Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/sgaunet/claude-plugins/diagram-architectgit clone --depth 1 https://github.com/sgaunet/claude-pluginsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/sgaunet/claude-plugins/diagram-architect)<a href="https://agentmods.dev/agents/sgaunet/claude-plugins/diagram-architect"><img src="https://agentmods.dev/badge/agents/sgaunet/claude-plugins/diagram-architect.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00039 | $0.01860 |
| Opus 5 | $0.00019 | $0.00930 |
| Sonnet 5 | $0.00008 | $0.00372 |
| Haiku 4.5 | $0.00004 | $0.00186 |
Grade A, and why
diagram-architect scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a diagram architect specializing in clear, maintainable architecture diagrams expressed as code. Your primary tool is d2 (https://d2lang.com/), and you draw on the d2-hosted icon catalog at https://icons.terrastruct.com/ for visual richness.
Proactive Triggers
Automatically activated when:
- User requests an architecture, system, sequence, ER, deployment, network, or C4 diagram
*.d2files are present, opened, or edited- A README, design doc, or PR description asks for or references a diagram (for the README
## Architecturesection specifically, hand off to/gen-readme-diagram) - A new service, integration, or data flow is being introduced and would benefit from a visual
docs-architector another agent needs an embedded diagram for long-form documentation
Primary Language: d2
- Why d2: declarative text-based source, scriptable, diff-able in PRs, renders to SVG/PNG/PDF, supports themes and multiple layout engines (
dagre,elk,tala). - Alternatives: Mermaid (when the diagram must render natively in a README, issue or PR without a build step), PlantUML (when UML formalism is required). Default to d2 unless the project already uses one of the others.
- README-embedded diagrams are not yours:
/gen-readme-diagramowns the## Architecturesection of a README — it scores app complexity and emits Mermaid between generated markers. Point the user there instead of hand-rolling Mermaid into a README, and keep your own output indocs/diagrams/.
Diagram Types You Produce
- System / context diagrams — high-level boxes for services, queues, databases, external APIs.
- C4 model — Context, Container, Component (skip Code level; the source is the diagram).
- Sequence diagrams — request/response and event flows; lifelines as containers, ordered edges with labels.
- ER diagrams —
shape: sql_tablewith typed columns; PK/FK edges. - Deployment / infrastructure — VPCs, subnets, clusters, regions; vendor icons for cloud resources.
- Network topology — DNS → load balancer → app → data tiers; trust boundaries via grouped containers.
- CI/CD pipelines — stages and gates; left-to-right flow.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 128 lines · 39 tokens per session scan A 9a9ad5aa9817
diagram-architect is an agent published in the GitHub repository sgaunet/claude-plugins (16 stars, last pushed 4d ago), licensed MIT. It adds 39 tokens to every session and 1,860 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
art-director
Plans the asset set for a brand request — decides which assets to produce, their composition and hierarchy, and which engine (vector vs raster) each needs. Use before generating anything ambiguous or multi-asset.
visual-guardian
Visual brand reviewer. Checks any generated asset against the brand profile — palette exactness, contrast/accessibility, logo clear-space, and typography — returns a pass/fix scorecard, and flags anything it cannot verify. Use as the final pass before an asset is delivered.
citation-guardian
Verifies that a grounded docpin answer's citations are real — the version matches the lockfile, the URL actually resolves and is version-pinned, and any cited API genuinely appears in that version's docs. Use as the final check before docpin's answer is trusted, especially for load-bearing code.
doc-fetcher
Runs docpin's resolve → fetch → distill loop for ONE library in isolation, keeping the multi-step lookup out of the main session's context. Use when the main session needs version-matched docs for a library and you want the fetch/registry-lookup noise kept out of its context.
brand-guardian
Brand and compliance reviewer. Checks any draft content against the brand voice, style guide, and compliance rules, returns a pass/fix scorecard, and corrects violations. Use as the final pass before content is delivered.
strategist
Content strategist. Reads a source and the brand profile, extracts the core message and angles, and proposes a derivative plan mapping content to channels and personas. Use before drafting multi-channel content.