Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Diagram

ASecurity

Diagrams of this repo's architecture and infrastructure, kept fresh in ./diagrams. Use when asked to diagram, map or document the system's architecture or infrastructure, or when ./diagrams is reported stale.

16 stars
0 votes
0 copies
0 views
Added 9/24/2026
ai-agentspythongobashgit

Security Analysis

A100/100

Scanned 9/24/2026

Install to Claude Code

$npx -y skills add heathdutton/claude-d2-diagrams --skill diagram --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Diagram?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Diagram
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/heathdutton-diagram/badge)](https://www.skillsdirectory.com/skills/heathdutton-diagram)

More formats (shields.io, HTML) on the badges page.

Download with Pro
Files
SKILL.md
---
name: diagram
description: Diagrams of this repo's architecture and infrastructure, kept fresh in ./diagrams. Use when asked to diagram, map or document the system's architecture or infrastructure, or when ./diagrams is reported stale.
argument-hint: "[--full] [--check] [--ci] [--infrastructure-only | --architecture-only] [--scope=<path>]"
allowed-tools:
  - Bash(${CLAUDE_PLUGIN_ROOT}/scripts/*)
  - Bash("${CLAUDE_PLUGIN_ROOT}/scripts/*)
  - Bash(d2 *)
  - Bash(git diff *)
  - Bash(git log *)
  - Read
  - Glob
  - Grep
  - Edit(diagrams/**)
  - Edit(README.md)
---

# Diagrams

`./diagrams` holds four docs, four D2 sources, eight SVGs (light and dark), a landing page, and `manifest.json`, which records the commit they were built from and the paths they depend on. The code is the ground truth. Existing docs and diagrams are hints that may have gone stale.

## Status

!`"${CLAUDE_PLUGIN_ROOT}/scripts/status.sh"`

Arguments: $ARGUMENTS

## Mode

- `--check`: report the status above in two lines and stop.
- `--ci`: follow [ci.md](ci.md) and stop.
- `state: fresh` without `--full`: say the diagrams are current as of the manifest commit and stop.
- `state: new`, `unknown-base` or `bad-sources`, or `--full`: **build** every artifact from scratch.
- `state: stale` or `legacy`: **refresh**, a surgical update driven by the changed files listed above.

`--infrastructure-only` and `--architecture-only` limit every step to one family. `--scope=<path>` goes into every documenter brief.

If the status has an `animations:` line, delete the file it names before rendering. If it has an `icons:` line, move every icon to the new host, whose paths are identical: `perl -pi -e 's#icons\.terrastruct\.com#icons.d2lang.com#g' diagrams/*.d2`.

The `d2:` line must end in `ok` before any step runs. Otherwise run the install or upgrade command it names, then re-run `"${CLAUDE_PLUGIN_ROOT}/scripts/status.sh"`. The `python3:` line must end in `ok` too, since icons are inlined with it: run the command it names.

## Steps

Dispatch the agents with the Agent tool in the foreground: several Agent calls in one message already run in parallel, and a headless run such as CI stops waiting on background agents after ten minutes. Every brief carries the mode and, for a refresh, the changed-file lines from the status above. The agents' results are short blocks: keep them for the report and the stamp.

1. **Document.** In one message, dispatch `d2:documenter` for `infrastructure` and for `architecture`. In a refresh, skip a family only when none of the changed files could bear on it.
   Done when each returns a status. A family that returns `unchanged` skips step 2, and step 3 too unless the state is `legacy`.
2. **Simplify.** In one message, dispatch `d2:documenter` with `model: sonnet` for the `-simplified` kind of each family whose detailed doc was written or updated, and name the sections that changed.
   Done when each returns a status.
3. **Render.** Dispatch one `d2:renderer` with every kind whose doc was written or updated. In a build, and in a `legacy` refresh, whose SVGs came from an older pipeline, that's every kind plus any other `.d2` the project keeps in `./diagrams`.
   Done when it returns `rendered` for each of those kinds.
4. **Verify.** Dispatch `d2:verifier` with the same kinds, the mode, the renderer's `png` directory, and the changed sections.
   Hand the `fix` findings to a fresh dispatch of whoever owns each file, a documenter for a `.md` and the renderer for a `.d2` (and the renderer again for any doc that changed). Then re-dispatch the verifier on those kinds, once. A `fix` that survives the second pass goes in the report.
5. **Embed.** In a build, or when `verify.sh` below reports a missing embed, follow [readme-embeds.md](readme-embeds.md).
6. **Gate.** Run `"${CLAUDE_PLUGIN_ROOT}/scripts/verify.sh"`, adding `--only <family>` for a single-family run, and fix what it names until it exits 0.
7. **Stamp.** Run `"${CLAUDE_PLUGIN_ROOT}/scripts/stamp.sh"` with the union of the documenters' `sources`. In a refresh, include the pathspecs already in `./diagrams/manifest.json`, so a family skipped this run keeps its sources.
   Skip the stamp on an `--infrastructure-only` or `--architecture-only` run: one commit covers both families, so stamping would mark the other family's drift as handled. Say so in the report.

## Report

Lead with the outcome: built, refreshed (which kinds changed), or already current. Then list only what needs the reader: surviving `fix` findings, `Unverified` counts, icons rendered without an image, and in a build, the one-line pointer that `/d2:diagram --ci` keeps diagrams fresh on every push.

Attribution

heathduttonheathdutton
View sourceMore from heathdutton →
SSkills DirectorySkills Directory

Know which skills are safe — weekly.

Best new skills + every skill we flagged as malicious. From the team that scanned 103,619.

Join free

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Know which skills are safe — weekly.

Best new skills + every skill we flagged as malicious. From the team that scanned 103,619.

Join free

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1074701 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

693161 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

691 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →