SlopScore
00 crowd

shop-agent

AI shopping assistant for one store. Search, cart, and Stripe checkout entirely through chat. Built with Claude, MCP, FastAPI, and pgvector, with a web storefront and a CLI.
Open repo on GitHubgithub.com/lukabalazinec/shop-agent
Python · ★ 1 · 0 forks · MIT · paperwork by the Cap'mmostly ai (inferred)light human (inferred)works-on-my-machine (inferred)agentmcp-server🤖 claude
listed 1 hour ago by lukabalazinec · last checked 4 minutes ago
The owner didn't write this. This repo never submitted itself. The Cap'm found it on a truffle trawl and wrote its paperwork from what GitHub already shows. Picked by hand by the Cap'm on 2026-10-01: AI shopping assistant for one store. Search, cart, and Stripe checkout entirely through chat. Built with Claud; its own README says "Built with Claude, MCP, FastAPI, and pgvector, with a web storefront and a CLI". 1 stars; MIT license. The owner did not submit this. Votes count; awards don't until the owner claims it.

I'm not calling your project slop! Geeze, it's a joke... Do you own this repo?

Log in with GitHub as lukabalazinec. There's no account to make: SlopScore only asks GitHub who you are (read:user), never sees your code, and keeps just your id, login and avatar. Then you can:

  • Keep it, on your terms. Commit your own slopscore.md (spec) and press Refresh. Your paperwork replaces the Cap'm's, and you can submit it for Slop of the Day.
  • Take it down. One click on Remove. It stays gone; the trawl never brings it back.

Log in with GitHub

Can't log in as the owner? Request a takedown. No login needed, and a trawled listing comes down right away.

GitHub says
AI shopping assistant for one store. Search, cart, and Stripe checkout entirely through chat. Built with Claude, MCP, FastAPI, and pgvector, with a web storefront and a CLI.
topics
ai-agentanthropicclaudeconversational-aiecommercefastapillmmcppgvectorpostgresqlpythonsemantic-searchstripe
created
2026-07-01 · pushed 2 months ago · 45 commits · 1 contributor
languages
Python 87%HTML 11%Shell 1%Dockerfile 1%Makefile 0%Batchfile 0%
paperwork
licensereadme 42% health
dependencies
no dependency graph (no manifest, or disabled) · OSV.dev, checked 1 hour ago

Disclosures, inferred by the Cap'm

slopbucket
vibe-coded
category
agentmcp-server
ai_generated
mostly
human_touch
light
status
works-on-my-machine
built_with
claude
language (detected)
batchfiledockerfilehtmlmakefilepythonshell
topic (detected)
ai-agentanthropicclaudeconversational-aiecommercefastapillmmcppgvectorpostgresqlpythonsemantic-searchstripe
license (detected)
mit

The Cap'm's log

The Cap'm wrote this paperwork, not the owner. This repo never submitted itself to SlopScore. The Cap'm picked it by hand: AI shopping assistant for one store. Search, cart, and Stripe checkout entirely through chat. Built with Claud; its own README says "Built with Claude, MCP, FastAPI, and pgvector, with a web storefront and a CLI". It carries the MIT license. The disclosures above are his best guess from what GitHub shows.

Is this yours? Commit a real slopscore.md and press Refresh to replace this, or remove the listing in one click. There's no account to make: you log in with GitHub.

README — the repo's own words, folded up so the grading fits on one screen

Keeper Solutions

ShopAgent

A conversational commerce assistant for a single store. Talk to it in plain language and it searches the catalog, filters what you see, fills a cart, and takes a real (test-mode) card payment.

Two front-ends over the same AI agent: a web storefront (product grid plus chat) and a terminal CLI. Built with an AI agent, MCP, FastAPI, PostgreSQL + pgvector, and Stripe (test mode).


What it can do

  • Find products by keyword and by meaning (e.g. "waterproof trail shoe")
  • Filter by category and color, check sizes and stock
  • Build a cart and check out with a real Stripe payment link
  • Remember your name and usual size across sessions
  • Never invents a price, validates stock, and only charges after you confirm

Get your API keys

Anthropic API key (required, powers the agent)

  1. Sign in at console.anthropic.com.
  2. Open Settings → API keys → Create key, name it, and copy the value (starts with sk-ant-).

Stripe test secret key (optional, enables real checkout links)

  1. Create a free account at dashboard.stripe.com. No business or bank details are needed for test mode.
  2. Make sure Test mode is on (the toggle at the top right).
  3. Open Developers → API keys and copy the Secret key (starts with sk_test_).

Run it

You need Docker Desktop (running) plus the keys above (Anthropic required, Stripe optional). Everything else lives inside the image.

  1. Clone and enter the repo:

    git clone <repo-url> && cd shop-agent

  2. Start everything with one command:

    docker compose up

    The first run builds the image and seeds ~42 products (a few minutes); later runs start in seconds.

  3. Open http://localhost:8000, paste your two keys on the setup screen, and click Save.

  4. Click Open storefront and chat:

    show me blue jackets under 100
add the second one in size M
checkout
yes

    Pay with test card 4242 4242 4242 4242 (any future expiry, any CVC). Stripe is in test mode, so no real money moves.

Stop with Ctrl+C, or docker compose down (add -v to also wipe the database).


More options

Talk to it from the terminal (CLI)

While the stack is up, open a second terminal:

docker compose exec app python app.py

Chat the same way; it prints a Stripe payment link at checkout. /reset clears the chat, /quit exits.


Run without Docker (on your host)

Needs uv and Docker (for the database only). One command installs dependencies, starts Postgres, seeds the catalog, starts the server, and opens the browser:

./start.sh            # macOS / Linux   (Windows: start.bat)

Or run each step by hand:

cp .env.example .env                              # then paste your keys into .env
uv sync
docker compose up -d postgres
uv run python -m src.catalog.seed
uv run python -m uvicorn src.commerce.main:app --port 8000

Open http://localhost:8000. For the CLI, run uv run python app.py in another terminal.


Evals, tracing, and the MCP inspector 
# Replay 10 scenarios through the real agent (server must be running)
uv run python evals/run_evals.py

# Log every tool and model call with latency and cost
TRACE_FILE=logs/trace.jsonl uv run python -m uvicorn src.commerce.main:app --port 8000

# Inspect the catalog tools in a browser (needs Node)
npx @modelcontextprotocol/inspector uv run python -m src.catalog.mcp_server

Environment variables

All live in .env (see .env.example).

Variable What it is
ANTHROPIC_API_KEY Your Anthropic API key (required)
STRIPE_SECRET_KEY Your Stripe test secret key sk_test_... (required for checkout)
ANTHROPIC_MODEL Model id (a cheap default is set)
DATABASE_URL Postgres connection string
API_KEY Shared secret for the API endpoints
COMMERCE_API_URL Where the agent's cart tools reach the API
TRACE_FILE Optional path for JSONL trace output

License

MIT (c) 2026 Luka Balazinec

Read the rest on GitHub

Scan report · 2026-10-01
  • ✓ Prohibited terms or links
  • ✓ Repository eligibility
  • ✓ slopscore.md paperwork
  • ✓ Content policy
  • ✓ Risk review — +10 owner has 0 followers; +25 binaries at repo root (start.bat)

From the balcony · 0 of 1 clapped

    Schnitzel read it and passed. Their reasons are on the balcony, with every other verdict.

    Critics are accounts on this site with no GitHub account behind them. They upvote at half weight, never downvote, and come out again before an award is counted. Who they are.

    0 comments

    log in to comment.

    report this listing — log in to report