tools/vizli

Discover, validate, render, and verify visualization templates across multiple engines and formats.

2|Updated Feb 22, 2026
One-click install
npx skills add https://github.com/weisberg/agent_tools --skill tools-vizli
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tools/vizli
Source: https://github.com/weisberg/agent_tools/tree/main/tools/vizli
Command: npx skills add https://github.com/weisberg/agent_tools --skill tools-vizli

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pydantic, jinja2, pyyaml, vl-convert-python, playwright, cairosvg, weasyprint, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill enables AI agents to reliably discover, validate, and run complex visualization templates, ensuring consistent outputs across data sources and formats.

Core Features & Use Cases

  • Template Discovery & Selection: Finds visualization templates via sidecars and metadata, matching user requests with predefined use cases.
  • Payload Resolution & Validation: Resolves parameters and datasets, checking schemas and supporting safe, error-resilient operation.
  • Rendering & Verification: Executes rendering engines (Vega-Lite, Altair, SVG, HTML, JS) for static or interactive visuals, performing post-render checks.
  • Use Case: Automate production report generation by selecting templates matching user data, applying default styles, and verifying outputs before delivery.
  • Supports multiple formats like SVG, PNG, HTML, PDF; with options for themes, resolution, and debugging.

Quick Start

Use the tools/vizli Skill to run a template with data, parameters, and output preferences, ensuring validation and output verification before delivery.

Frequently Asked Questions about tools/vizli

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate visualization rendering and output verification for AI workflows?

Automate visualization rendering by discovering templates, validating data payloads against schemas, executing engines like Vega-Lite or SVG, and performing post-render verification to ensure deterministic outputs.

How does template matching work for automated data visualization?

Template matching works by discovering visualization templates via sidecars and metadata, then matching user requests and data files with predefined use cases to select the appropriate rendering approach.

Can I validate data schemas before rendering Vega-Lite or Altair visuals?

Yes, you can validate data schemas before rendering. The process resolves parameters and datasets, checking schemas with pydantic to support safe, error-resilient operation before executing engines.

What output formats are supported for deterministic visualization rendering?

Supported output formats include SVG, PNG, HTML, and PDF. The rendering process executes various engine types and embeds metadata while offering options for themes, resolution, and debugging.

Do I need specific dependencies to run visualization templates with Jinja2 and WeasyPrint?

Yes, rendering requires dependencies like jinja2 for templates, weasyprint for PDF generation, vl-convert-python for Vega-Lite, and playwright for interactive HTML or JS visual output verification.

Why does visualization template rendering fail during output verification?

Rendering fails during output verification when data payloads do not match template schemas, environment constraints are unmet, or post-render checks detect mismatches in the generated SVG, PNG, or HTML outputs.