N8n Workflows Maker
Terminal agent + Prompts = ready to import n8n workflow JSON files
Install
git clone https://github.com/jorgevz/n8n-workflows-makerInstall command taken from the project's README or registry entry. Always review code before running it.
README excerpt
Read full README on GitHub →Terminal agent + prompts = n8n Automation Workflows
This repo contains high-precision AI agent prompts to generate ready-to-import n8n workflow JSON files.
The goal is to make it easy to:
- Use a terminal AI agent as a workflow engineer to build production-ready n8n automations in JSON format.
- Validate and harden those workflows with dedicated QA and Security agents.
- Store and version-control both the prompts (brains) and the workflows (artifacts).
- Side note: This is a small personal project aimed to automated automations ( yes you read that right ) improvements will be actively made to the prompts and is not stable yet.
Repository Structure
n8n/
├── prompts/
│ ├── n8n_Build_Captain.md
│ ├── n8n_QA_Compliance.md
│ └── n8n_Security_Architect.md
│
├── workflows/
│ ├── workflow-1.json
│ ├── workflow-2.json
│
├── samples/
│ ├── n8n_sample.json
│ └── (future ones samples here)
│
└── README.md
prompts/
Contains three specialized Claude Code system prompts:
n8n_Build_Captain.md
- The orchestrator agent.
- Builds full n8n workflows from a plain-language spec.
- Outputs valid, import-ready JSON + README + sample payloads.
n8n_QA_Compliance.md
- The quality assurance agent.
- Validates JSON importability, node conventions, error handling, and adherence to n8n best practices.
- Returns targeted JSON patches and test plans.
n8n_Security_Architect.md
- The security review agent.
- Identifies risks in workflow design (auth, PII, DoS, SSRF, etc.).
- Suggests and provides JSON fragments for hardening.
workflows/
- This is where all importable
.jsonn8n workflows live for you. - Each JSON file should be valid for direct import into n8n via UI or CLI.
- These files are to remain private for your own use and not to be committed. If you wish to share workflows for the community you can place that
.jsonfile into the templates directory through a PR. Please remember to not share any of your secrets or private keys with your workflow. Once reviewed it will be merged.
templates/
The
Templatesdirectory contains community-contributed workflow JSON files.
If you’ve created a workflow you’d like to share, submit it via a Pull Request (PR) with your JSON file placed in theTemplatesdirectory.All submissions are reviewed before being merged to ensure quality and functionality.
Once approved, your workflow will be available for others to use and adapt.
samples/
- Stores canonical "reference" exports for n8n workflows — known-good examples that follow best practices and pass QA/Security reviews.
- These samples are not production workflows (skeleton); they exist to prime AI agents and serve as schema/style references.
- Example:
n8n_sample.jsoncontains a minimal but complete pattern (Trigger → HTTP → Transform → Respond + Error handling). - If you expand to other platforms (Make.com, Zapier, etc.), their samples can also live here, using their respective formats.
How to Use the Agents
- Open three Claude Code chats(optional, can be done with just one chat):
- Chat A → Paste
n8n_Build_Captain.md(Build Captain) - Chat B → Paste
n8n_QA_Compliance.md(QA & Compliance) - Chat C → Paste
n8n_Security_Architect.md(Security Architect)
- Chat A → Paste
Similar plugs
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1
Documented system prompts from Anthropic - Claude Fable 5.1, Opus 5.5, Claude Design, Claude Code. OpenAI - ChatGPT GPT-6-Astra, Codex. Google - Gemini 3.8 Flash, 3.1 Pro,…
Wrap Antigravity, ChatGPT Codex, Claude Code, Grok Build, Muse Code, Devin as an OpenAI/Gemini/Claude/Codex compatible API service, allowing you to enjoy the free Gemini…