@cat-factory/app 0.187.0 → 0.188.0

package/app/utils/catalog.ts

@@ -30,6 +30,7 @@ export const AGENT_CATEGORIES: { id: AgentCategory; label: string }[] = [
 export const AGENT_ARCHETYPES: AgentArchetype[] = [
   {
     kind: 'requirements-review',
+    tier: 'basic',
     label: 'Requirements Reviewer',
     icon: 'i-lucide-clipboard-check',
     color: '#f59e0b',
@@ -42,6 +43,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'clarity-review',
+    tier: 'basic',
     label: 'Clarity Reviewer',
     icon: 'i-lucide-bug',
     color: '#f59e0b',
@@ -58,6 +60,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
     // structured triage opens in the shared generic viewer; the clarity gate consumes its
     // `clarity`/`questions` server-side.
     kind: 'bug-investigator',
+    tier: 'intermediate',
     label: 'Bug Investigator',
     icon: 'i-lucide-search-code',
     color: '#38bdf8',
@@ -73,6 +76,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
     // resolves to the findings-selection window on every surface — not just when the manifest
     // is hydrated. Mirrors the backend `presentation` in `pr-reviewer.ts`.
     kind: 'pr-reviewer',
+    tier: 'basic',
     label: 'PR Reviewer',
     icon: 'i-lucide-clipboard-check',
     color: '#6366f1',
@@ -89,6 +93,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
     // shared generic viewer; the findings document is committed by a backend post-op (delivered
     // as a pull request by default, or straight to base via the direct pipeline).
     kind: 'spike',
+    tier: 'intermediate',
     label: 'Spike',
     icon: 'i-lucide-flask-conical',
     color: '#22d3ee',
@@ -99,6 +104,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'task-estimator',
+    tier: 'intermediate',
     label: 'Task Estimator',
     icon: 'i-lucide-gauge',
     color: '#eab308',
@@ -108,6 +114,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'requirements-brainstorm',
+    tier: 'intermediate',
     label: 'Requirements Brainstorm',
     icon: 'i-lucide-lightbulb',
     color: '#f59e0b',
@@ -120,6 +127,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'architecture-brainstorm',
+    tier: 'advanced',
     label: 'Architecture Brainstorm',
     icon: 'i-lucide-compass',
     color: '#a78bfa',
@@ -130,6 +138,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'architect',
+    tier: 'basic',
     label: 'Architect',
     icon: 'i-lucide-drafting-compass',
     color: '#a78bfa',
@@ -138,6 +147,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'researcher',
+    tier: 'intermediate',
     label: 'Researcher',
     icon: 'i-lucide-telescope',
     color: '#38bdf8',
@@ -146,6 +156,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'coder',
+    tier: 'basic',
     label: 'Coder',
     icon: 'i-lucide-code-xml',
     color: '#34d399',
@@ -154,6 +165,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'integrator',
+    tier: 'advanced',
     label: 'Integrator',
     icon: 'i-lucide-plug-zap',
     color: '#fb923c',
@@ -162,6 +174,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'mocker',
+    tier: 'advanced',
     label: 'Mock Builder',
     icon: 'i-lucide-server-cog',
     color: '#fb7185',
@@ -170,6 +183,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'tester-api',
+    tier: 'basic',
     label: 'API Tester',
     icon: 'i-lucide-flask-conical',
     color: '#fbbf24',
@@ -181,6 +195,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'tester-ui',
+    tier: 'intermediate',
     label: 'UI Tester',
     icon: 'i-lucide-camera',
     color: '#fbbf24',
@@ -192,6 +207,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'playwright',
+    tier: 'intermediate',
     label: 'Acceptance Test Author',
     icon: 'i-lucide-theater',
     color: '#e879f9',
@@ -201,6 +217,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'human-test',
+    tier: 'intermediate',
     label: 'Human Testing',
     icon: 'i-lucide-user-check',
     color: '#f59e0b',
@@ -213,6 +230,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'visual-confirmation',
+    tier: 'advanced',
     label: 'Visual Confirmation',
     icon: 'i-lucide-image-play',
     color: '#f59e0b',
@@ -225,6 +243,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'documenter',
+    tier: 'basic',
     label: 'Documenter',
     icon: 'i-lucide-book-open-text',
     color: '#818cf8',
@@ -233,6 +252,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'business-documenter',
+    tier: 'intermediate',
     label: 'Domain Rules Documenter',
     icon: 'i-lucide-scroll-text',
     color: '#84cc16',
@@ -242,6 +262,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'business-reviewer',
+    tier: 'advanced',
     label: 'Domain Rules Reviewer',
     icon: 'i-lucide-shield-alert',
     color: '#ef4444',
@@ -264,6 +285,7 @@ export const AGENT_ARCHETYPES: AgentArchetype[] = [
 export const COMPANION_ARCHETYPES: AgentArchetype[] = [
   {
     kind: 'reviewer',
+    tier: 'basic',
     label: 'Reviewer (companion)',
     icon: 'i-lucide-scan-eye',
     color: '#f472b6',
@@ -272,6 +294,7 @@ export const COMPANION_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'architect-companion',
+    tier: 'intermediate',
     label: 'Architect Companion',
     icon: 'i-lucide-bug-play',
     color: '#c084fc',
@@ -280,6 +303,7 @@ export const COMPANION_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'spec-companion',
+    tier: 'intermediate',
     label: 'Spec Reviewer',
     icon: 'i-lucide-list-checks',
     color: '#2dd4bf',
@@ -288,6 +312,7 @@ export const COMPANION_ARCHETYPES: AgentArchetype[] = [
   },
   {
     kind: 'doc-reviewer',
+    tier: 'intermediate',
     label: 'Doc Reviewer',
     icon: 'i-lucide-file-search',
     color: '#818cf8',
@@ -366,13 +391,19 @@ export function __resetCustomAgentKindMetaForTest(): void {
  * Agent kinds eligible for the optional consensus mechanism (the pipeline builder shows an
  * "Enable Consensus" toggle for these). Mirrors the backend default-eligible set assigned by
  * `registerConsensusTraits()` in `@cat-factory/consensus` — hand-synced, like the other
- * frontend mirrors. In CONSENSUS mode `architect`/`analysis` reason over the provided context
- * rather than exploring a checkout (a deliberate trade, gated by the task estimate).
+ * frontend mirrors. In CONSENSUS mode a container-backed kind (`architect`/`analysis`, and the
+ * checkout-exploring reviewers) reasons over the provided context rather than exploring a
+ * checkout (a deliberate trade, gated by the task estimate); the review kinds carry the bulk of
+ * the value, since a review is a judgement and a panel judges better than one model.
  */
 export const CONSENSUS_ELIGIBLE_KINDS: ReadonlySet<string> = new Set([
   'architect',
   'analysis',
   'reviewer',
+  'pr-reviewer',
+  'doc-reviewer',
+  'architect-companion',
+  'spec-companion',
   'task-estimator',
 ])
 
@@ -400,6 +431,7 @@ export function isTesterKind(kind: string): boolean {
 export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   'spec-writer': {
     kind: 'spec-writer',
+    tier: 'intermediate',
     label: 'Spec Writer',
     icon: 'i-lucide-clipboard-list',
     color: '#c084fc',
@@ -408,6 +440,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   },
   blueprints: {
     kind: 'blueprints',
+    tier: 'intermediate',
     label: 'Blueprinter',
     icon: 'i-lucide-map',
     color: '#22d3ee',
@@ -420,6 +453,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   // different (stronger) model than the reviewer. Its output renders in the pr-review window.
   'challenge-investigator': {
     kind: 'challenge-investigator',
+    tier: 'advanced',
     label: 'Challenge Investigator',
     icon: 'i-lucide-gavel',
     color: '#6366f1',
@@ -433,6 +467,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   // built-in pipelines, so it needs display metadata (else it renders as a generic gray "Agent").
   deployer: {
     kind: 'deployer',
+    tier: 'intermediate',
     label: 'Deployer',
     icon: 'i-lucide-cloud-upload',
     color: '#34d399',
@@ -445,6 +480,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   // interviewer, so the interview covers only what the code cannot answer.
   'initiative-interviewer': {
     kind: 'initiative-interviewer',
+    tier: 'advanced',
     label: 'Initiative Interviewer',
     icon: 'i-lucide-messages-square',
     color: '#818cf8',
@@ -455,6 +491,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   },
   'initiative-analyst': {
     kind: 'initiative-analyst',
+    tier: 'advanced',
     label: 'Initiative Analyst',
     icon: 'i-lucide-microscope',
     color: '#818cf8',
@@ -464,6 +501,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   },
   'initiative-planner': {
     kind: 'initiative-planner',
+    tier: 'advanced',
     label: 'Initiative Planner',
     icon: 'i-lucide-milestone',
     color: '#818cf8',
@@ -475,6 +513,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   },
   'initiative-committer': {
     kind: 'initiative-committer',
+    tier: 'advanced',
     label: 'Initiative Committer',
     icon: 'i-lucide-git-commit-horizontal',
     color: '#818cf8',
@@ -487,6 +526,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   // for run-timeline / saved-pipeline display rather than in AGENT_ARCHETYPES.
   analysis: {
     kind: 'analysis',
+    tier: 'intermediate',
     label: 'Analyst',
     icon: 'i-lucide-search-code',
     color: '#818cf8',
@@ -498,6 +538,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   // tech-debt pipeline, so it is a display-metadata system kind, not a palette archetype.
   tracker: {
     kind: 'tracker',
+    tier: 'advanced',
     label: 'Issue Tracker',
     icon: 'i-lucide-ticket',
     color: '#fb923c',
@@ -511,6 +552,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   // `tracker`), not a palette archetype.
   'bug-intake': {
     kind: 'bug-intake',
+    tier: 'advanced',
     label: 'Bug Intake',
     icon: 'i-lucide-inbox',
     color: '#fb923c',
@@ -519,6 +561,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   },
   conflicts: {
     kind: 'conflicts',
+    tier: 'intermediate',
     label: 'Conflicts Gate',
     icon: 'i-lucide-git-merge',
     color: '#f97316',
@@ -529,6 +572,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   },
   'conflict-resolver': {
     kind: 'conflict-resolver',
+    tier: 'intermediate',
     label: 'Conflict Resolver',
     icon: 'i-lucide-git-merge',
     color: '#f97316',
@@ -536,6 +580,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   },
   ci: {
     kind: 'ci',
+    tier: 'basic',
     label: 'CI Gate',
     icon: 'i-lucide-shield-check',
     color: '#38bdf8',
@@ -550,6 +595,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   // is a registered kind, so it arrives via the workspace snapshot's `customAgentKinds`.
   'doc-quality': {
     kind: 'doc-quality',
+    tier: 'intermediate',
     label: 'Doc Quality Gate',
     icon: 'i-lucide-file-check-2',
     color: '#818cf8',
@@ -560,6 +606,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   },
   'ci-fixer': {
     kind: 'ci-fixer',
+    tier: 'basic',
     label: 'CI Fixer',
     icon: 'i-lucide-wrench',
     color: '#38bdf8',
@@ -567,6 +614,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   },
   fixer: {
     kind: 'fixer',
+    tier: 'basic',
     label: 'Fixer',
     icon: 'i-lucide-wrench',
     color: '#fbbf24',
@@ -575,6 +623,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   },
   merger: {
     kind: 'merger',
+    tier: 'basic',
     label: 'Merger',
     icon: 'i-lucide-git-pull-request',
     color: '#a3e635',
@@ -585,6 +634,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   },
   'human-review': {
     kind: 'human-review',
+    tier: 'basic',
     label: 'Human Review Gate',
     icon: 'i-lucide-users',
     color: '#c084fc',
@@ -600,6 +650,7 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   // so it needs display metadata here and a per-workspace model in Model Configuration.
   kaizen: {
     kind: 'kaizen',
+    tier: 'advanced',
     label: 'Kaizen',
     icon: 'i-lucide-sparkles',
     color: '#2dd4bf',
@@ -613,6 +664,11 @@ export const SYSTEM_AGENT_META: Record<string, AgentArchetype> = {
   // but it still needs display metadata here so timelines/saved pipelines render it.
   'post-release-health': {
     kind: 'post-release-health',
+    // Intermediate rather than advanced: the palette already offers this gate ONLY when the
+    // workspace has an observability integration connected, and that connection is a stronger
+    // statement of intent than the tier — burying a gate the operator just wired two levels
+    // down would hide the capability they came for.
+    tier: 'intermediate',
     label: 'Post-Release Health',
     icon: 'i-lucide-activity',
     color: '#f43f5e',

package/i18n/locales/de.json

@@ -971,6 +971,47 @@
       "saved": "Standard-Testumgebung gespeichert.",
       "savedState": "Gespeichert.",
       "saveFailed": "Die Standard-Testumgebung konnte nicht gespeichert werden."
+    },
+    "consensusGroups": {
+      "title": "Konsensgruppen",
+      "intro": "Wiederverwendbare Review-Panels, auf die ein Pipeline-Schritt hochstufen kann. Jede Gruppe trägt die Schätzungsschwelle, die eine Aufgabe erreichen muss — so bekommt schwerere Arbeit ein größeres Panel, ohne eigene Pipeline.",
+      "new": "Neue Gruppe",
+      "list": {
+        "empty": "Noch keine Konsensgruppen.",
+        "always": "Immer",
+        "bar": "Ab {bar}",
+        "editTitle": "Gruppe bearbeiten",
+        "deleteTitle": "Gruppe löschen",
+        "participantCount": "keine Teilnehmer | ein Teilnehmer | {count} Teilnehmer"
+      },
+      "editor": {
+        "nameLabel": "Name",
+        "namePlaceholder": "Tiefes Review-Panel",
+        "strategyLabel": "Strategie",
+        "descriptionLabel": "Beschreibung",
+        "descriptionPlaceholder": "Wofür dieses Panel gedacht ist",
+        "participantsLabel": "Teilnehmer",
+        "participantsHint": "Mindestens zwei. Vielfalt bei Rolle und Modell macht ein Panel besser als ein einzelnes Modell.",
+        "addParticipant": "Teilnehmer hinzufügen",
+        "framingPlaceholder": "Perspektive, aus der dieser Teilnehmer argumentiert",
+        "stepModel": "Modell des Schritts",
+        "synthesizerLabel": "Synthese-Modell",
+        "gatedLabel": "Nur für Aufgaben über einer Schwelle",
+        "gatedHint": "Die Schätzung, die eine Aufgabe erreichen muss, um dieses Panel zu bekommen. Aus lassen, damit diese Gruppe immer gilt.",
+        "complexityThreshold": "Komplexität",
+        "gatingIncomplete": "Setze mindestens eine Schwelle oder schalte die Bedingung aus.",
+        "unnamedRole": "Reviewer",
+        "cancel": "Abbrechen",
+        "save": "Gruppe speichern"
+      },
+      "confirmDelete": {
+        "title": "Konsensgruppe löschen?",
+        "body": "Schritte, die „{name}“ nennen, fallen auf ihre verbleibenden Gruppen zurück."
+      },
+      "toast": {
+        "saveFailed": "Konsensgruppe konnte nicht gespeichert werden",
+        "deleteFailed": "Konsensgruppe konnte nicht gelöscht werden"
+      }
     }
   },
   "inspector": {
@@ -3618,7 +3659,11 @@
       "skillNeedsPick": "Ein Skill-Schritt braucht einen ausgewählten Skill, bevor du speichern kannst.",
       "purposeStepsConflict": "Dieser Zweck schreibt keinen Code und führt keine Tests aus, doch die Pipeline enthält noch Implementierungs- oder Testschritte. Entfernen Sie diese oder setzen Sie den Zweck auf Entwicklung.",
       "promptEditTooltip": "Systemprompt dieses Agenten bearbeiten",
-      "promptEditedTooltip": "Der Systemprompt dieses Agenten ist für diesen Arbeitsbereich bearbeitet"
+      "promptEditedTooltip": "Der Systemprompt dieses Agenten ist für diesen Arbeitsbereich bearbeitet",
+      "consensusGroups": "Konsensgruppen",
+      "consensusGroupsHint": "Wähle die wiederverwendbaren Panels, auf die dieser Schritt hochstufen darf. Jedes hat seine eigene Schätzungsschwelle; es läuft das anspruchsvollste, das die Aufgabe erreicht.",
+      "consensusGroupsActive": "Dieser Schritt führt die ausgewählte Gruppe aus. Hebe die Auswahl aller Gruppen auf, um die Teilnehmer hier zu konfigurieren.",
+      "consensusGroupAlways": "immer"
     },
     "progress": {
       "status": {
@@ -3727,6 +3772,17 @@
       "conflict": "Jemand anderes hat diesen Prompt geändert. Neu geladen - wenden Sie Ihre Änderung erneut an."
     }
   },
+  "agentTier": {
+    "label": "Agentenstufe",
+    "basic": "Basis",
+    "intermediate": "Fortgeschritten",
+    "advanced": "Experte",
+    "basicHint": "Nur die alltäglichen Agenten, aus denen eine Pipeline normalerweise besteht.",
+    "intermediateHint": "Ergänzt die Agenten, die du regelmäßig einsetzt.",
+    "advancedHint": "Zeigt alle Agenten, auch die spezialisierten.",
+    "tooltip": "Wähle, wie viel des Agentenkatalogs angezeigt wird",
+    "hidden": "{count} Agent ist auf dieser Stufe ausgeblendet. | {count} Agenten sind auf dieser Stufe ausgeblendet."
+  },
   "requirements": {
     "title": "Anforderungsprüfung",
     "iteration": "Iteration {current} / {max}",
@@ -4864,7 +4920,8 @@
       "synthesizing": "Wird synthetisiert",
       "done": "Fertig",
       "failed": "Fehlgeschlagen"
-    }
+    },
+    "groupTitle": "Die Konsensgruppe, auf die dieser Lauf hochgestuft hat"
   },
   "spec": {
     "title": "Anforderungen",

package/i18n/locales/en.json

@@ -3261,6 +3261,47 @@
       "saved": "Default test environment saved.",
       "savedState": "Saved.",
       "saveFailed": "Could not save the default test environment."
+    },
+    "consensusGroups": {
+      "title": "Consensus groups",
+      "intro": "Reusable review panels a pipeline step can escalate to. Each group carries the estimate bar a task must clear to earn it, so heavier work gets a bigger panel without a separate pipeline.",
+      "new": "New group",
+      "list": {
+        "empty": "No consensus groups yet.",
+        "always": "Always",
+        "bar": "From {bar}",
+        "editTitle": "Edit group",
+        "deleteTitle": "Delete group",
+        "participantCount": "no participants | one participant | {count} participants"
+      },
+      "editor": {
+        "nameLabel": "Name",
+        "namePlaceholder": "Deep review panel",
+        "strategyLabel": "Strategy",
+        "descriptionLabel": "Description",
+        "descriptionPlaceholder": "What this panel is for",
+        "participantsLabel": "Participants",
+        "participantsHint": "At least two. Diversity of role and model is what makes a panel better than one model.",
+        "addParticipant": "Add participant",
+        "framingPlaceholder": "Perspective this participant argues from",
+        "stepModel": "Step's model",
+        "synthesizerLabel": "Synthesizer model",
+        "gatedLabel": "Only for tasks above a threshold",
+        "gatedHint": "The estimate a task must reach to earn this panel. Leave off to make this the group that always applies.",
+        "complexityThreshold": "Complexity",
+        "gatingIncomplete": "Set at least one threshold, or turn the gate off.",
+        "unnamedRole": "Reviewer",
+        "cancel": "Cancel",
+        "save": "Save group"
+      },
+      "confirmDelete": {
+        "title": "Delete consensus group?",
+        "body": "Steps that name \"{name}\" fall back to their remaining groups."
+      },
+      "toast": {
+        "saveFailed": "Could not save the consensus group",
+        "deleteFailed": "Could not delete the consensus group"
+      }
     }
   },
   "providers": {
@@ -4049,7 +4090,11 @@
       "skillNeedsPick": "A Skill step needs a skill selected before you can save.",
       "purposeStepsConflict": "This purpose writes no code and runs no tests, but the pipeline still has implementation or testing steps. Remove them or switch the purpose to Build.",
       "promptEditTooltip": "Edit this agent's system prompt",
-      "promptEditedTooltip": "This agent's system prompt is edited for this workspace"
+      "promptEditedTooltip": "This agent's system prompt is edited for this workspace",
+      "consensusGroups": "Consensus groups",
+      "consensusGroupsHint": "Pick the reusable panels this step may escalate to. Each carries its own estimate bar; the most demanding one the task clears is the one that runs.",
+      "consensusGroupsActive": "This step runs the selected group. Deselect every group to configure participants here instead.",
+      "consensusGroupAlways": "always"
     },
     "progress": {
       "status": {
@@ -4167,6 +4212,23 @@
       "conflict": "Someone else changed this prompt. Reloaded - re-apply your edit."
     }
   },
+  "agentTier": {
+    "label": "Agent tier",
+    "@label": {
+      "description": "Label of the control that chooses how much of the agent catalog is shown. \"Tier\" here means level of depth/specialisation, never a pricing plan."
+    },
+    "basic": "Basic",
+    "intermediate": "Intermediate",
+    "advanced": "Advanced",
+    "@advanced": {
+      "description": "The widest of the three levels: selecting it shows every agent, not only the advanced ones."
+    },
+    "basicHint": "Only the everyday agents a pipeline is normally built from.",
+    "intermediateHint": "Adds the agents you reach for regularly.",
+    "advancedHint": "Shows every agent, including the specialist ones.",
+    "tooltip": "Choose how much of the agent catalog to show",
+    "hidden": "{count} agent hidden at this tier. | {count} agents hidden at this tier."
+  },
   "riskPolicy": {
     "health": {
       "title": "Risk policy updates",
@@ -4582,7 +4644,8 @@
       "synthesizing": "Synthesizing",
       "done": "Done",
       "failed": "Failed"
-    }
+    },
+    "groupTitle": "The consensus group this run escalated to"
   },
   "brainstorm": {
     "title": {

package/i18n/locales/es.json

@@ -3167,6 +3167,47 @@
       "saved": "Entorno de pruebas predeterminado guardado.",
       "savedState": "Guardado.",
       "saveFailed": "No se ha podido guardar el entorno de pruebas predeterminado."
+    },
+    "consensusGroups": {
+      "title": "Grupos de consenso",
+      "intro": "Paneles de revisión reutilizables a los que puede escalar un paso de la canalización. Cada grupo lleva el umbral de estimación que una tarea debe superar, así el trabajo más pesado recibe un panel mayor sin necesidad de otra canalización.",
+      "new": "Nuevo grupo",
+      "list": {
+        "empty": "Todavía no hay grupos de consenso.",
+        "always": "Siempre",
+        "bar": "Desde {bar}",
+        "editTitle": "Editar grupo",
+        "deleteTitle": "Eliminar grupo",
+        "participantCount": "sin participantes | un participante | {count} participantes"
+      },
+      "editor": {
+        "nameLabel": "Nombre",
+        "namePlaceholder": "Panel de revisión profunda",
+        "strategyLabel": "Estrategia",
+        "descriptionLabel": "Descripción",
+        "descriptionPlaceholder": "Para qué sirve este panel",
+        "participantsLabel": "Participantes",
+        "participantsHint": "Al menos dos. La diversidad de rol y de modelo es lo que hace que un panel supere a un solo modelo.",
+        "addParticipant": "Añadir participante",
+        "framingPlaceholder": "Perspectiva desde la que argumenta este participante",
+        "stepModel": "Modelo del paso",
+        "synthesizerLabel": "Modelo sintetizador",
+        "gatedLabel": "Solo para tareas por encima de un umbral",
+        "gatedHint": "La estimación que una tarea debe alcanzar para merecer este panel. Déjalo desactivado para que este grupo se aplique siempre.",
+        "complexityThreshold": "Complejidad",
+        "gatingIncomplete": "Fija al menos un umbral o desactiva la condición.",
+        "unnamedRole": "Revisor",
+        "cancel": "Cancelar",
+        "save": "Guardar grupo"
+      },
+      "confirmDelete": {
+        "title": "¿Eliminar el grupo de consenso?",
+        "body": "Los pasos que nombran «{name}» recurrirán a sus grupos restantes."
+      },
+      "toast": {
+        "saveFailed": "No se pudo guardar el grupo de consenso",
+        "deleteFailed": "No se pudo eliminar el grupo de consenso"
+      }
     }
   },
   "providers": {
@@ -3934,7 +3975,11 @@
       "skillNeedsPick": "Un paso Skill necesita una habilidad seleccionada antes de poder guardar.",
       "purposeStepsConflict": "Este propósito no escribe código ni ejecuta pruebas, pero la canalización todavía tiene pasos de implementación o de prueba. Quítalos o cambia el propósito a Desarrollo.",
       "promptEditTooltip": "Editar el prompt del sistema de este agente",
-      "promptEditedTooltip": "El prompt del sistema de este agente está editado para este espacio de trabajo"
+      "promptEditedTooltip": "El prompt del sistema de este agente está editado para este espacio de trabajo",
+      "consensusGroups": "Grupos de consenso",
+      "consensusGroupsHint": "Elige los paneles reutilizables a los que este paso puede escalar. Cada uno lleva su propio umbral de estimación; se ejecuta el más exigente que la tarea supere.",
+      "consensusGroupsActive": "Este paso ejecuta el grupo seleccionado. Deselecciona todos los grupos para configurar aquí los participantes.",
+      "consensusGroupAlways": "siempre"
     },
     "progress": {
       "status": {
@@ -4043,6 +4088,17 @@
       "conflict": "Otra persona cambió este prompt. Se recargó: vuelve a aplicar tu edición."
     }
   },
+  "agentTier": {
+    "label": "Nivel de agentes",
+    "basic": "Básico",
+    "intermediate": "Intermedio",
+    "advanced": "Avanzado",
+    "basicHint": "Solo los agentes de uso diario con los que se arma una pipeline.",
+    "intermediateHint": "Añade los agentes que usas con regularidad.",
+    "advancedHint": "Muestra todos los agentes, incluidos los especializados.",
+    "tooltip": "Elige cuánto del catálogo de agentes se muestra",
+    "hidden": "{count} agente oculto en este nivel. | {count} agentes ocultos en este nivel."
+  },
   "palette": {
     "hint": "Haz clic en un agente para añadirlo al pipeline.",
     "customAgents": "Agentes personalizados"
@@ -4386,7 +4442,8 @@
       "synthesizing": "Sintetizando",
       "done": "Completado",
       "failed": "Fallido"
-    }
+    },
+    "groupTitle": "El grupo de consenso al que escaló esta ejecución"
   },
   "brainstorm": {
     "title": {