Status
Live
Production-ready
Data & Analytics
Chart Validation Agent
Deterministic validation agent that ensures every chart has the correct type for its data shape and the correct configuration keys for its frontend component. Validates data fitness, normalizes config keys across 15+ cha
Enterprise AssistantLiveInternal (Colaberry Enterprise)Verified
Department
Data & Analytics
Enterprise Assistant department for Colaberry Enterprise agents
Source
Internal (Colaberry Enterprise)
Built by Colaberry
About
About the Agent
What this agent does, the challenges it addresses, and where it delivers value.
Deterministic validation agent that ensures every chart has the correct type for its data shape and the correct configuration keys for its frontend component. Validates data fitness, normalizes config keys across 15+ chart types, applies business-friendly titles, removes empty charts, and ensures a minimum chart count. No LLM calls - pure functions with sub-2ms latency.
Challenges This Agent Addresses
- 1**Business Intelligence**: Ensures dashboard charts always render correctly regardless of data shape variations
- 2**Education Analytics**: Validates student performance charts have appropriate visualization types for the data
- 3**Executive Reporting**: Guarantees professional, business-friendly chart titles instead of technical column names
Workflow
How the Agent Works
Step-by-step operational flow showing how this agent processes tasks end-to-end.
1
Step 1
**Data shape validation**: Checks each chart against shape rules (minimum/maximum rows, required numeric columns). Demotes charts to "bar" type if their data does not fit the selected type (e.g., radar requires 3 to 8 rows).
2
Step 2
**Config key normalization**: Maps each chart type to its frontend component's expected config keys. For example, a bar chart gets `x_axis`, `xKey`, `category`, `y_axes`, and `bars` keys. This is the core fix for empty-rendering charts.
3
Step 3
**Business title cleanup**: Replaces technical titles (SQL fragments, query references) with business-friendly labels using the data analyst dictionary.
4
Step 4
**Empty chart removal**: Filters out charts with no data or all-zero values.
5
Step 5
**Minimum chart guarantee**: If fewer than 2 charts remain, synthesizes additional bar charts from SQL results by finding string/numeric column pairs.
6
Step 6
Caps output at 4 charts maximum.
Execution Modes
Trigger: event (invoked during query response assembly)
Data
Inputs & Outputs
What data this agent consumes and the artifacts or actions it produces.
Input Data
- List of `ChartConfig` objects (type, title, data, labelKey, valueKey)
- List of `SqlResult` objects (for synthesizing additional charts if needed)
Deliverables
- Validated and normalized list of `ChartConfig` objects (max 4 charts), each with:
- Correct chart type for the data shape
- Normalized frontend config keys
- Business-friendly title
- Non-empty, non-zero data
Core Tasks
- Data & Analytics
Integrations
Systems Connected
Internal systems, APIs, and tools this agent integrates with.
Tools & APIs
Invoked by the **Query Engine** during response assemblyUses **Data Analyst Agent** (`getBusinessLabel`) for title generationReads `SqlResult` data for chart synthesis fallbackNormalized config is extracted via `extractNormalizedConfig` by the query engine
Specifications
Agent Specs
Technical specifications, requirements, and deployment details.
Related
Related Agents
Other agents in the same department or industry.
Enterprise AI
Ready to deploy this agent?
Schedule a walkthrough with our team to see how this agent integrates with your workflows.