- Home
- Skills
- Agents & Orchestration
- create hld
create hld
Generate and update detailed, editable draw.io architecture diagrams from your source code.
$5
Works with the AI Agents you already use
create hld
Example session with this skill installed
/create-hld auth-service-repo
- Read your context and instructions
- Compiled the create hld
- Generated the app files
Architecture analysis for auth-service-repo is complete. I have generated a four-page editable HLD based on the detected Node.js/Express stack and PostgreSQL database.
Generated HLD: auth-service-repo-hld.drawio
The file has been saved to docs/architecture/auth-service-repo-hld.drawio.
Page 1: Overview
- Tech Stack: Node.js v20.x, Express v4.18, TypeScript.
- Git: Commit
a1b2c3d- "feat: add jwt rotation". - Layers: Client -> Middleware (JWT, Rate Limit) -> Routes -> Controllers -> Services -> DAO -> PostgreSQL.
Page 2: API Contract
- GET
/health,/api/v1/users/:id - POST
/api/v1/auth/login,/api/v1/auth/register - DELETE
/api/v1/auth/logout - Total of 12 endpoints mapped.
Page 3: Dependencies & External Services
- Database: PostgreSQL (Primary), Redis (Session Cache).
- Integrations: SendGrid (Mail), Stripe (Billing).
- Key Packages:
prismav5.x,jsonwebtokenv9.x.
Page 4: Infrastructure & Environment
- Runtime: Docker (Node-Alpine image).
- Orchestration: Kubernetes (Deployment, Service, Ingress manifests).
- CI/CD: GitHub Actions (.github/workflows/deploy.yml).
- Env Vars:
DATABASE_URL,REDIS_HOST,JWT_SECRET(redacted).
Next steps
- Open the file in VS Code using the Draw.io Integration extension.
- Alternatively, run
open docs/architecture/auth-service-repo-hld.drawioand upload it toapp.diagrams.net. - Review the
Not detectedsections on Page 4 to manually add internal network details.
create-hld-app.zip
ZIP · project files
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Maintaining up-to-date architecture diagrams is a manual chore that often falls behind code changes. This skill automates the generation of professional, four-page, editable .drawio High-Level Design (HLD) documents by analyzing your source code.
What it does
- Source code analysis scans manifests, routes, and controllers to map your application architecture.
- Native draw.io output generates a fully editable XML file with four distinct pages: Overview, API Contract, Dependencies, and Infrastructure.
- Component mapping identifies layers like middleware, services, DAOs, and event consumers based on repository evidence.
- API discovery extracts REST endpoints and groups them by HTTP method and controller association.
- Infrastructure visualization maps Docker, Kubernetes, CI/CD pipelines, and monitoring tools found in the repo.
- Security-first documentation redacts secrets and credentials while documenting environment variable names and external service URLs.
How it works
- Use the
/create-hldcommand followed by a repository name or path. - The agent inspects
package.json,requirements.txt, and source directories to detect frameworks and architecture layers. - It generates a
.drawiofile containing the four standardized pages with a consistent, color-coded legend. - The agent provides the file path and instructions to open the diagram in a browser or VS Code.
Frameworks & tools
Works with any language or framework including Node.js (Express, NestJS, Next.js), Python (FastAPI, Django, Flask), and Java (Spring Boot). Supports infrastructure tools like Docker, Kubernetes, and GitHub Actions.
Why this beats prompting it yourself
Generic prompts often result in flat, non-editable text or simple Mermaid diagrams that lack repository-specific versions and deep architectural layers. This skill enforces a professional four-page structure and produces native XML that you can immediately edit and present to stakeholders.
Use cases
- Documenting a legacy repository for new team members.
- Refreshing architectural diagrams before a major system audit.
- Standardizing HLDs across a microservices monorepo.
Known limitations
Does not scan unrelated folders outside the workspace. Cannot detect architectural layers if code structure deviates significantly from standard patterns without directory hints.
Requires repository access. Annotations for layers depend on standard naming conventions (e.g., /services, /controllers). Cannot extract logic from obfuscated or binary files.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 26 days ago
Requires