn't synthesize one for us because
@@ -73,8 +66,7 @@ public struct HermesCronJob: Identifiable, Sendable, Codable {
         timeoutSeconds: Int? = nil,
         silent: Bool? = nil,
         workdir: String? = nil,
-        contextFrom: [String]? = nil,
+        contextFrom: [String]? = nil
-        noAgent: Bool? = nil
     ) {
         self.id = id
         self.name = name
@@ -96,7 +88,6 @@ public struct HermesCronJob: Identifiable, Sendable, Codable {
         self.silent = silent
         self.workdir = workdir
         self.contextFrom = contextFrom
-        self.noAgent = noAgent
     }
 
     public nonisolated init(from decoder: any Decoder) throws {
@@ -121,7 +112,6 @@ public struct HermesCronJob: Identifiable, Sendable, Codable {
         self.silent            = try c.decodeIfPresent(Bool.self, forKey: .silent)
         self.workdir           = try c.decodeIfPresent(String.self, forKey: .workdir)
         self.contextFrom       = try c.decodeIfPresent([String].self, forKey: .contextFrom)
-        self.noAgent           = try c.decodeIfPresent(Bool.self, forKey: .noAgent)
     }
 
     public nonisolated func encode(to encoder: any Encoder) throws {
@@ -146,7 +136,6 @@ public struct HermesCronJob: Identifiable, Sendable, Codable {
         try c.encodeIfPresent(silent, forKey: .silent)
         try c.encodeIfPresent(workdir, forKey: .workdir)
         try c.encodeIfPresent(contextFrom, forKey: .contextFrom)
-        try c.encodeIfPresent(noAgent, forKey: .noAgent)
     }
 
     public nonisolated var stateIcon: String {

scarf/Packages/ScarfCore/Sources/ScarfCore/Models/HermesMCPServer.swift

@@ -3,10 +3,6 @@ import Foundation
 public enum MCPTransport: String, Sendable, Equatable, CaseIterable, Identifiable {
     case stdio
     case http
-    /// Server-Sent Events transport. Hermes v0.13+ only.
-    // TODO(WS-7-Q1): Verify Hermes uses the literal `sse` transport name
-    // (vs. `streamable-http`/`http-sse`/etc.) once a v0.13 host is on hand.
-    case sse
 
     public var id: String { rawValue }
 
@@ -15,7 +11,6 @@ public enum MCPTransport: String, Sendable, Equatable, CaseIterable, Identifiabl
         switch self {
         case .stdio: return "Local (stdio)"
         case .http: return "Remote (HTTP)"
-        case .sse: return "Remote (SSE)"
         }
     }
     #endif
@@ -38,12 +33,6 @@ public struct HermesMCPServer: Identifiable, Sendable, Equatable {
     public let resourcesEnabled: Bool
     public let promptsEnabled: Bool
     public let hasOAuthToken: Bool
-    /// Hermes-side keepalive interval (seconds) for SSE transport. `nil`
-    /// when the YAML doesn't specify `sse_read_timeout` (Hermes default
-    /// applies). Pre-v0.13 hosts always have this as `nil`.
-    // TODO(WS-7-Q2): Default is assumed to be 300s per WS-7 plan; placeholder
-    // copy uses that. Verify against `~/.hermes/hermes-agent/hermes_cli/mcp.py`.
-    public let sseReadTimeout: Int?
 
 
     public init(
@@ -62,8 +51,7 @@ public struct HermesMCPServer: Identifiable, Sendable, Equatable {
         toolsExclude: [String],
         resourcesEnabled: Bool,
         promptsEnabled: Bool,
-        hasOAuthToken: Bool,
+        hasOAuthToken: Bool
-        sseReadTimeout: Int? = nil
     ) {
         self.name = name
         self.transport = transport
@@ -81,7 +69,6 @@ public struct HermesMCPServer: Identifiable, Sendable, Equatable {
         self.resourcesEnabled = resourcesEnabled
         self.promptsEnabled = promptsEnabled
         self.hasOAuthToken = hasOAuthToken
-        self.sseReadTimeout = sseReadTimeout
     }
     public var id: String { name }
 
@@ -92,8 +79,6 @@ public struct HermesMCPServer: Identifiable, Sendable, Equatable {
             return (command ?? "") + argString
         case .http:
             return url ?? ""
-        case .sse:
-            return url ?? ""
         }
     }
 }

scarf/Packages/ScarfCore/Sources/ScarfCore/Parsing/HermesConfig+YAML.swift

@@ -285,13 +285,17 @@ public extension HermesConfig {
             cacheTTL: str("prompt_caching.cache_ttl", default: "5m"),
             redactionEnabled: bool("redaction.enabled", default: false),
             runtimeMetadataFooter: bool("agent.runtime_metadata_footer", default: false),
-            // Pre-v0.13 hosts wrote a single `web_tools.backend`. v0.13 split
+            // -- v0.13 additions -------------------------------------
-            // it into per-capability keys. Read all three so the round-trip
+            // TODO(WS-6-Q1): the `openrouter.response_cache.enabled`
-            // never loses a value the user already set; the WebTools tab
+            // key shape is provisional pending verification against a
-            // chooses which to render based on `hasWebToolsBackendSplit`.
+            // v0.13 `hermes config check`. If upstream uses a different
-            webToolsBackend: str("web_tools.backend", default: "duckduckgo"),
+            // path (e.g. `providers.openrouter.response_cache_enabled`
-            webToolsSearchBackend: str("web_tools.search.backend", default: "duckduckgo"),
+            // or nested under `prompt_caching`), update this single
-            webToolsExtractBackend: str("web_tools.extract.backend", default: "reader")
+            // line + the matching `setSetting` key in
+            // `SettingsViewModel.setOpenRouterResponseCache`. Default
+            // is `false` per WS-6-plan §Open Questions #2.
+            imageGenModel: str("image_gen.model", default: ""),
+            openrouterResponseCacheEnabled: bool("openrouter.response_cache.enabled", default: false)
         )
     }
 }

scarf/Packages/ScarfCore/Sources/ScarfCore/Services/ModelCatalogService.swift

@@ -155,9 +155,20 @@ public struct ModelCatalogService: Sendable {
             )
         }
         return byID.values.sorted { lhs, rhs in
+            // Subscription-gated first (Nous Portal).
             if lhs.subscriptionGated != rhs.subscriptionGated {
                 return lhs.subscriptionGated
             }
+            // Demoted last (Vercel AI Gateway, per Hermes v0.13). The
+            // axis is unconditional — we don't gate on the Hermes
+            // version because "Vercel mid-alphabet on v0.12, bottom on
+            // v0.13" would be more confusing than the consistent
+            // "Vercel last" treatment for everyone.
+            let lDemoted = Self.demotedProviders.contains(lhs.providerID)
+            let rDemoted = Self.demotedProviders.contains(rhs.providerID)
+            if lDemoted != rDemoted {
+                return !lDemoted
+            }
             return lhs.providerName.localizedCaseInsensitiveCompare(rhs.providerName) == .orderedAscending
         }
     }
@@ -235,7 +246,10 @@ public struct ModelCatalogService: Sendable {
     public func provider(for modelID: String) -> HermesProviderInfo? {
         guard let catalog = loadCatalog() else { return nil }
         for (providerID, p) in catalog {
-            if p.models?[modelID] != nil {
+            // Resolve any model-rename alias for this provider before
+            // checking the catalog — see `modelAliases` for rationale.
+            let resolved = resolveModelAlias(providerID: providerID, modelID: modelID)
+            if p.models?[resolved] != nil {
                 return HermesProviderInfo(
                     providerID: providerID,
                     providerName: p.name ?? providerID,
@@ -299,14 +313,17 @@ public struct ModelCatalogService: Sendable {
     /// Look up a specific model by provider + ID. Returns nil if not in the
     /// catalog (e.g., free-typed custom model).
     public func model(providerID: String, modelID: String) -> HermesModelInfo? {
+        // Resolve any model-rename alias for this provider before
+        // checking the catalog — see `modelAliases` for rationale.
+        let resolved = resolveModelAlias(providerID: providerID, modelID: modelID)
         guard let catalog = loadCatalog(),
               let provider = catalog[providerID],
-              let raw = provider.models?[modelID] else { return nil }
+              let raw = provider.models?[resolved] else { return nil }
         return HermesModelInfo(
             providerID: providerID,
             providerName: provider.name ?? providerID,
-            modelID: modelID,
+            modelID: resolved,
-            modelName: raw.name ?? modelID,
+            modelName: raw.name ?? resolved,
             contextWindow: raw.limit?.context,
             maxOutput: raw.limit?.output,
             costInput: raw.cost?.input,
@@ -344,10 +361,14 @@ public struct ModelCatalogService: Sendable {
     /// HTTP 404 at runtime. Catch that at save time, not 6 hours later.
     public func validateModel(_ modelID: String, for providerID: String) -> ModelValidation {
         ScarfMon.measure(.diskIO, "modelCatalog.validateModel") {
-            let trimmed = modelID.trimmingCharacters(in: .whitespacesAndNewlines)
+            let raw = modelID.trimmingCharacters(in: .whitespacesAndNewlines)
-            guard !trimmed.isEmpty else {
+            guard !raw.isEmpty else {
                 return .invalid(providerName: providerID, suggestions: [])
             }
+            // Resolve any model-rename alias before lookup so configs
+            // referencing a deprecated ID (e.g. `x-ai/grok-4.20-beta`)
+            // validate against the canonical successor.
+            let trimmed = resolveModelAlias(providerID: providerID, modelID: raw)
 
             // Overlay-only providers (Nous Portal, OpenAI Codex, Qwen
             // OAuth, …) serve their own catalogs that aren't mirrored to
@@ -433,6 +454,78 @@ public struct ModelCatalogService: Sendable {
         let output: Int?
     }
 
+    // MARK: - Model aliases (model rename resolution)
+
+    /// Hermes deprecates model IDs across releases. When a stored config
+    /// `model.default` references a deprecated ID, resolve to its
+    /// canonical successor. Lossless — we never rewrite the user's
+    /// `config.yaml`; the alias just lets `validateModel` /
+    /// `model(providerID:modelID:)` / `provider(for:)` succeed against
+    /// the new ID.
+    ///
+    /// Keys are slash-joined `providerID/modelID` to disambiguate
+    /// across providers — even if `vercel` later adds a `grok-4.20-beta`
+    /// alias on its own, the openrouter resolution shouldn't fire.
+    /// Values are the bare resolved model ID (no provider prefix).
+    ///
+    /// **Schema is Swift-primary.** Mirror new entries into Hermes's
+    /// upstream deprecation map in `hermes_cli/providers.py` if/when
+    /// upstream tracks renames in code (today they're release-notes
+    /// only).
+    public static let modelAliases: [String: String] = [
+        // v0.13: x-ai dropped the `-beta` suffix once Grok 4.20 GA'd.
+        // The model is the same one served at the same OpenRouter slot;
+        // only the marketing identifier changed.
+        // TODO(WS-6-Q4): verify whether OpenRouter retired the
+        // `x-ai/grok-4.20-beta` slot entirely. Either way the alias is
+        // correct (cosmetic if old slot stays live, load-bearing if it
+        // 404s).
+        "openrouter/x-ai/grok-4.20-beta": "x-ai/grok-4.20",
+        "xai/grok-4.20-beta": "grok-4.20",
+        "vercel/xai/grok-4.20-beta": "xai/grok-4.20",
+    ]
+
+    /// Resolve a stored model identifier through the alias map. Returns
+    /// the input unchanged when no alias exists. Pure function — used at
+    /// read time everywhere a config'd model ID is rendered, validated,
+    /// or sent to Hermes.
+    public func resolveModelAlias(providerID: String, modelID: String) -> String {
+        let composite = "\(providerID)/\(modelID)"
+        return Self.modelAliases[composite] ?? modelID
+    }
+
+    // MARK: - Demoted providers (sort tail)
+
+    /// Provider IDs that Hermes v0.13 explicitly deprioritizes in the
+    /// picker. `loadProviders()` sorts these to the tail of the list,
+    /// after the alphabetical group, so users who haven't manually
+    /// chosen Vercel as their gateway don't end up there by default.
+    /// Mirrors Hermes's deprioritized-provider list in
+    /// `hermes-agent/hermes_cli/providers.py`.
+    public static let demotedProviders: Set<String> = [
+        "vercel",
+    ]
+
+    // MARK: - Image-generation model allowlist (curated)
+
+    /// Known image-generation models, used to pre-populate the
+    /// `image_gen.model` picker on the Auxiliary tab. The list is
+    /// curated — `models_dev_cache.json` doesn't tag image-capable
+    /// models, so we maintain this by hand on Hermes version bumps.
+    /// Always free-form-typeable on the picker too, so missing entries
+    /// don't block users with non-listed image providers.
+    ///
+    /// Order: most-likely-to-be-chosen first.
+    public static let imageGenModels: [HermesImageGenModel] = [
+        .init(modelID: "openai/gpt-image-1", display: "OpenAI · gpt-image-1", providerHint: "openai"),
+        .init(modelID: "google/imagen-4", display: "Google · Imagen 4", providerHint: "google-vertex"),
+        .init(modelID: "google/imagen-3", display: "Google · Imagen 3", providerHint: "google-vertex"),
+        .init(modelID: "stability/stable-image-ultra", display: "Stability · Stable Image Ultra", providerHint: "stability"),
+        .init(modelID: "fal-ai/flux-pro-1.1", display: "fal · FLUX 1.1 Pro", providerHint: "fal"),
+        .init(modelID: "black-forest-labs/flux-1.1-pro", display: "Black Forest Labs · FLUX 1.1 Pro", providerHint: "openrouter"),
+        .init(modelID: "openai/dall-e-3", display: "OpenAI · DALL·E 3", providerHint: "openai"),
+    ]
+
     // MARK: - Hermes overlay providers
 
     /// The 11 providers Hermes surfaces via `hermes model` that have no
@@ -538,6 +631,27 @@ public struct ModelCatalogService: Sendable {
     ]
 }
 
+/// Curated entry for the `image_gen.model` picker on the Auxiliary
+/// tab. Hermes v0.13 honors a top-level `image_gen.model` key but the
+/// models.dev catalog has no `image: true` tag, so we maintain a
+/// short hand-curated allowlist keyed by display order. The picker
+/// always allows free-form-typing too, so any provider's model ID
+/// works regardless of whether it appears here.
+public struct HermesImageGenModel: Sendable, Identifiable, Hashable {
+    public let modelID: String
+    public let display: String
+    /// Hint at which provider serves this model — surfaced as a
+    /// "Configure provider X first" advisory but never enforced.
+    public let providerHint: String?
+    public var id: String { modelID }
+
+    public init(modelID: String, display: String, providerHint: String?) {
+        self.modelID = modelID
+        self.display = display
+        self.providerHint = providerHint
+    }
+}
+
 /// Scarf-side mirror of `HermesOverlay` from hermes-agent's
 /// `hermes_cli/providers.py`. Describes a provider that isn't in the
 /// models.dev catalog.

scarf/Packages/ScarfCore/Tests/ScarfCoreTests/M0cServicesTests.swift

@@ -310,6 +310,74 @@ import Foundation
         }
     }
 
+    // MARK: - ModelCatalogService — WS-6 (v0.13)
+
+    @Test func vercelAIGatewayDemotedToBottom() throws {
+        // Build a minimal catalog with vercel + alphabetically-later
+        // providers, then assert vercel sorts after them. Locks the
+        // demoted-axis sort comparator added in WS-6.
+        let json = """
+        {
+          "anthropic": { "name": "Anthropic", "models": {} },
+          "vercel":    { "name": "Vercel AI Gateway", "models": {} },
+          "zonk":      { "name": "Zonk Provider", "models": {} }
+        }
+        """
+        let tmp = FileManager.default.temporaryDirectory
+            .appendingPathComponent("scarf-models-\(UUID().uuidString).json")
+        try json.write(to: tmp, atomically: true, encoding: .utf8)
+        defer { try? FileManager.default.removeItem(at: tmp) }
+        let svc = ModelCatalogService(path: tmp.path)
+        let providers = svc.loadProviders().filter { !$0.isOverlay }
+        let names = providers.map(\.providerName)
+        // anthropic first (alpha), zonk next (alpha), vercel last
+        // (demoted) — even though `vercel` < `zonk` alphabetically.
+        #expect(names.last == "Vercel AI Gateway")
+        let vercelIdx = names.firstIndex(of: "Vercel AI Gateway") ?? -1
+        let zonkIdx = names.firstIndex(of: "Zonk Provider") ?? -1
+        #expect(vercelIdx > zonkIdx)
+    }
+
+    @Test func grok420BetaAliasResolvesToGrok420() {
+        let svc = ModelCatalogService(path: "/tmp/scarf-nonexistent-\(UUID().uuidString).json")
+        // OpenRouter's old `-beta` ID resolves to the GA name.
+        #expect(svc.resolveModelAlias(providerID: "openrouter", modelID: "x-ai/grok-4.20-beta")
+                == "x-ai/grok-4.20")
+        // xAI direct provider keeps the same shape minus prefix.
+        #expect(svc.resolveModelAlias(providerID: "xai", modelID: "grok-4.20-beta")
+                == "grok-4.20")
+        // Non-aliased ID passes through unchanged.
+        #expect(svc.resolveModelAlias(providerID: "anthropic", modelID: "claude-4.7-opus")
+                == "claude-4.7-opus")
+        // Cross-provider isolation: same modelID on a different
+        // provider isn't aliased — composite key in `modelAliases`
+        // disambiguates by providerID.
+        #expect(svc.resolveModelAlias(providerID: "fictional", modelID: "x-ai/grok-4.20-beta")
+                == "x-ai/grok-4.20-beta")
+    }
+
+    @Test func imageGenModelAllowlistShape() {
+        // Lock the curated list size + a few sentinel entries so
+        // unintentional edits get caught in review. Free-form-typing
+        // bypasses the allowlist, so additions/removals here are
+        // purely UX (which models surface as picker rows).
+        let models = ModelCatalogService.imageGenModels
+        #expect(models.count >= 5)
+        #expect(models.contains(where: { $0.modelID == "openai/gpt-image-1" }))
+        #expect(models.contains(where: { $0.modelID == "google/imagen-4" }))
+        // Every entry has a non-empty display + a non-empty modelID.
+        for m in models {
+            #expect(!m.modelID.isEmpty)
+            #expect(!m.display.isEmpty)
+        }
+    }
+
+    @Test func demotedProvidersContainsVercel() {
+        // Minimal lock-in for the demoted-providers static set. Mirrors
+        // Hermes's deprioritized-provider list in providers.py.
+        #expect(ModelCatalogService.demotedProviders.contains("vercel"))
+    }
+
     // MARK: - ProjectDashboardService
 
     @Test func projectDashboardServiceRegistryRoundTrip() throws {

scarf/Packages/ScarfCore/Tests/ScarfCoreTests/M6ConfigCronTests.swift

@@ -92,6 +92,27 @@ import Foundation
         #expect(c.security.redactSecrets == true)
         #expect(c.compression.enabled == true)
         #expect(c.voice.ttsProvider == "edge")
+        // v0.13 additions default to empty / off when the YAML omits
+        // them — pre-v0.13 hosts produce this exact shape.
+        #expect(c.imageGenModel == "")
+        #expect(c.openrouterResponseCacheEnabled == false)
+    }
+
+    @Test func parsesImageGenAndOpenRouterCache() {
+        // WS-6: round-trip the two new top-level v0.13 keys. If the
+        // OpenRouter key shape changes upstream (see TODO(WS-6-Q1)),
+        // this test is the single touchpoint that pins the parser
+        // line + setter key + UI binding to a single shape.
+        let yaml = """
+        image_gen:
+          model: openai/gpt-image-1
+        openrouter:
+          response_cache:
+            enabled: true
+        """
+        let c = HermesConfig(yaml: yaml)
+        #expect(c.imageGenModel == "openai/gpt-image-1")
+        #expect(c.openrouterResponseCacheEnabled == true)
     }
 
     @Test func parsesTopLevelModel() {

scarf/scarf/Core/Services/HermesFileService.swift

@@ -599,8 +599,7 @@ struct HermesFileService: Sendable {
                 toolsExclude: server.toolsExclude,
                 resourcesEnabled: server.resourcesEnabled,
                 promptsEnabled: server.promptsEnabled,
-                hasOAuthToken: hasToken,
+                hasOAuthToken: hasToken
-                sseReadTimeout: server.sseReadTimeout
             )
         }
     }
@@ -631,37 +630,6 @@ struct HermesFileService: Sendable {
         return runHermesCLI(args: cliArgs, timeout: 45, stdinInput: "y\ny\ny\n")
     }
 
-    /// Adds an SSE-transport MCP server. v0.13+ only — caller is responsible
-    /// for capability-gating; pre-v0.13 hosts will reject the `--transport`
-    /// flag at argparse time. The optional `sseReadTimeout` is passed via
-    /// `--sse-read-timeout <int>` and persisted as `sse_read_timeout: <int>`
-    /// in the YAML entry.
-    // TODO(WS-7-Q3): Verify exact CLI flag spelling against `hermes mcp add --help`
-    // on a v0.13 install. Plan assumes `--transport sse` + `--sse-read-timeout`;
-    // alternatives could be `--sse` (boolean) + `--read-timeout`.
-    @discardableResult
-    nonisolated func addMCPServerSSE(name: String, url: String, sseReadTimeout: Int?) -> (exitCode: Int32, output: String) {
-        var cliArgs: [String] = ["mcp", "add", name, "--url", url, "--transport", "sse"]
-        if let timeout = sseReadTimeout {
-            cliArgs.append(contentsOf: ["--sse-read-timeout", String(timeout)])
-        }
-        return runHermesCLI(args: cliArgs, timeout: 45, stdinInput: "y\ny\ny\n")
-    }
-
-    /// Updates the `sse_read_timeout` scalar in-place via the same surgical
-    /// patcher used by `setMCPServerTimeouts`. Pass `nil` to remove the
-    /// scalar entirely (Hermes default applies).
-    @discardableResult
-    nonisolated func setMCPServerSSETimeout(name: String, sseReadTimeout: Int?) -> Bool {
-        patchMCPServerField(name: name) { entryLines in
-            if let timeout = sseReadTimeout {
-                Self.replaceOrInsertScalar(key: "sse_read_timeout", value: String(timeout), in: &entryLines)
-            } else {
-                Self.removeScalar(key: "sse_read_timeout", in: &entryLines)
-            }
-        }
-    }
-
     @discardableResult
     nonisolated func setMCPServerArgs(name: String, args: [String]) -> Bool {
         patchMCPServerField(name: name) { entryLines in
@@ -844,23 +812,11 @@ struct HermesFileService: Sendable {
 
         func flush() {
             guard let name = currentName else { return }
-            // 3-way transport discriminator: an explicit `transport: sse` scalar
+            let transport: MCPTransport = fields["url"] != nil ? .http : .stdio
-            // wins (Hermes v0.13+ emits it for SSE servers); otherwise URL-bearing
-            // entries fall back to .http (v0.12 shape) and command-bearing entries
-            // to .stdio. This preserves byte-for-byte round-trip on existing files
-            // — pre-v0.13 entries have no `transport:` key so they parse identically.
-            // TODO(WS-7-Q1): Verify Hermes v0.13 actually emits `transport: sse`
-            // (vs. inferring from the schema/url shape) once a v0.13 host is on hand.
-            let transport: MCPTransport = {
-                if fields["transport"]?.lowercased() == "sse" { return .sse }
-                if fields["url"] != nil { return .http }
-                return .stdio
-            }()
             let enabledStr = fields["enabled"]?.lowercased()
             let enabled = enabledStr != "false"
             let timeout = fields["timeout"].flatMap(Int.init)
             let connectTimeout = fields["connect_timeout"].flatMap(Int.init)
-            let sseReadTimeout = fields["sse_read_timeout"].flatMap(Int.init)
             let server = HermesMCPServer(
                 name: name,
                 transport: transport,
@@ -877,8 +833,7 @@ struct HermesFileService: Sendable {
                 toolsExclude: excludeList,
                 resourcesEnabled: resources,
                 promptsEnabled: prompts,
-                hasOAuthToken: false,
+                hasOAuthToken: false
-                sseReadTimeout: sseReadTimeout
             )
             servers.append(server)
 

scarf/scarf/Features/Cron/ViewModels/CronViewModel.swift

@@ -146,7 +146,7 @@ final class CronViewModel {
         }
     }
 
-    func createJob(schedule: String, prompt: String, name: String, deliver: String, skills: [String], script: String, repeatCount: String, workdir: String = "", noAgent: Bool = false) {
+    func createJob(schedule: String, prompt: String, name: String, deliver: String, skills: [String], script: String, repeatCount: String, workdir: String = "") {
         var args = ["cron", "create"]
         if !name.isEmpty { args += ["--name", name] }
         if !deliver.isEmpty { args += ["--deliver", deliver] }
@@ -158,25 +158,12 @@ final class CronViewModel {
         // know the flag — argparse rejects unknown args, so the form
         // omits the flag when the field is empty.
         if !workdir.isEmpty { args += ["--workdir", workdir] }
-        // v0.13+: --no-agent runs the pre-run script and skips the AI turn.
-        // Caller (CronView) strips this on pre-v0.13 hosts so the flag is
-        // never emitted to a Hermes that can't parse it.
-        if noAgent { args.append("--no-agent") }
         args.append(schedule)
-        // TODO(WS-7-Q5): When --no-agent is set Hermes ignores the prompt arg,
+        if !prompt.isEmpty { args.append(prompt) }
-        // but argparse still wants positional args to line up with the
-        // schedule. The plan recommends passing an empty string explicitly so
-        // the positional parser doesn't treat the prompt as missing — verify
-        // this behaviour against `hermes cron create --help` on a v0.13 host.
-        if noAgent {
-            args.append("")
-        } else if !prompt.isEmpty {
-            args.append(prompt)
-        }
         runAndReload(args, success: "Job created")
     }
 
-    func updateJob(id: String, schedule: String?, prompt: String?, name: String?, deliver: String?, repeatCount: String?, newSkills: [String]?, clearSkills: Bool, script: String?, workdir: String? = nil, noAgent: Bool? = nil) {
+    func updateJob(id: String, schedule: String?, prompt: String?, name: String?, deliver: String?, repeatCount: String?, newSkills: [String]?, clearSkills: Bool, script: String?, workdir: String? = nil) {
         var args = ["cron", "edit", id]
         if let schedule, !schedule.isEmpty { args += ["--schedule", schedule] }
         if let prompt, !prompt.isEmpty { args += ["--prompt", prompt] }
@@ -193,16 +180,6 @@ final class CronViewModel {
         // = user cleared an existing workdir; Hermes documents `--workdir ""`
         // on edit as the explicit clear gesture, mirroring the `--script` shape.
         if let workdir { args += ["--workdir", workdir] }
-        // TODO(WS-7-Q4): The toggle-off shape of `--no-agent` on edit is
-        // unverified. Plan assumes Hermes accepts `--agent` to flip the flag
-        // back; if the CLI is one-way (`--no-agent` only), the edit-mode
-        // toggle should disable itself with a tooltip explaining the
-        // limitation. Send the flag in the assumed shape for now and adjust
-        // post-integration.
-        if let noAgent {
-            if noAgent { args.append("--no-agent") }
-            else { args.append("--agent") }
-        }
         runAndReload(args, success: "Updated")
     }
 

scarf/scarf/Features/Cron/Views/CronView.swift

@@ -25,10 +25,6 @@ struct CronView: View {
         capabilitiesStore?.capabilities.hasCronWorkdir ?? false
     }
 
-    private var hasCronNoAgent: Bool {
-        capabilitiesStore?.capabilities.hasCronNoAgent ?? false
-    }
-
     var body: some View {
         VStack(spacing: 0) {
             pageHeader
@@ -51,7 +47,7 @@ struct CronView: View {
         // polling timer. Same wiring ActivityView uses.
         .onChange(of: fileWatcher.lastChangeDate) { viewModel.load() }
         .sheet(isPresented: $viewModel.showCreateSheet) {
-            CronJobEditor(mode: .create, availableSkills: viewModel.availableSkills, supportsWorkdir: hasCronWorkdir, supportsNoAgent: hasCronNoAgent) { form in
+            CronJobEditor(mode: .create, availableSkills: viewModel.availableSkills, supportsWorkdir: hasCronWorkdir) { form in
                 viewModel.createJob(
                     schedule: form.schedule,
                     prompt: form.prompt,
@@ -60,12 +56,7 @@ struct CronView: View {
                     skills: form.skills,
                     script: form.script,
                     repeatCount: form.repeatCount,
-                    workdir: hasCronWorkdir ? form.workdir : "",
+                    workdir: hasCronWorkdir ? form.workdir : ""
-                    // Mirrors the workdir strip-on-pre-version pattern: pre-v0.13
-                    // hosts get a hard `false`, so a stale form value (or a
-                    // hand-edited jobs.json round-tripped through edit-mode)
-                    // can't sneak `--no-agent` into a CLI that doesn't grok it.
-                    noAgent: hasCronNoAgent ? form.noAgent : false
                 )
                 viewModel.showCreateSheet = false
             } onCancel: {
@@ -73,7 +64,7 @@ struct CronView: View {
             }
         }
         .sheet(item: $viewModel.editingJob) { job in
-            CronJobEditor(mode: .edit(job), availableSkills: viewModel.availableSkills, supportsWorkdir: hasCronWorkdir, supportsNoAgent: hasCronNoAgent) { form in
+            CronJobEditor(mode: .edit(job), availableSkills: viewModel.availableSkills, supportsWorkdir: hasCronWorkdir) { form in
                 viewModel.updateJob(
                     id: job.id,
                     schedule: form.schedule,
@@ -84,8 +75,7 @@ struct CronView: View {
                     newSkills: form.skills,
                     clearSkills: form.clearSkills,
                     script: form.script,
-                    workdir: hasCronWorkdir ? form.workdir : nil,
+                    workdir: hasCronWorkdir ? form.workdir : nil
-                    noAgent: hasCronNoAgent ? form.noAgent : nil
                 )
                 viewModel.editingJob = nil
             } onCancel: {
@@ -653,9 +643,6 @@ struct CronJobEditor: View {
         /// v0.12+ workdir flag — fills `--workdir <path>`. Empty string
         /// preserves the v0.11 behaviour of running with no cwd hint.
         var workdir: String = ""
-        /// v0.13+ `--no-agent` flag — script-only watchdog mode. Hermes
-        /// runs the pre-run script and skips the AI turn.
-        var noAgent: Bool = false
     }
 
     let mode: Mode
@@ -663,10 +650,6 @@ struct CronJobEditor: View {
     /// Pass `false` on pre-v0.12 hosts; the `--workdir` field is hidden and
     /// the form's value is dropped when the parent calls `createJob`/`updateJob`.
     let supportsWorkdir: Bool
-    /// Pass `false` on pre-v0.13 hosts; the `--no-agent` toggle is hidden
-    /// and the parent strips the form's value before calling
-    /// `createJob`/`updateJob`. Mirrors the `supportsWorkdir` pattern.
-    let supportsNoAgent: Bool
     let onSave: (FormState) -> Void
     let onCancel: () -> Void
 
@@ -698,25 +681,12 @@ struct CronJobEditor: View {
                     )
                     .scrollContentBackground(.hidden)
             }
-            .opacity(form.noAgent ? 0.4 : 1.0)
-            .disabled(form.noAgent)
             formField("Deliver", text: $form.deliver, placeholder: "origin | local | discord:CHANNEL | telegram:CHAT", mono: true)
             formField("Repeat", text: $form.repeatCount, placeholder: "Optional count")
             formField("Script path", text: $form.script, placeholder: "Python script whose stdout is injected", mono: true)
             if supportsWorkdir {
                 formField("Workdir", text: $form.workdir, placeholder: "Absolute path; pulls AGENTS.md/CLAUDE.md context", mono: true)
             }
-            if supportsNoAgent {
-                Toggle("Run script only (no agent call)", isOn: $form.noAgent)
-                    .scarfStyle(.body)
-                    .tint(ScarfColor.accent)
-                if form.noAgent {
-                    Text("Watchdog mode — Hermes runs the pre-run script and skips the AI turn. Prompt + skills are ignored.")
-                        .scarfStyle(.caption)
-                        .foregroundStyle(ScarfColor.foregroundMuted)
-                        .padding(.leading, ScarfSpace.s3)
-                }
-            }
             if !availableSkills.isEmpty {
                 VStack(alignment: .leading, spacing: 4) {
                     Text("Skills")
@@ -753,8 +723,6 @@ struct CronJobEditor: View {
                             .tint(ScarfColor.accent)
                     }
                 }
-                .opacity(form.noAgent ? 0.4 : 1.0)
-                .disabled(form.noAgent)
             }
             HStack {
                 Spacer()
@@ -778,7 +746,6 @@ struct CronJobEditor: View {
                 form.skills = job.skills ?? []
                 form.script = job.preRunScript ?? ""
                 form.workdir = job.workdir ?? ""
-                form.noAgent = job.noAgent ?? false
             }
         }
     }

scarf/scarf/Features/MCPServers/ViewModels/MCPServerEditorViewModel.swift

@@ -21,9 +21,6 @@ final class MCPServerEditorViewModel {
     var promptsEnabled: Bool
     var timeoutDraft: String
     var connectTimeoutDraft: String
-    /// SSE-only — renders as a third numeric on `.sse` servers. Empty string
-    /// means "use Hermes default" (writer drops the scalar).
-    var sseReadTimeoutDraft: String
     var showSecrets: Bool = false
     var isSaving: Bool = false
     var saveError: String?
@@ -40,7 +37,6 @@ final class MCPServerEditorViewModel {
         self.promptsEnabled = server.promptsEnabled
         self.timeoutDraft = server.timeout.map { String($0) } ?? ""
         self.connectTimeoutDraft = server.connectTimeout.map { String($0) } ?? ""
-        self.sseReadTimeoutDraft = server.sseReadTimeout.map { String($0) } ?? ""
     }
 
     func appendEnvRow() {
@@ -73,8 +69,6 @@ final class MCPServerEditorViewModel {
         let exclude = excludeDraft.split(separator: ",").map { $0.trimmingCharacters(in: .whitespaces) }.filter { !$0.isEmpty }
         let timeoutValue = Int(timeoutDraft.trimmingCharacters(in: .whitespaces))
         let connectValue = Int(connectTimeoutDraft.trimmingCharacters(in: .whitespaces))
-        let trimmedSSE = sseReadTimeoutDraft.trimmingCharacters(in: .whitespaces)
-        let sseTimeoutValue: Int? = trimmedSSE.isEmpty ? nil : Int(trimmedSSE)
 
         let service = fileService
         let transport = server.transport
@@ -93,11 +87,6 @@ final class MCPServerEditorViewModel {
                     if !service.setMCPServerEnv(name: name, env: envMap) { ok = false }
                 case .http:
                     if !service.setMCPServerHeaders(name: name, headers: headerMap) { ok = false }
-                case .sse:
-                    // SSE servers carry headers like .http does, plus an
-                    // optional sse_read_timeout written below.
-                    if !service.setMCPServerHeaders(name: name, headers: headerMap) { ok = false }
-                    if !service.setMCPServerSSETimeout(name: name, sseReadTimeout: sseTimeoutValue) { ok = false }
                 }
                 if !service.updateMCPToolFilters(
                     name: name,

scarf/scarf/Features/MCPServers/ViewModels/MCPServersViewModel.swift

@@ -42,10 +42,6 @@ final class MCPServersViewModel {
         filteredServers.filter { $0.transport == .http }
     }
 
-    var sseServers: [HermesMCPServer] {
-        filteredServers.filter { $0.transport == .sse }
-    }
-
     var selectedServer: HermesMCPServer? {
         guard let name = selectedServerName else { return nil }
         return servers.first(where: { $0.name == name })
@@ -171,11 +167,6 @@ final class MCPServersViewModel {
                     url: preset.url ?? "",
                     auth: preset.auth
                 )
-            case .sse:
-                // No SSE-transport presets ship today; the preset picker
-                // only surfaces stdio/http servers. Treat as a no-op
-                // failure if a preset somehow declares .sse.
-                addResult = (exitCode: 1, output: "SSE-transport presets are not supported.")
             }
             guard addResult.exitCode == 0 else {
                 await MainActor.run {
@@ -205,11 +196,6 @@ final class MCPServersViewModel {
                 result = fileService.addMCPServerStdio(name: name, command: command, args: args)
             case .http:
                 result = fileService.addMCPServerHTTP(name: name, url: url, auth: auth)
-            case .sse:
-                // Routed through addCustomSSE; this branch is unreachable from
-                // the add-server form (which dispatches per-transport in submit())
-                // but kept so the switch is exhaustive without `@unknown default`.
-                result = (exitCode: 1, output: "SSE servers must be added via addCustomSSE.")
             }
             await MainActor.run {
                 if result.exitCode == 0 {
@@ -225,28 +211,6 @@ final class MCPServersViewModel {
         }
     }
 
-    /// v0.13+ SSE-transport server creation. Caller is responsible for
-    /// capability-gating; the form filters `.sse` out of `availableTransports`
-    /// when `hasMCPSSETransport` is false, so this method is unreachable
-    /// from the UI on pre-v0.13 hosts.
-    func addCustomSSE(name: String, url: String, sseReadTimeout: Int?) {
-        let fileService = self.fileService
-        Task.detached {
-            let result = fileService.addMCPServerSSE(name: name, url: url, sseReadTimeout: sseReadTimeout)
-            await MainActor.run {
-                if result.exitCode == 0 {
-                    self.flashStatus("Added \(name)")
-                    self.load()
-                    self.selectedServerName = name
-                    self.showRestartBanner = true
-                    self.showAddCustom = false
-                } else {
-                    self.activeError = "Add failed: \(result.output)"
-                }
-            }
-        }
-    }
-
     func restartGateway() {
         let fileService = self.fileService
         Task.detached {

scarf/scarf/Features/MCPServers/Views/MCPServerAddCustomView.swift

@@ -6,26 +6,12 @@ struct MCPServerAddCustomView: View {
     let viewModel: MCPServersViewModel
 
     @Environment(\.dismiss) private var dismiss
-    @Environment(\.hermesCapabilities) private var capabilitiesStore
     @State private var name: String = ""
     @State private var transport: MCPTransport = .stdio
     @State private var command: String = "npx"
     @State private var argsText: String = ""
     @State private var url: String = ""
     @State private var auth: String = "none"
-    @State private var sseReadTimeout: String = ""
-
-    /// `.sse` is a v0.13+ surface; pre-v0.13 hosts only see stdio + http.
-    /// Iterating `MCPTransport.allCases` directly would render the SSE
-    /// segment unconditionally and Hermes would reject the resulting CLI
-    /// invocation at argparse time.
-    private var availableTransports: [MCPTransport] {
-        var t: [MCPTransport] = [.stdio, .http]
-        if capabilitiesStore?.capabilities.hasMCPSSETransport ?? false {
-            t.append(.sse)
-        }
-        return t
-    }
 
     var body: some View {
         VStack(spacing: 0) {
@@ -58,20 +44,17 @@ struct MCPServerAddCustomView: View {
                     }
                     sectionBox(title: "Transport") {
                         Picker("", selection: $transport) {
-                            ForEach(availableTransports) { t in
+                            ForEach(MCPTransport.allCases) { t in
                                 Text(t.displayName).tag(t)
                             }
                         }
                         .pickerStyle(.segmented)
                         .labelsHidden()
                     }
-                    switch transport {
+                    if transport == .stdio {
-                    case .stdio:
                         stdioSection
-                    case .http:
+                    } else {
                         httpSection
-                    case .sse:
-                        sseSection
                     }
                     Text("Env vars, headers, and tool filters can be edited after the server is added.")
                         .font(.caption)
@@ -129,28 +112,6 @@ struct MCPServerAddCustomView: View {
         }
     }
 
-    private var sseSection: some View {
-        sectionBox(title: "Endpoint (SSE)") {
-            VStack(alignment: .leading, spacing: 8) {
-                VStack(alignment: .leading, spacing: 4) {
-                    Text("URL").font(.caption.bold())
-                    TextField("https://.../sse", text: $url)
-                        .textFieldStyle(.roundedBorder)
-                        .font(.system(.body, design: .monospaced))
-                }
-                VStack(alignment: .leading, spacing: 4) {
-                    Text("SSE Read Timeout (seconds)").font(.caption.bold())
-                    TextField("default 300", text: $sseReadTimeout)
-                        .textFieldStyle(.roundedBorder)
-                        .frame(maxWidth: 140)
-                    Text("Hermes-side keepalive interval. Leave blank to use the default.")
-                        .font(.caption2)
-                        .foregroundStyle(.secondary)
-                }
-            }
-        }
-    }
-
     private var canSubmit: Bool {
         let trimmedName = name.trimmingCharacters(in: .whitespaces)
         guard !trimmedName.isEmpty else { return false }
@@ -159,8 +120,6 @@ struct MCPServerAddCustomView: View {
             return !command.trimmingCharacters(in: .whitespaces).isEmpty
         case .http:
             return !url.trimmingCharacters(in: .whitespaces).isEmpty
-        case .sse:
-            return !url.trimmingCharacters(in: .whitespaces).isEmpty
         }
     }
 
@@ -171,8 +130,6 @@ struct MCPServerAddCustomView: View {
             .map { $0.trimmingCharacters(in: .whitespaces) }
             .filter { !$0.isEmpty }
         let resolvedAuth: String? = (auth == "none") ? nil : auth
-        switch transport {
-        case .stdio, .http:
         viewModel.addCustom(
             name: trimmedName,
             transport: transport,
@@ -181,15 +138,6 @@ struct MCPServerAddCustomView: View {
             url: url.trimmingCharacters(in: .whitespaces),
             auth: resolvedAuth
         )
-        case .sse:
-            let trimmedTimeout = sseReadTimeout.trimmingCharacters(in: .whitespaces)
-            let parsedTimeout: Int? = trimmedTimeout.isEmpty ? nil : Int(trimmedTimeout)
-            viewModel.addCustomSSE(
-                name: trimmedName,
-                url: url.trimmingCharacters(in: .whitespaces),
-                sseReadTimeout: parsedTimeout
-            )
-        }
         dismiss()
     }
 

scarf/scarf/Features/MCPServers/Views/MCPServerDetailView.swift

@@ -127,11 +127,6 @@ struct MCPServerDetailView: View {
                 if let auth = server.auth, !auth.isEmpty {
                     summaryRow(label: "Auth", value: auth)
                 }
-            case .sse:
-                summaryRow(label: "URL", value: server.url ?? "—")
-                if let timeout = server.sseReadTimeout {
-                    summaryRow(label: "Read TO", value: "\(timeout)s")
-                }
             }
         }
         .padding(ScarfSpace.s3)

scarf/scarf/Features/MCPServers/Views/MCPServerEditorView.swift

@@ -186,16 +186,6 @@ struct MCPServerEditorView: View {
                         .textFieldStyle(.roundedBorder)
                         .frame(maxWidth: 140)
                 }
-                if viewModel.server.transport == .sse {
-                    VStack(alignment: .leading, spacing: 4) {
-                        Text("SSE read timeout")
-                            .font(.caption)
-                            .foregroundStyle(.secondary)
-                        TextField("default 300", text: $viewModel.sseReadTimeoutDraft)
-                            .textFieldStyle(.roundedBorder)
-                            .frame(maxWidth: 140)
-                    }
-                }
                 Spacer()
             }
         }

scarf/scarf/Features/MCPServers/Views/MCPServersView.swift

@@ -132,14 +132,6 @@ struct MCPServersView: View {
                     }
                 }
             }
-            if !viewModel.sseServers.isEmpty {
-                Section("Remote (SSE)") {
-                    ForEach(viewModel.sseServers) { server in
-                        serverRow(server)
-                            .tag(server.name as String?)
-                    }
-                }
-            }
             if viewModel.servers.isEmpty && !viewModel.isLoading {
                 Section {
                     Text("No servers configured yet")

scarf/scarf/Features/Profiles/ViewModels/ProfilesViewModel.swift

@@ -112,17 +112,10 @@ final class ProfilesViewModel {
         }
     }
 
-    func create(name: String, cloneConfig: Bool, cloneAll: Bool, noSkills: Bool = false) {
+    func create(name: String, cloneConfig: Bool, cloneAll: Bool) {
         var args = ["profile", "create", name]
         if cloneAll { args.append("--clone-all") }
         else if cloneConfig { args.append("--clone") }
-        // v0.13+: Empty-profile creation. The wire is independent of
-        // --clone / --clone-all per the v0.13 release notes — the user
-        // can stack `--clone --no-skills` to clone config but skip
-        // skills, which is a plausible workflow. The UI still disables
-        // the toggle under --clone-all (Decision H, see ProfilesView)
-        // but the wire is permissive.
-        if noSkills { args.append("--no-skills") }
         runAndReload(args, success: "Profile '\(name)' created")
     }
 

scarf/scarf/Features/Profiles/Views/ProfilesView.swift

@@ -11,12 +11,7 @@ struct ProfilesView: View {
     @State private var createName = ""
     @State private var createCloneConfig = true
     @State private var createCloneAll = false
-    /// v0.13+ `--no-skills` toggle. Mutually exclusive with `--clone-all`
-    /// at the UX layer (Decision H from the WS-7 plan): a full clone
-    /// copies skills wholesale — `--no-skills` would be a contradiction.
-    @State private var createNoSkills = false
     @State private var showRename = false
-    @Environment(\.hermesCapabilities) private var capabilitiesStore
 
     init(context: ServerContext) {
         _viewModel = State(initialValue: ProfilesViewModel(context: context))
@@ -128,7 +123,7 @@ struct ProfilesView: View {
                 }
                 Spacer()
                 Button {
-                    createName = ""; createCloneConfig = true; createCloneAll = false; createNoSkills = false
+                    createName = ""; createCloneConfig = true; createCloneAll = false
                     showCreate = true
                 } label: {
                     Label("Create", systemImage: "plus")
@@ -305,31 +300,11 @@ struct ProfilesView: View {
             Toggle("Clone config, .env, SOUL.md from active profile", isOn: $createCloneConfig)
                 .disabled(createCloneAll)
             Toggle("Full copy of active profile (all state)", isOn: $createCloneAll)
-            // TODO(WS-7-Q8): Decision H — disable --no-skills when --clone-all
-            // is on. A full clone copies skills wholesale; --no-skills would
-            // be a contradiction. Verify Hermes's behaviour with both flags
-            // (argparse mutual exclusion vs. last-flag-wins vs. clone-but-
-            // skip-skills) and relax the disabled state if Hermes does
-            // something useful with the combination.
-            if capabilitiesStore?.capabilities.hasProfileNoSkills ?? false {
-                Toggle("Empty profile (no skills)", isOn: $createNoSkills)
-                    .disabled(createCloneAll)
-            }
             HStack {
                 Spacer()
                 Button("Cancel") { showCreate = false }
                 Button("Create") {
-                    viewModel.create(
+                    viewModel.create(name: createName, cloneConfig: createCloneConfig, cloneAll: createCloneAll)
-                        name: createName,
-                        cloneConfig: createCloneConfig,
-                        cloneAll: createCloneAll,
-                        // Defensive: if the toggle isn't visible (pre-v0.13)
-                        // the state is always `false`, but read it through
-                        // the capability gate anyway so a stale state value
-                        // can't sneak `--no-skills` to a CLI that doesn't
-                        // know it.
-                        noSkills: (capabilitiesStore?.capabilities.hasProfileNoSkills ?? false) ? createNoSkills : false
-                    )
                     showCreate = false
                 }
                 .buttonStyle(.borderedProminent)

scarf/scarf/Features/Settings/ViewModels/SettingsViewModel.swift

@@ -143,16 +143,6 @@ final class SettingsViewModel {
     func setBrowserAllowPrivateURLs(_ value: Bool) { setSetting("browser.allow_private_urls", value: value ? "true" : "false") }
     func setCamofoxManagedPersistence(_ value: Bool) { setSetting("browser.camofox.managed_persistence", value: value ? "true" : "false") }
 
-    // MARK: - Web Tools
-
-    /// Pre-v0.13 combined backend. Pre-v0.13 hosts read this; v0.13+
-    /// hosts read it for back-compat but the WebToolsTab gates writes
-    /// on `hasWebToolsBackendSplit` so the tab only writes the split
-    /// keys on v0.13.
-    func setWebToolsBackend(_ value: String) { setSetting("web_tools.backend", value: value) }
-    func setWebToolsSearchBackend(_ value: String) { setSetting("web_tools.search.backend", value: value) }
-    func setWebToolsExtractBackend(_ value: String) { setSetting("web_tools.extract.backend", value: value) }
-
     // MARK: - Voice / TTS / STT
 
     func setAutoTTS(_ value: Bool) { setSetting("voice.auto_tts", value: value ? "true" : "false") }
@@ -205,6 +195,24 @@ final class SettingsViewModel {
         setSetting("auxiliary.\(task).timeout", value: String(value))
     }
 
+    // MARK: - Image generation (v0.13+)
+
+    /// `image_gen.model` — overrides the per-provider default image
+    /// model (Hermes v0.13+). Empty string clears the override.
+    /// Capability-gated in `AuxiliaryTab` so pre-v0.13 hosts never
+    /// invoke this setter.
+    func setImageGenModel(_ value: String) { setSetting("image_gen.model", value: value) }
+
+    /// `openrouter.response_cache.enabled` — toggles OpenRouter
+    /// response caching for repeat prompts (Hermes v0.13+).
+    /// Capability-gated in `AuxiliaryTab` so pre-v0.13 hosts never
+    /// invoke this setter.
+    // TODO(WS-6-Q1): the YAML key path is provisional — keep in lockstep
+    // with `HermesConfig+YAML.swift`'s parser line.
+    func setOpenRouterResponseCache(_ value: Bool) {
+        setSetting("openrouter.response_cache.enabled", value: value ? "true" : "false")
+    }
+
     // MARK: - Security / Privacy
 
     func setRedactSecrets(_ value: Bool) { setSetting("security.redact_secrets", value: value ? "true" : "false") }

scarf/scarf/Features/Settings/Views/SettingsView.swift

@@ -26,7 +26,6 @@ struct SettingsView: View {
         case agent = "Agent"
         case terminal = "Terminal"
         case browser = "Browser"
-        case webTools = "Web Tools"
         case voice = "Voice"
         case memory = "Memory"
         case auxiliary = "Aux Models"
@@ -42,7 +41,6 @@ struct SettingsView: View {
             case .agent: return "Agent"
             case .terminal: return "Terminal"
             case .browser: return "Browser"
-            case .webTools: return "Web Tools"
             case .voice: return "Voice"
             case .memory: return "Memory"
             case .auxiliary: return "Aux Models"
@@ -58,7 +56,6 @@ struct SettingsView: View {
             case .agent: return "brain.head.profile"
             case .terminal: return "terminal"
             case .browser: return "globe"
-            case .webTools: return "globe.americas"
             case .voice: return "mic"
             case .memory: return "memorychip"
             case .auxiliary: return "sparkles.rectangle.stack"
@@ -174,7 +171,6 @@ struct SettingsView: View {
         case .agent:     AgentTab(viewModel: viewModel)
         case .terminal:  TerminalTab(viewModel: viewModel)
         case .browser:   BrowserTab(viewModel: viewModel)
-        case .webTools:  WebToolsTab(viewModel: viewModel)
         case .voice:     VoiceTab(viewModel: viewModel)
         case .memory:    MemoryTab(viewModel: viewModel)
         case .auxiliary: AuxiliaryTab(viewModel: viewModel)

scarf/scarf/Features/Settings/Views/Tabs/AuxiliaryTab.swift

@@ -139,6 +139,23 @@ struct AuxiliaryTab: View {
                 auxRows(for: task.key)
             }
         }
+        // -- Hermes v0.13 additions ---------------------------------
+        // Image-gen model picker. Hermes v0.13 honors `image_gen.model`
+        // as a top-level YAML key; pre-v0.13 hosts ignore it silently.
+        // Hide the section on pre-v0.13 hosts to spare users a
+        // "I set this and nothing happened" trap.
+        if capabilitiesStore?.capabilities.hasImageGenModel ?? false {
+            SettingsSection(title: "Image Generation", icon: "photo") {
+                imageGenRow
+            }
+        }
+        // OpenRouter response caching toggle (v0.13+). Same hide-on-
+        // pre-v0.13 rationale: the toggle no-ops on older Hermes hosts.
+        if capabilitiesStore?.capabilities.hasOpenRouterResponseCache ?? false {
+            SettingsSection(title: "OpenRouter", icon: "shippingbox") {
+                openRouterResponseCacheRow
+            }
+        }
         // Unknown / unrecognised aux tasks present in config.yaml.
         // Shown only when at least one such key is present so the
         // typical user with a clean config never sees this section.
@@ -225,6 +242,60 @@ struct AuxiliaryTab: View {
         }
     }
 
+    // MARK: - v0.13 surfaces
+
+    /// Image-gen model picker — curated allowlist + free-form custom
+    /// entry. Capability-gated by the caller; this view assumes the
+    /// host honors `image_gen.model` (Hermes v0.13+).
+    @ViewBuilder
+    private var imageGenRow: some View {
+        let value = viewModel.config.imageGenModel
+        Picker("Model", selection: Binding(
+            get: { value },
+            set: { viewModel.setImageGenModel($0) }
+        )) {
+            Text("Provider default").tag("")
+            Divider()
+            ForEach(ModelCatalogService.imageGenModels) { model in
+                Text(model.display).tag(model.modelID)
+            }
+            // User has set a custom value not in the curated list;
+            // preserve it as a tagged option so the picker renders the
+            // actual selection rather than collapsing to "Provider
+            // default".
+            if !value.isEmpty
+                && !ModelCatalogService.imageGenModels.contains(where: { $0.modelID == value }) {
+                Divider()
+                Text(value + "  (custom)").tag(value)
+            }
+        }
+        .pickerStyle(.menu)
+        EditableTextField(label: "Custom model ID", value: value) { newValue in
+            viewModel.setImageGenModel(newValue.trimmingCharacters(in: .whitespaces))
+        }
+        Text("Used for image generation calls. Leave as Provider default unless your provider documents a specific model ID for image-gen.")
+            .font(.caption2)
+            .foregroundStyle(.tertiary)
+            .padding(.horizontal, 12)
+            .padding(.bottom, 4)
+    }
+
+    /// OpenRouter response-caching toggle (Hermes v0.13+). Off by
+    /// default; surfaced for users with highly repeated prompts who
+    /// want OpenRouter to cache identical-prompt responses.
+    @ViewBuilder
+    private var openRouterResponseCacheRow: some View {
+        let isOn = viewModel.config.openrouterResponseCacheEnabled
+        ToggleRow(label: "Response caching", isOn: isOn) { newValue in
+            viewModel.setOpenRouterResponseCache(newValue)
+        }
+        Text("OpenRouter caches identical prompts within a session to reduce token costs. Off by default — enable when your workload has highly repeated prompts.")
+            .font(.caption2)
+            .foregroundStyle(.tertiary)
+            .padding(.horizontal, 12)
+            .padding(.bottom, 4)
+    }
+
     private func auxModel(for key: String) -> AuxiliaryModel {
         switch key {
         case "vision": return viewModel.config.auxiliary.vision

scarf/scarf/Features/Settings/Views/Tabs/WebToolsTab.swift

@@ -1,76 +0,0 @@
-import SwiftUI
-import ScarfCore
-import ScarfDesign
-
-/// Web Tools tab — search + extract backend pickers. Pre-v0.13 hosts
-/// see a single "Combined backend" row writing to the legacy
-/// `web_tools.backend` key. v0.13+ hosts see two rows writing to the
-/// per-capability split keys (`web_tools.search.backend` +
-/// `web_tools.extract.backend`); SearXNG appears in the search picker
-/// only because Hermes registers it as a search-only backend.
-struct WebToolsTab: View {
-    @Bindable var viewModel: SettingsViewModel
-    @Environment(\.hermesCapabilities) private var capabilitiesStore
-
-    private var split: Bool {
-        capabilitiesStore?.capabilities.hasWebToolsBackendSplit ?? false
-    }
-
-    // TODO(WS-7-Q6): Backend lists are curated inline based on the v0.13
-    // release notes ("SearXNG joined search-only"). The exact dispatch
-    // table lives in `~/.hermes/hermes-agent/hermes_cli/web_tools.py` —
-    // verify during integration. A wrong entry just produces a
-    // `hermes config set` failure on save (recoverable, not silent).
-    private static let searchBackends: [String] = [
-        "duckduckgo", "tavily", "brave", "exa", "you", "searxng"
-    ]
-    private static let extractBackends: [String] = [
-        "reader", "browserless", "trafilatura", "firecrawl"
-    ]
-    /// v0.12 combined-backend list — superset of the v0.13 search list
-    /// minus SearXNG (which only dispatches as search) plus the v0.13
-    /// extract-only entries that pre-v0.13 hosts handled under the
-    /// combined key.
-    private static let combinedBackends: [String] = [
-        "duckduckgo", "tavily", "brave", "exa", "you",
-        "reader", "browserless", "trafilatura", "firecrawl"
-    ]
-
-    var body: some View {
-        if split {
-            SettingsSection(title: "Web Tools", icon: "globe.americas") {
-                PickerRow(
-                    label: "Search backend",
-                    selection: viewModel.config.webToolsSearchBackend,
-                    options: Self.searchBackends
-                ) { viewModel.setWebToolsSearchBackend($0) }
-                PickerRow(
-                    label: "Extract backend",
-                    selection: viewModel.config.webToolsExtractBackend,
-                    options: Self.extractBackends
-                ) { viewModel.setWebToolsExtractBackend($0) }
-            }
-            Text("SearXNG joined v0.13 as a search-only backend. Backend-specific tuning (host URLs, API keys) lives in the raw YAML editor for now.")
-                .scarfStyle(.caption)
-                .foregroundStyle(ScarfColor.foregroundMuted)
-                .padding(.horizontal, ScarfSpace.s4)
-        } else {
-            // TODO(WS-7-Q7): Pre-v0.13 hosts fall back to the legacy single
-            // backend. v0.13 may or may not honour `web_tools.backend` as a
-            // fallback when the split keys are absent — verify with Hermes
-            // and consider a one-time migration prompt in a follow-up if
-            // upgrading from v0.12 silently resets the user's backend.
-            SettingsSection(title: "Web Tools", icon: "globe.americas") {
-                PickerRow(
-                    label: "Backend",
-                    selection: viewModel.config.webToolsBackend,
-                    options: Self.combinedBackends
-                ) { viewModel.setWebToolsBackend($0) }
-            }
-            Text("Hermes v0.13 splits search and extract into separate backends. Update Hermes to access the per-capability picker.")
-                .scarfStyle(.caption)
-                .foregroundStyle(ScarfColor.foregroundFaint)
-                .padding(.horizontal, ScarfSpace.s4)
-        }
-    }
-}