architecture-diagrams
Skill3 starsChanged 5 months agoMITView on GitHub
Turn a natural-language cloud architecture description into an editable diagram whose nodes are official vendor service icons (Azure, AWS, GCP, Kubernetes), exported as PDF, draw.io, or VSDX (Lucid-importable). Use when the user asks to draw, diagram, or visualize a cloud/system architecture, or to produce a .drawio/.vsdx/.pdf architecture diagram with real service icons.
- Installs packages
8 other projects use a version of this.
Adopted by
Most recently taken up by
- excalimate/excalimatetheir own version, first seen Oct 31, 2026
- microsoft/hve-coretheir own version, first seen Oct 13, 2026
- quangdangle/architecture-diagrams-skilltheir own version, first seen Sep 25, 2026
- Leostruka/devin-bundletheir own version, first seen Aug 15, 2026
- craig-ai-tooling/claude-skillstheir own version, first seen Jul 10, 2026
- nctiggy/claude-skillstheir own version, first seen Jul 10, 2026
Other versions8 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 8
- microsoft/hve-core★ 1.6ksame name
- excalimate/excalimate★ 99same name
- nilecui/SkillsBase★ 27same name
- vishalc412/Architecture-ClaudeSkill★ 5same name
- craig-ai-tooling/claude-skills★ 1same name
- nctiggy/claude-skills★ 1same name
- Leostruka/devin-bundle★ 0same name
- quangdangle/architecture-diagrams-skill★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 8 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 97
If the engine is running as an MCP server (python -m mcp_server.server), call the tools directly: validate_spec(spec_json), …
“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Architecture Diagrams (vendor-icon)
Translate an architecture description into a small JSON spec, then render it
with the archdiagram engine. Nodes become the real vendor service icons, not
labelled boxes.
Workflow
- [ ] 1. Read the architecture (text, image, or existing diagram)
- [ ] 2. Map each component to a catalog service id (search the catalog)
- [ ] 3. Write the JSON spec (nodes, edges, groups)
- [ ] 4. Validate the spec
- [ ] 5. Render to the requested format(s)Setup (once)
git clone https://github.com/bharathgnana/archdiagram.git
cd archdiagram
python -m tools.download_icons # fetch vendor icons into ./icons (not committed)
cd node && npm install && cd .. # optional: enables PDF + crisp raster iconsThe Python core is stdlib-only. Node is only needed for .pdf and for
rasterising SVG icons into .vsdx/.pdf.
Step 2: find service ids
python -m archdiagram.cli catalog search "kubernetes"
python -m archdiagram.cli catalog list --vendor azureA service id is "<vendor>.<service_key>", e.g. azure.aks, aws.s3,
gcp.cloud_run. Unknown ids still render as a clean vendor-accented fallback
box, so prefer the closest real id and proceed.
Step 3: spec format
{
"title": "Web platform on Azure",
"direction": "LR",
"groups": [
{ "id": "app", "label": "Application tier", "vendor": "azure" }
],
"nodes": [
{ "id": "web", "service": "azure.app_service", "label": "Web App", "group": "app" },
{ "id": "aks", "service": "azure.aks", "label": "AKS", "group": "app" },
{ "id": "kv", "service": "azure.key_vault", "label": "Key Vault" }
],
"edges": [
{ "source": "web", "target": "aks", "label": "api" },
{ "source": "aks", "target": "kv", "label": "secrets" }
]
}Rules:
direction:LR,RL,TB, orBT(defaultLR).nodes[].service: required,"<vendor>.<service_key>".nodes[].group: optional; must match agroups[].id.nodes[].x/nodes[].y: optional explicit coordinates (override auto-layout).edges[].source/target: must reference node ids.directeddefaults true.- Layout is automatic and deterministic; only set coordinates to override.
Step 4-5: validate and render
python -m archdiagram.cli validate spec.json
python -m archdiagram.cli render spec.json -f pdf -o out/arch.pdf
python -m archdiagram.cli render spec.json -f drawio -o out/arch.drawio --drawio-mode portable
python -m archdiagram.cli render spec.json -f vsdx -o out/arch.vsdxChoosing a format
| Want | Use |
| Reliable visual fidelity in Lucidchart (icons always render) | pdf |
| Editable in draw.io / diagrams.net | drawio --drawio-mode native |
| Editable after Lucid/Visio import (embedded icons) | drawio --drawio-mode portable or vsdx |
Lucid does not map draw.io named stencils, so use portable (embedded icons)
or pdf when the target is Lucidchart.
Using via MCP instead of the CLI
If the engine is running as an MCP server (python -m mcp_server.server), call
the tools directly: validate_spec(spec_json), render_diagram(spec_json, fmt,
out_path, drawio_mode), search_catalog(query), list_catalog(vendor).
Additional resources
- Spec schema and engine details: ../README.md
- Extend the catalog: add a service entry under
archdiagram/registry/catalog_data/<vendor>.jsonand an icon source intools/icon_sources.json.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

