# Slide Excellence

Multi-agent comprehensive slide review (visual + pedagogy + proofreading, plus TikZ /...

- Repository: https://github.com/pedrohcgs/claude-code-my-workflow (1676 stars)
- Category: Documents
- License: MIT
- Last updated: 2026-09-27
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add pedrohcgs/claude-code-my-workflow --skill slide-excellence -g -a claude-code -y`
Codex: `npx skills add pedrohcgs/claude-code-my-workflow --skill slide-excellence -g -a codex -y`
Cursor: `npx skills add pedrohcgs/claude-code-my-workflow --skill slide-excellence -g -a cursor -y`
Gemini CLI: `npx skills add pedrohcgs/claude-code-my-workflow --skill slide-excellence -g -a gemini-cli -y`

## Description

Multi-agent comprehensive slide review (visual + pedagogy + proofreading, plus TikZ / parity / substance conditionally). Use when user says "full review", "excellence pass", "comprehensive check", "review everything", "pre-release review", "slide excellence", or before teaching / shipping a deck. Fanout wrapper, for a single lens, use `/visual-audit`, `/pedagogy-review`, or `/proofread` directly.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# Slide Excellence Review

Run a comprehensive multi-dimensional review of lecture slides. Multiple agents analyze the file independently, then results are synthesized.

> **Which slide-review skill do I want?**
>
> - **`/slide-excellence`** (this skill), multi-agent fanout (visual + pedagogy + proofread, plus TikZ / parity / substance conditionally). Best for **pre-teaching** or **pre-release** checks.
> - **`/visual-audit`**, single lens, layout/overflow/font/spacing only. Fast.
> - **`/pedagogy-review`**, single lens, narrative/prerequisites/worked-examples/notation.
> - **`/proofread`**, single lens, grammar/typos/overflow/terminology.
> - **`/qa-quarto`**, adversarial Beamer ↔ Quarto parity (critic-fixer loop).
> - **`/devils-advocate`**, 5-7 pointed challenges, not a full review.

**Important:** this orchestrator does **conditional** dispatch, it only spawns the subagents that can actually produce useful output for the given file. It does not run `tikz-reviewer` on a file with zero TikZ, or `quarto-critic` on a deck without a counterpart.

## Step 1: Identify the File

Parse `$ARGUMENTS` for the filename. Resolve path in `Quarto/` or `Slides/`.

Determine the file type:

- `.tex` → Beamer
- `.qmd` → Quarto
- `.md` → Markdown slides

## Step 2: Pre-flight, Detect Conditions

Before spawning any agent, probe the file to determine which reviews make sense:

```bash
FILE="$resolved_path"

# Has TikZ diagrams?
has_tikz=$(grep -c '\\begin{tikzpicture}' "$FILE" 2>/dev/null); has_tikz=${has_tikz:-0}

# For .qmd: is there a paired .tex in Slides/?
has_tex_pair="false"
if [[ "$FILE" == *.qmd ]]; then
  base="$(basename "$FILE" .qmd)"
  # Common Beamer suffixes: _Topic, _Lecture, exact match
  for candidate in "Slides/${base}.tex" "Slides/Lecture${base}.tex"; do
    if [ -f "$candidate" ]; then
      has_tex_pair="true"
      tex_pair="$candidate"
      break
    fi
  done
fi

The rest: https://github.com/pedrohcgs/claude-code-my-workflow/blob/HEAD/.claude/skills/slide-excellence/SKILL.md