amazon-assistant
Headless Amazon assistant — cookie-based session reuse to search, compare, sort, manage cart, track orders, surface deals, and analyze spending across amazon.in / amazon.com / amazon.co.uk / amazon.de / amazon.co.jp / etc. Checkout is prepare-and-handoff: the skill builds the cart and opens the final checkout URL; the user clicks Place Order.
Profile →npx skills add cosmicstack-labs/mercury-agent-skills --skill amazon-assistant --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Automates interactions with Amazon storefronts by reusing a user-supplied cookie-based session in a headed Chromium browser. It supports searching, comparing, viewing product details and reviews, managing the cart (view, add, remove, clear, sum), and preparing a checkout that opens the final checkout URL for human completion. It also offers deals and spending insights and can list or track orders.
How it works
The skill operates in three phases:
- Setup: capture cookies by opening a visible Chromium window, logging in manually, and saving cookies to ~/.mercury/amazon/cookies.json when login is detected within 6 minutes. The user is informed that the script saves cookies and closes the window.
- Operate: load saved cookies and perform scripted operations (search, cart, orders, deals, insights) against the appropriate amazon domain map (e.g., amazon.in, amazon.com, etc.). For read/write actions, the agent uses the amazon.py commands like search, product, compare, cart, deals, insights, orders, and checkout.
- Checkout: build the cart and open the checkout page (/spc/handlers/display.html) in the user’s browser via webbrowser.open(), then exit without placing the order; final submission must be done by the user.
Path A (primary) is the recommended login-capture path; Path B (paste) is a fallback that accepts cookie data exported via EditThisCookie when Path A is impractical.
When to use it
Use when you want to automate Amazon interactions across multiple storefronts (IN, US, UK, DE, JP, etc.) while preserving a real browser session with user cookies. Trigger phrases map to specific actions: search, compare, product details, cart operations, checkout, orders, deals, and insights. The flow enforces that final order submission is human-driven at checkout.
What it can touch
The skill operates by executing Python commands in the local environment, using a configured domain map and a cookie file at ~/.mercury/amazon/cookies.json. It references files and commands exactly as described in Setup and the core commands (e.g., amazon.py setup, amazon.py search, amazon.py cart add, amazon.py checkout, amazon.py whoami).
Caveats
Amazon prohibits automated access in its Terms of Use; the skill uses a real browser session with user cookies and may still trigger CAPTCHAs or account actions. The final submission is always performed by the user. Path A polls for login state with a 6-minute cap; if login isn’t detected, it instructs re-running setup. Path B requires explicit user actions, including exporting cookies from EditThisCookie and providing them via a paste flow. The tool requires explicit user consent and awareness of these limitations.
# Amazon Assistant 🛒 > **⚠ Terms-of-Service notice.** Amazon's [Conditions of Use](https://www.amazon.com/gp/help/customer/display.html?nodeId=508088) prohibit automated access, scraping, and bots. This skill drives a real browser session with the user's own cookies — Amazon may still treat that as a ToS violation, throw CAPTCHAs, or in repeated cases suspend the account. Use it on your own account, at a human pace, and accept the risk. The skill never places orders unattended — final submission is always a human click. --- ## Core Principle Amazon has no public consumer API for cart/checkout/order operations. The only viable automation path is **cookie-based session reuse with a headed (visible) browser**: 1. **Phase 1 — Setup (one-time per device/account):** open a visible Chromium, user logs in manually, cookies are captured. 2. **Phase 2 — Operate (reusable):** load cookies, run scripted operations (search, cart, orders, deals, insights). 3. **Phase 3 — Checkout (always human-confirmed):** the script builds the cart and opens the checkout page. **The user clicks "Place Order".** Why headed and not headless: Amazon's anti-bot stack (PerimeterX-style fingerprinting) detects `nav
- Core Principle
- When To Use This Skill
- Setup-state precedence (agent MUST follow this order)
- Trigger phrases
- When NOT to use
- Setup
- 1. Resolve the marketplace domain
- 2. Install runtime
- 3. Capture cookies — two paths, in order of preference
- 4. Verify (works for both paths)
- Core Capabilities
- Search
- Product detail + reviews
- Compare items
pip install playwright beautifulsoup4 lxml rich playwright install chromium python3 amazon.py setup python3 amazon.py setup --paste python3 amazon.py whoami python3 amazon.py search "wireless noise cancelling headphones" \ python3 amazon.py product B0BDHB9Y8H --reviews 10 python3 amazon.py compare B0BDHB9Y8H B0CHX1W1XY B09XS7JWHH --reviews 5 python3 amazon.py cart view # list items + line totals + grand total python3 amazon.py cart add B0BDHB9Y8H --qty 1
What does the amazon-assistant skill do?
Headless Amazon assistant — cookie-based session reuse to search, compare, sort, manage cart, track orders, surface deals, and analyze spending across amazon.in / amazon.com / amazon.co.uk / amazon.de / amazon.co.jp / etc. Checkout is prepare-and-handoff: the skill builds the cart and opens the final checkout URL; the user clicks Place Order.
How do I install it?
Run `npx skills add cosmicstack-labs/mercury-agent-skills --skill amazon-assistant --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From cosmicstack-labs/mercury-agent-skills, a repository with 364 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.