
Visualize
- 296 installs
- 1 repo stars
- Updated May 26, 2026
- camacho/ai-skills
visualize is an AI agent skill that converts messy reasoning traces, datasets, and system designs into charts, diagrams, and other visual artifacts developers can review during implementation.
About
visualize is a Camacho AI skill for turning unstructured agent reasoning, tabular datasets, and architecture notes into charts, diagrams, and visual artifacts both humans and coding agents can inspect. It fits build-phase work when an implementation needs a shared picture of logic flow, data shape, or component relationships before or while code is written. Developers reach for visualize when text-only explanations are slowing review, when stakeholders need quick visuals from agent output, or when system design sketches must be generated from partial specs. The skill emphasizes reviewable visual deliverables rather than production rendering pipelines or BI dashboards.
- Generates architecture and flow diagrams from prompts
- Supports data charts and explanatory visuals
- Improves agent-to-human communication
- Useful for design and technical documentation
- Speeds alignment before coding
Visualize by the numbers
- 296 all-time installs (skills.sh)
- Ranked #2,276 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Jul 28, 2026 (Skillselion catalog sync)
npx skills add https://github.com/camacho/ai-skills --skill visualizeAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 296 |
|---|---|
| repo stars | ★ 1 |
| Last updated | May 26, 2026 |
| Repository | camacho/ai-skills ↗ |
How do you chart AI reasoning into reviewable diagrams?
Turn messy AI reasoning, datasets, or system designs into charts, diagrams, and visual artifacts agents and humans can review during implementation.
Who is it for?
Developers and agent workflows that need fast visual explanations of reasoning, data, or system design during implementation.
Skip if: Skip visualize when you need production-grade analytics dashboards, pixel-perfect UI mockups, or automated chart rendering inside a shipped application.
When should I use this skill?
Trigger when the user asks to diagram AI reasoning, visualize a dataset, sketch system architecture, or turn messy agent output into reviewable charts.
What you get
Mermaid or chart definitions, architecture diagrams, and visual summary artifacts ready for human or agent review.
Files
/visualize
Auto-detect content shape and render the appropriate visualization. Follows the active-selection rule from ai-workspace/rules/visualizations.md.
Usage
/visualize # visualize the last discussed concept
/visualize <topic or file> # visualize specific content
/visualize --format mermaid <topic> # force a specific formatSteps
1. Identify content. From args, recent conversation context, or a file path. Determine what needs visualizing.
2. Classify content shape. Use this decision table:
| Content shape | Signal | Format |
|---|---|---|
| Graph with nodes + edges | architecture, dependencies, flow between components | Mermaid graph or flowchart |
| Sequence / interaction | request flow, API calls, multi-step protocol | Mermaid sequenceDiagram |
| State transitions | modes, lifecycle, status changes | Mermaid stateDiagram-v2 |
| Comparison (3+ items) | options, tradeoffs, feature matrix | Markdown table |
| Hierarchy / tree | file structure, org chart, simple nesting | ASCII art |
| Contrast (2 items) | before/after, good/bad, do/don't | Inline pairs |
| Entity relationships | data model, schema | Mermaid erDiagram |
3. Check render surface. Mermaid in committed .md files renders on iOS/web/GitHub/VS Code. In terminal chunks, Mermaid is literal text — use ASCII instead. When surface is unknown, default to ASCII.
4. Render.
- ASCII / table / inline pairs: output directly in the response.
- Mermaid: output as a fenced
mermaidcode block in the response. If high-resolution PNG is needed:
node --import tsx "${SKILL_DIR}/scripts/render-mermaid.ts" --input <mermaid-file> --output <png-path> --width 2400 --height 1800- Format override: if user passed
--format, use that format regardless of auto-detection.
5. Present. Show the visualization. If Mermaid PNG was rendered, display the path.
Failure modes
| Condition | Behavior |
|---|---|
| Content shape ambiguous | Ask: "This could be a [X] or [Y] — which fits better?" |
| Mermaid syntax error | Show the error, offer to fix |
| mmdc not available | Fall back to fenced Mermaid block (renders on GitHub/web) |
Cross-tool notes
- Codex / Cursor: run
scripts/render-mermaid.tsdirectly for PNG rendering. Content-shape detection is agent reasoning, not script logic.
#!/usr/bin/env -S node --import tsx
// render-mermaid.ts — Render Mermaid diagram to PNG via @mermaid-js/mermaid-cli.
// Run from this skill: node --import tsx "${SKILL_DIR}/scripts/render-mermaid.ts" --input <file> --output <png> [--width N] [--height N]
// Or: "${SKILL_DIR}/scripts/render-mermaid.ts" --input <file> --output <png> (if chmod +x and tsx installed)
import { execFileSync } from "node:child_process";
import { readFileSync, unlinkSync, writeFileSync } from "node:fs";
import { resolve } from "node:path";
const args = process.argv.slice(2);
if (args.includes("--help") || args.length === 0) {
console.log(`Usage: node --import tsx "\${SKILL_DIR}/scripts/render-mermaid.ts" --input <file> --output <png> [options]
Renders a Mermaid diagram file to PNG using @mermaid-js/mermaid-cli (mmdc).
Options:
--input <file> Mermaid source file (or - for stdin)
--output <path> Output PNG path (required)
--width <px> Width in pixels (default: 2400)
--height <px> Height in pixels (default: 1800)
--help Show this help
Examples:
node --import tsx "\${SKILL_DIR}/scripts/render-mermaid.ts" --input diagram.mmd --output diagram.png
echo 'graph LR; A-->B' | node --import tsx "\${SKILL_DIR}/scripts/render-mermaid.ts" --input - --output out.png`);
process.exit(0);
}
function getArg(name: string, fallback?: string): string {
const idx = args.indexOf(name);
if (idx === -1 || idx + 1 >= args.length) {
if (fallback !== undefined) return fallback;
process.stderr.write(`Error: ${name} is required.\n`);
process.exit(1);
}
return args[idx + 1] as string;
}
const inputPath = getArg("--input");
const outputPath = getArg("--output");
const width = getArg("--width", "2400");
const height = getArg("--height", "1800");
let mermaidSource: string;
if (inputPath === "-") {
mermaidSource = readFileSync(0, "utf-8");
} else {
mermaidSource = readFileSync(inputPath, "utf-8");
}
// Write to temp file if reading from stdin
let inputFile = inputPath;
if (inputPath === "-") {
inputFile = resolve(outputPath + ".tmp.mmd");
writeFileSync(inputFile, mermaidSource, "utf-8");
}
try {
execFileSync("npx", [
"mmdc",
"-i", inputFile,
"-o", outputPath,
"-w", width,
"-H", height,
"--quiet",
], {
stdio: ["pipe", "pipe", "pipe"],
timeout: 30_000,
});
console.log(outputPath);
} catch (err) {
const error = err as { stderr?: Buffer };
process.stderr.write(`Error rendering Mermaid: ${error.stderr?.toString() ?? "unknown error"}\n`);
process.exit(1);
} finally {
if (inputPath === "-") {
try { unlinkSync(inputFile); } catch { /* ignore */ }
}
}
Related skills
How it compares
Choose visualize for quick review diagrams from agent or design input; use dedicated charting libraries or BI tools when shipping interactive analytics in production.
FAQ
What does the visualize skill produce?
The visualize skill from camacho/ai-skills generates charts, diagrams, and visual artifacts from AI reasoning traces, datasets, or system designs. Outputs are meant for developer and agent review during implementation, not as production dashboard code.
When should a developer use visualize?
Developers should use visualize when text explanations of agent reasoning, data shape, or architecture are insufficient for review. The skill fits build-phase documentation needs before or while implementation proceeds.