Visual Explainer

Extension

Adds a sequential prepare-and-render tool that guides visual explanation planning, validates complete HTML documents, and writes them inside a fixed diagrams directory.

What it does

Visual Explainer registers a sequential prepare-and-render tool that guides visual explanation planning and writes supplied complete HTML documents.

Its workflow separates preparation from rendering and requires a complete HTML document before writing output.

When to reach for it

Invocation: Ask the model to prepare a visual explanation, then render the completed document.

Architecture or workflow relationships are hard to scan in prose
Create a focused visual map with supporting explanation.
A plan or diff needs a reviewable visual recap
Turn the relevant evidence into a complete HTML artifact.
Data or technical concepts need an audience-specific explanation
Prepare for the audience before authoring the final document.

What it adds

  • A sequential prepare-and-render tool contract.
  • Validated output naming, directory containment, and optional browser opening.

How it fits

  • Pi Subagents: Preparation can recommend a subagent scout when more context should be gathered first.
  • Tufte visualization: The visualization skill can critique quantitative displays inside a visual explanation.

Common questions

What remains the model's responsibility?
  • The model authors the HTML, output goes to a fixed local directory, and browser opening is best-effort.

It’s working if

  • Preparation returns a bounded explanation plan before rendering.
  • Rendering writes one complete HTML document inside the diagrams directory.

Source and setup

Provenance

Origin
Nico Bailon
Current maintainer
Vedang Manerikar
Relationship
fork
Installed source
git:github.com/vedang/visual-explainer
Source path
View source (plugins/visual-explainer/extension.ts)
Compatibility
  • Pi package APIs
  • Optional browser opener
  • Optional subagent-aware preparation
License
MIT
License publication scope
Source link, independently written summary, original-author/fork attribution, and reviewed behavior claims.
Reviewed revision
0d8d26471ca0d8d50ffe1eb7009c62f56c092c5c
Catalog revision
e0c035a297462c519315abfa1af31a1ed3bd9eef

Setup notes

  • Created by Nico Bailon; Vedang maintains the installed MIT-licensed fork and its visual-learning additions.
  • The package supports an optional browser opener and optional subagent-aware preparation.
  • Behavior claims were reviewed against the installed fork revision, not current upstream.