ai-inventory-agent

KO | EN

gitlite — search

#monitoring #claude #json #rust #cpp #developer-tools #rag #video #dotnet #typescript

ai-inventory-agent

(★ 9)

No description available.

File Explorer

ZIP Download

- styles.css
- sample-products.json
- add-product.html
- chat.html
- dashboard.html
- edit-product.html
- login.html
- products.html
- reports.html
- app.js
.gitignore
package-lock.json
package.json
README.md
server.js

# Use via CDN

jsDelivr

jsDelivr serves any public GitHub repository as a CDN with zero setup. Pick a version and a file to get a ready-to-paste link and snippet.

Version

File

Link

Example

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

This is your frontend (`frontend/`, `css/`, `js/` — copied unchanged from your repo) wired up
to a working backend. The original repo only had a Spring Boot backend that wasn't buildable
(missing source structure), so instead of trying to reconstruct that, I wrote a small Node/Express
backend that implements every API endpoint your frontend already calls:

- `POST /api/login`, `GET /api/session-check`, `POST /api/logout`
- `GET /api/dashboard`
- `GET /api/products` (pagination, search, category/status filter, sorting)
- `POST /api/products`, `GET/PUT/DELETE /api/products/:id`
- `GET /api/products/low-stock`, `GET /api/products/out-of-stock`
- `POST /api/products/import-sample`
- `POST /api/chat` (rule-based inventory Q&A — no external AI API key needed)

Data is stored in `data/products.json` (plain JSON file, created automatically). No database
setup, no native module compilation — just Node.

## Run it

```bash
npm install
npm start
```

Then open **http://localhost:3000/login.html**

On first run there's no data — go to **Products** and click **Seed Sample Data**, or add a
product yourself.

## Notes / things you should know

- The chat endpoint is NOT a real LLM. It's keyword matching (low stock / out of stock / value /
  category / total). It works, but don't call it "AI-powered" in a resume bullet without being
  clear that it's rule-based, not an LLM integration. If you want it to actually be AI, swap the
  `/api/chat` handler to call the Anthropic or OpenAI API instead — the frontend already expects
  `{ response: "markdown text" }` back, so the contract won't change.
- Sessions are in-memory (`express-session` default store). Restarting the server logs everyone
  out — fine for a demo/project, not fine for production.
- No password hashing, single hardcoded admin account, no rate limiting on login. This is a
  student-project-grade backend, not something to expose on the public internet as-is.
- `data/products.json` is created at runtime — delete it to reset the app to a clean slate.

ai-inventory-agent

Owner Repos

File Explorer

# Use via CDN

Was this content helpful?

Recent Feedback

ai-inventory-agent

Owner Repos

File Explorer

# Use via CDN

Was this content helpful?

Recent Feedback

Download README