Zustand Stores

Hand-written reference

This page is manually maintained. See frontend/src/stores/ for the definitive source.

All stores are barrel-exported from @/stores:

import { useUIStore, useDataStore, useMonitoringStore } from '@/stores';

Store List

Store	File	Description
useUIStore	ui-store.ts	UI preferences, filters, modal state
useDataStore	data-store.ts	Evaluation data from uploads
useMonitoringStore	monitoring-store.ts	Monitoring/observability data
useMemoryStore	memory-store.ts	Decision memory rules and batches
useHumanSignalsStore	human-signals-store.ts	Human signals data
useAnnotationStore	annotation-store.ts	Human annotation state
useCalibrationStore	calibration-store.ts	LLM judge calibration, pattern discovery, learning insights
useCopilotStore	copilot-store.ts	AI copilot chat state
useDatabaseStore	database-store.ts	DB connection wizard state
useEvalRunnerStore	eval-runner-store.ts	Evaluation runner workflow
useThemeStore	theme-store.ts	Theme/branding configuration

---

useUIStore

Manages global UI preferences. Uses persist middleware for localStorage persistence.

Key State:

Field	Type	Description
sidebarCollapsed	boolean	Sidebar open/closed
copilotOpen	boolean	AI copilot panel visibility
selectedExperiment	string \| null	Active experiment filter
selectedExperiments	string[]	Experiments for comparison
selectedMetrics	string[]	Active metric filters
currentPage	number	Current pagination page
itemsPerPage	number	Items per page
viewMode	'list' \| 'grid'	Data view mode
visualizeSubTab	string	Active visualize sub-tab

---

useDataStore

Manages evaluation data loaded from CSV uploads or DB imports.

Key State:

Field	Type	Description
data	EvaluationRecord[]	Raw evaluation records
format	DataFormat \| null	Detected data format
columns	string[]	Column names
metricColumns	string[]	Metric column names
summary	MetricSummary[]	Computed metric summaries
isLoading	boolean	Loading state
error	string \| null	Error message

Key Actions:

• setData(data, format) — Store uploaded data
• clearData() — Reset to empty state

---

useMonitoringStore

Manages production monitoring data with time-series filtering and source metadata.

Key State:

Field	Type	Description
data	MonitoringRecord[]	Raw monitoring records
metricColumns	string[]	Detected metric columns
timeRange	MonitoringTimeRange	Active time filter
selectedEnvironment	string \| null	Environment filter
selectedSourceName	string \| null	Source name filter
selectedSourceComponent	string \| null	Component filter
selectedSourceType	string \| null	Source type filter
availableEnvironments	string[]	Unique environments in data
availableSourceNames	string[]	Unique source names

---

useMemoryStore

Manages decision memory rules, hard stops, batches, and decision quality data.

Key State:

Field	Type	Description
rules	MemoryRuleRecord[]	All memory rules
activeTab	string	Active tab (rules, hardstops, batches, quality, graph)
searchQuery	string	Rule search filter
selectedCategory	string \| null	Category filter

---

useHumanSignalsStore

Manages Human Signals (HITL) data with source and metric filtering. Uses persist middleware.

Key State:

Field	Type	Description
cases	SignalsCaseRecord[]	Flattened case records
metricSchema	SignalsMetricSchema \| null	Auto-discovered metric schema
displayConfig	SignalsDisplayConfig \| null	YAML-driven display layout
selectedSourceName	string	Source name filter
selectedSourceComponent	string	Source component filter
selectedEnvironment	string	Environment filter
metricFilters	Record<string, string[]>	Per-metric signal filters (keyed by metric__signal)
timeRange	HumanSignalsTimeRange	Time range with presets (7d, 30d, 90d, 6m, 1y, custom)
datasetReady	boolean	Whether DuckDB dataset is available
selectedCaseId	string \| null	Case selected for detail modal
caseDetailModalOpen	boolean	Whether case detail modal is open
selectedSignalKpi	string \| null	KPI selected for expanded trend
sortColumn	string \| null	Active sort column in case table
visibleColumns	string[] \| null	Visible table columns (null = defaults)

---

useCalibrationStore

Manages LLM judge calibration state including evaluation results, pattern discovery, and learning insights from the EvidencePipeline.

Key State:

Field	Type	Description
step	number	Current workflow step (0-2)
evaluationResults	AlignEvaluationResult[] \| null	Judge evaluation results
clusterPatterns	ClusterPattern[] \| null	Discovered error patterns
learningArtifacts	LearningArtifact[] \| null	Actionable learning insights from EvidencePipeline
pipelineMetadata	PipelineMetadata \| null	Metadata about the clustering pipeline run
domainContext	string	Optional domain context for pattern discovery
selectedProvider	string	Active LLM provider
selectedModel	string	Active LLM model

---

Pattern

All stores follow the Zustand create<State>()((set) => ({...})) pattern:

import { create } from 'zustand';

interface ExampleState {
  count: number;
  increment: () => void;
  reset: () => void;
}

export const useExampleStore = create<ExampleState>()((set) => ({
  count: 0,
  increment: () => set((state) => ({ count: state.count + 1 })),
  reset: () => set({ count: 0 }),
}));

For persisted stores (e.g., useUIStore):

import { create } from 'zustand';
import { persist } from 'zustand/middleware';

export const useUIStore = create<UIState>()(
  persist(
    (set) => ({
      // ...state and actions
    }),
    {
      name: 'axis-ui-store',
      partialize: (state) => ({
        sidebarCollapsed: state.sidebarCollapsed,
        // only persist selected fields
      }),
    }
  )
);