China Shopping

Recommend suitable Chinese shopping platforms for a user-provided product type, product name, or shopping need. Use when the user asks where to buy something...

Security Scan

The skill is internally coherent: it uses a local Python script and bundled JSON data to recommend Chinese shopping platforms, requests only python3, and does not ask for credentials or network access.

Purpose & Capability

Name/description match the actual implementation. The skill only requires python3 and the included data files to map product names to categories and recommend platforms — all appropriate for a shopping-recommendation skill.

Instruction Scope

SKILL.md instructs the agent to run the local CLI (python3 china-shopping.py ...), which matches the code. Minor inconsistencies: the README and SKILL.md list data/product_mapping.json and data/general_fallback.json as used files, but the CLI loads data/categories.json and uses an internal PRODUCT_KEYWORDS table; product_mapping.json and general_fallback.json are present but not actually read by the code. This is a coherence/maintenance issue but not a security risk. The instructions do not direct reading unrelated system files or environment variables.

Install Mechanism

No install spec; this is instruction-only with a bundled Python script and local JSON files. Nothing is downloaded or executed from external URLs at install time.

Credentials

The skill requests no environment variables, no credentials, and no config paths. This is proportionate to its stated function.

Persistence & Privilege

always:false and no requests to modify other skills or system configuration. The skill runs only when invoked and does not demand elevated persistence.

Assessment

This skill appears to be what it claims: a local, offline recommender that reads bundled data and prints suggestions. It does not request credentials or make network calls. Before installing, consider: (1) review the bundled data files if you need to ensure recommendations match your expectations (there are duplicate/unused data files present), (2) the code is local Python — run it in a safe environment if you want to validate outputs, and (3) the SKILL.md and code have minor maintenance mismatches (unused JSON files and duplicated mappings) but nothing indicating exfiltration or malicious behavior.

Recommend suitable Chinese shopping platforms based on product category and shopping intent.

This is a lightweight Python-backed skill. It uses a local Python script plus bundled JSON data to map product names to shopping categories and recommend suitable Chinese e-commerce platforms.

It does not perform live browsing, real-time price checks, or seller verification. For live page inspection, real-time pricing, or store-level comparison, switch to browser-based workflows instead of pretending this skill does live retrieval.

Runtime requirement

Require:

• python3

Do not require:

• jq
• shell helper scripts
• install scripts
• writable config or log paths
• credentials or API keys

Files used by this skill

• china-shopping.py — local CLI implementation
• data/categories.json — category and platform recommendation data
• data/product_mapping.json — product keyword mapping
• data/general_fallback.json — fallback recommendations

Read these references as needed:

• references/category-guide.md for category-to-platform guidance
• references/output-patterns.md for answer structure

Workflow

1. Identify the product category or shopping intent.

   • Accept a product type, shopping need, or product name.
   • If the request is too broad, ask one short clarifying question.

2. Use the local Python implementation when execution is appropriate.

   • Run python3 china-shopping.py recommend "<product>" for the default recommendation flow.
   • Use python3 china-shopping.py categories when the user wants to inspect supported categories.

3. Explain the recommendation.

   • Say why the recommended platforms fit.
   • Mention meaningful trade-offs when useful.

4. Add practical guidance.

   • Suggest what the user should check next, such as official stores, seller reputation, user reviews, authenticity, or delivery terms.

Output

Use this structure unless the user asks for something else:

Recommended Platforms

List the most suitable 2-4 platforms.

Why

Explain why each platform fits the product or need.

Best Choice

State which platform is the strongest default recommendation.

Caveats

Mention important cautions, such as seller quality differences, authenticity risk, or category-specific trade-offs.

Final Advice

Give a practical buying suggestion in plain language.

Quality bar

Do:

• recommend platforms by category fit
• explain trade-offs clearly
• mention official stores or trusted sellers when relevant
• stay honest about not doing real-time price retrieval

Do not:

• pretend to check live listings or prices
• claim a platform is cheapest without real-time evidence
• suggest weak-fit platforms just because they are famous