← Back to Skills Marketplace
554
Downloads
71
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install raccoon-data-analysis
Description
当用户需要使用小浣熊(Raccoon)进行数据分析会话管理、文件上传下载、数据可视化、数据分析交互时使用此技能。触发词包括"小浣熊数据分析"、"Raccoon数据分析"、"数据分析会话"。
Usage Guidance
This skill is internally consistent: it will read files you explicitly point it at and upload them to the configured RACCOON_API_HOST using the RACCOON_API_TOKEN. Before using it: (1) confirm you have permission to upload the files and they contain no sensitive information, (2) verify and trust the RACCOON_API_HOST endpoint (the SKILL suggests https://xiaohuanxiong.com but the host is configurable), (3) keep your token secret and rotate it if exposed, (4) run the provided scripts from the skill directory as instructed (avoid running broad filesystem searches), and (5) review scripts/main.py yourself if you want to inspect behavior — the script performs uploads, SSE streaming, and downloads but contains no opaque/obfuscated code. If any of these checks fail or you do not trust the remote service, do not use the skill.
Capability Analysis
Type: OpenClaw Skill
Name: raccoon-data-analysis
Version: 1.0.3
The skill is designed to perform remote data analysis by uploading local user files to a third-party API (xiaohuanxiong.com). While this behavior is explicitly documented and aligned with its stated purpose, the core functionality involves data exfiltration (uploading local files) and the execution of downloaded content via system openers (subprocess.run in scripts/main.py). Furthermore, SKILL.md contains high-priority instructions that explicitly forbid the agent from performing any local analysis, mandating that all user data be sent to the remote service, which increases the risk profile of the bundle.
Capability Assessment
Purpose & Capability
Name/description describe a remote-data-analysis skill. The skill requires python3 and two env vars (RACCOON_API_HOST, RACCOON_API_TOKEN) which are exactly what a remote-API client needs. Included scripts implement session creation, file upload, SSE-based chat, and artifact download — all coherent with the stated purpose.
Instruction Scope
SKILL.md explicitly restricts behavior to remote API use (forbids local data analysis) and instructs the agent to run the provided scripts; the instructions require absolute file paths for uploads and ask for explicit user consent before uploading. There is no instruction to read unrelated files, scan the filesystem, or exfiltrate unrelated secrets.
Install Mechanism
There is no install spec (instruction-only runtime) and the included Python script uses standard requests. No remote downloads or archive extraction occur during install. Risk from installation mechanisms is low.
Credentials
The skill asks only for RACCOON_API_HOST and RACCOON_API_TOKEN (primary credential). These are necessary and proportionate for authenticating to the remote API. The scripts do not request other unrelated credentials or config paths.
Persistence & Privilege
always is false and the skill does not request persistent system-wide privileges. It writes downloaded artifacts to ./raccoon/dataanalysis (per its purpose) and does not modify other skills or system-wide agent settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install raccoon-data-analysis - After installation, invoke the skill by name or use
/raccoon-data-analysis - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.3
Version 1.0.3
- 简化并优化 SKILL 描述与触发词，突出“小浣熊数据分析”核心用途。
- 删除了冗长的详细说明与例子，结构更紧凑，重点突出行为规则和要求。
- 保留核心规则、隐私保护与错误处理流程，内容更聚焦、易查阅。
- 使用场景、调用流程、注意事项等文档内容未做实质变更。
- 仅文档/说明调整，无脚本或功能改动。
v1.0.1
- Changed script organization: moved main.py to scripts/main.py for a clearer project structure.
- Removed the old main.py from the project root. No functional changes.
v1.0.0
Initial release of the raccoon-dataanalysis skill:
- Provides remote code interpretation and data visualization via the SenseTime-powered 小浣熊 (Raccoon) API.
- Users can upload Excel, CSV, and similar files for AI-driven analysis, chart generation, and automated insights.
- All analysis is performed remotely; local data analysis or visualization is strictly prohibited.
- Multi-turn data analysis sessions supported, including file handling and result downloading.
- Emphasizes security: requires explicit user consent, encrypted file transfer, and strict environment variable checks.
- Detailed privacy, error handling, and workflow guidelines included.
Metadata
Frequently Asked Questions
What is Raccoon Data Analysis?
当用户需要使用小浣熊(Raccoon)进行数据分析会话管理、文件上传下载、数据可视化、数据分析交互时使用此技能。触发词包括"小浣熊数据分析"、"Raccoon数据分析"、"数据分析会话"。 It is an AI Agent Skill for Claude Code / OpenClaw, with 554 downloads so far.
How do I install Raccoon Data Analysis?
Run "/install raccoon-data-analysis" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Raccoon Data Analysis free?
Yes, Raccoon Data Analysis is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Raccoon Data Analysis support?
Raccoon Data Analysis is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Raccoon Data Analysis?
It is built and maintained by 商汤小浣熊 (@raccoon-office); the current version is v1.0.3.
More Skills