← 返回 Skills 市场

Amazon Market Research

Name: Amazon Market Research
Author: lenger666

作者 lenger666 · GitHub ↗ · v1.0.5 · MIT-0

cross-platform ⚠ suspicious

364

总下载

当前安装

版本数

在 OpenClaw 中安装

/install amazon-market-research

功能描述

Generate professional Amazon market research reports using a complete 18-step product development framework. Output is optimized for direct Feishu display wi...

使用说明 (SKILL.md)

Amazon Market Research

Generate a professional Amazon market research report using a structured 18-step product development framework.

Output

This skill returns:

Full 18-step report directly in Feishu
Local markdown report file for reuse

Command

bash ~/.openclaw/workspace/skills/amazon-market-research/run.sh "{{input}}"

Example

/amazon-market-research 调研一下午餐盒在美国Amazon市场值不值得做 /amazon-market-research 分析一下 pet water fountain 在美国 Amazon 市场

Notes

Output is optimized for Feishu direct display
Report is returned as final answer, not summary
Full report is also saved as a markdown file locally

安全使用建议

Before installing, note these points: - The skill will send whatever prompt you provide to a third‑party model endpoint (default https://api.moonshot.cn/v1) and requires you to supply MARKET_API_KEY in a .env file. The registry did not declare these env vars, so the external network call is not obvious from the metadata. - If your prompts may contain sensitive or proprietary information, do not use this skill unless you trust the remote model provider and the API key scope. Consider testing with non‑sensitive examples first. - The skill auto‑creates a .env from ENV_TEMPLATE in the skill directory; review that file and set MARKET_BASE_URL to a provider you trust or remove the API key when not in use. - The skill runs run.sh which invokes python3 and skill.py. Ensure python3 is the expected interpreter and inspect the full skill.py file locally (you were shown a truncated excerpt) to confirm no other unexpected network calls or file operations. - If you want to limit blast radius, run the skill in an isolated environment, provide an API key with limited permissions, or host/point the skill to an in‑house model endpoint you control. Because of the undeclared credential requirement and the default external endpoint, I recommend reviewing and testing the code and .env behavior before trusting it with sensitive data.

功能分析

Type: OpenClaw Skill Name: amazon-market-research Version: 1.0.5 The skill is a legitimate tool designed to generate Amazon market research reports using an LLM API (defaulting to Moonshot AI). The code in skill.py and run.sh focuses on parsing user input, communicating with a user-configured API endpoint, and saving the resulting report as a local markdown file. It uses standard Python libraries, includes input sanitization for filenames, and shows no signs of data exfiltration, malicious execution, or persistence mechanisms.

能力评估

ℹ Purpose & Capability

The name/description (Amazon market research) matches the actual code: the skill parses a query and calls a model to produce an 18‑step report. However the registry metadata declares no required environment variables or binaries, while the shipped files clearly expect an external model API (MARKET_API_KEY, MARKET_BASE_URL, MARKET_MODEL) and execute via python3. That mismatch between declared requirements and actual files is noteworthy.

⚠ Instruction Scope

The SKILL.md tells the agent to run run.sh which sources .env and executes skill.py. skill.py will POST the full user input (and derived parameters) to an external chat/completions endpoint (cfg['base_url']/chat/completions) using MARKET_API_KEY. The SKILL.md and registry metadata do not explicitly declare this external network call or the need to provide an API key; user data will be sent to a third party by default (ENV_TEMPLATE defaults to https://api.moonshot.cn/v1). This is scope‑relevant: the skill transmits user prompts externally and saves a local markdown file in the skill directory.

ℹ Install Mechanism

There is no external install spec (no downloads), which is lower risk. But the skill is not purely instruction‑only: it ships run.sh and a ~18 KB Python script that will be executed. The runtime depends on python3 being available (run.sh invokes python3), though the registry declared no binaries. No remote archives or URL downloads are performed by the installer itself.

⚠ Credentials

The code requires MARKET_API_KEY, MARKET_BASE_URL, and MARKET_MODEL to function; these are reasonable for calling a model API. However the registry metadata lists no required env vars and no primary credential. The skill will auto-create a .env from ENV_TEMPLATE if missing and instructs the user to fill it — this hidden/undeclared credential requirement and the default base_url (api.moonshot.cn) are important to surface because they determine where data and any API key are sent.

✓ Persistence & Privilege

The skill is not marked always:true and does not request elevated persistence. It writes outputs (markdown report) into its own skill directory and reads/writes a local .env file within the skill folder — normal for this type of skill. It does not appear to modify other skills or global agent configs.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install amazon-market-research
安装完成后，直接呼叫该 Skill 的名称或使用 /amazon-market-research 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.5

- Output is now optimized for direct display in Feishu, providing the full report content immediately. - Skill now returns a local markdown file of the complete 18-step market research report for reuse. - Report is presented as the final answer, not just an executive summary. - Example commands and usage instructions updated for improved clarity.

v1.0.4

还原18条输出版本

v1.1.1

- 增加了“兼容模式”配置说明，当显式启用时可读取默认模型的认证资料来复用本机 API Key - 明确默认情况下不会读取其他 agent 的认证资料 - 其余功能与使用方式保持不变

v1.1.0

amazon-market-research v1.0.4 - Updated SKILL.md: replaced lengthy English documentation with a concise Chinese guide focused on product, usage, and integration details. - Added a CHANGELOG.md file for easier version tracking. - Clarified configuration and access scope, emphasizing support for default OpenClaw model settings. - Outlined structured JSON output format for integration. - Enhanced instructions on supported input styles and typical scenarios.

v1.0.3

Version 1.0.3 of amazon-market-research - No file changes detected in this version. - No feature, documentation, or configuration updates present. - Skill functionality and instructions remain unchanged from the previous release.

v1.0.2

- Expanded report structure to use a comprehensive 18-step professional Amazon product research framework. - Outputs now include executive summary, opportunity scores, recommended product directions, core risks, and full detailed analysis. - Enhanced descriptions and usage examples for clarity. - Added requirements for model configuration before use. - Refined intended audience to product researchers, managers, and e-commerce sellers.

v1.0.1

- Updated documentation: Added HOW_TO_USE.md, QUICK_START.md, and README.md for easier onboarding. - Improved instructions in SKILL.md for clearer usage, setup, and testing guidance. - Expanded description to specify supported use cases and recommended invocation method. - Removed CHANGELOG.md and PUBLISH_NOTES.md from the project.

v1.0.0

- Initial release of Amazon market research skill for cross-border ecommerce sellers - Generates detailed market research reports suitable for Feishu reading - Supports input of product name, target market, and sales platform - Useful for Amazon product selection, product development, competitor analysis, and market entry evaluation

元数据

Slug amazon-market-research

版本 1.0.5

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 8

常见问题