https://github.com/theprojectsx/postman_automate
An AI Automated runner to test postman collections. Simple agentic workflow to validated, create, execute requests.
https://github.com/theprojectsx/postman_automate
Last synced: 16 days ago
JSON representation
An AI Automated runner to test postman collections. Simple agentic workflow to validated, create, execute requests.
- Host: GitHub
- URL: https://github.com/theprojectsx/postman_automate
- Owner: TheProjectsX
- Created: 2026-04-07T05:14:25.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2026-04-18T10:57:36.000Z (2 months ago)
- Last Synced: 2026-04-18T12:32:48.548Z (2 months ago)
- Language: TypeScript
- Size: 185 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Postman AI Automator
**Status:** Beta v1
A small CLI that turns a Postman Collection into a step-by-step JSON workflow (REGISTER / INPUT / LOG / EXECUTE), using Gemini to plan the flow and then execute it. It also saves generated workflows for reuse and writes detailed logs while it runs.
Currently it is in Beta test version. Many more things there is to be fixed. And I will continue to work on this.
**Flow (what / where / why / how)**
- What: Convert a Postman Collection into an ordered JSON workflow of actions (REGISTER, INPUT, LOG, EXECUTE).
- Where: Input is your `collection.json`; output is stored in `workflows/` with run logs in `logs/process_logs/`.
- Why: Enforces consistent API execution order and variable handling, even for complex collections.
- How: Parses and simplifies the collection, sends a structured prompt to Gemini, validates the JSON response, then optionally executes each step.
**Quick start**
1. Create `.env` with `GEMINI_API_KEY`.
2. Run:
```bash
node index.ts .json
# With workflow
node index.ts .json --workflow workflows/workflow_2026-04-07_10-28-17.json
```
**Outputs**
- Generated workflows are saved under `workflows/`.
> AI generated workflow is saved so that it can be used later w/o calling AI for same request paths. Tokens saved!
- Run logs are written under `logs/process_logs/`.
> No thing is missed, every action, every response, every request is logged here to debug later
- Run errors are written under `logs/error_logs`.
> If a route faces error, detailed response with request is saved to check and debug later
- Response saved postman collections exported under `exports/`.
> The requests and responses of the executions are saved in the postman collection to import in postman and use it!
**Flags**
```bash
--delay= Delay between requests (default: 0)
--timeout= Request timeout in ms (default: 30000)
--skip= Skip requests whose URL contains pattern
--only= Only run requests whose URL contains pattern
--workflow= Use a pre-generated workflow JSON (alias: -wf=)
--context= Extra AI instructions (alias: -c=)
--dry Dry run - plan workflow but do not send requests
```