Configurable Dashboards for CCAI Insights: A Skill for Agents

Where this fits

A contact center analyst spends hours hand-tuning Vega-Lite specs and SQL queries to build a multi-tab dashboard for executive KPIs. Every change risks breaking the JSON, and deploying to a GCP project requires manual steps. The cxas-configurable-dashboards skill turns this into a declarative workflow: define dashboards in dashboards.yaml, let an agent validate and sync them, and iterate quickly without touching raw chart code.

Why agents benefit

• Author dashboards from natural language: An agent can generate a complete dashboards.yaml with multiple tabs, score cards, and charts based on a user's description of the KPIs they need.
• Validate schema and queries: The skill checks YAML structure, Vega-Lite syntax, and SQL references before deployment, catching errors early.
• Sync to GCP projects: Agents can push validated dashboards to the correct project and location, handling versioning and updates automatically.
• Iterate on visualizations: Users can request changes like "add a pie chart for sentiment" and the agent updates the YAML without touching the underlying query logic.
• Maintain declarative state: The skill keeps dashboards as code, enabling review, rollback, and reuse across environments.

Practical scenarios

Automating executive KPI dashboards

A manager asks for a weekly view of contact center metrics: handle time, CSAT, and call volume. The agent creates a multi-tab dashboard with score cards and trend lines, validates the SQL against sample data, and syncs it to the production project—all from a single prompt.

Iterating on visualization specs

An analyst wants to switch a bar chart to a Sankey diagram to show call flow. Instead of rewriting Vega-Lite manually, they ask the agent to update the chart type. The skill adjusts the spec, validates it, and previews the change.

Multi-project synchronization

A team manages dashboards across dev, staging, and prod environments. The agent reads the dashboards.yaml, compares versions, and syncs only the changed dashboards to each project, ensuring consistency without manual copy-paste.

Add it to your agent workflow

Install the skill with:

npx skills add GoogleCloudPlatform/cxas-scrapi --skill cxas-configurable-dashboards

Then use it in an agent prompt like:

Create a dashboard named "support_overview" with two tabs: one showing daily ticket volume as a line chart, another with a table of top issue categories. Use the project_id 'my-project' and location 'us-central1'. Validate and sync.

The agent will generate the YAML, run validation, and deploy it—no manual configuration.

Compare before adopting

Before committing, evaluate:

• Quality signals: Check the GitHub repo for recent commits, issue resolution, and test coverage. The skill is from GoogleCloudPlatform, which adds credibility.
• Maintenance freshness: Look at the last update date and whether it tracks CCAI Insights API changes.
• Alternatives: Compare with manual dashboard building or other automation tools. This skill is purpose-built for CCAI Insights, so it may be more efficient than generic scripting.
• Workflow fit: Does your team already use declarative configs? If not, the learning curve might be steeper.

Why it is worth tracking

With 95 stars and active maintenance under GoogleCloudPlatform, this skill shows steady community interest. It addresses a real pain point—declarative management of complex analytics dashboards—and aligns with the trend of infrastructure-as-code. If your agents handle contact center analytics or need to automate GCP dashboard deployments, this is worth evaluating now. For a deeper look, see the skill page on OpenAgentSkill.