Note: This repository contains Baidu AI Cloud's official collection of skills for Claude. For information about the Agent Skills standard, see agentskills.io.

BCE Skills

Skills are folders of instructions, scripts, and resources that Claude loads dynamically to improve performance on specialized tasks. Skills teach Claude how to complete specific tasks in a repeatable way, whether that's integrating with Baidu AI Cloud services, processing data using BCE-specific workflows, or automating cloud-native tasks.

For more information, check out:

• Agent Skills Specification
• What are skills?
• Using skills in Claude
• How to create custom skills

About This Repository

This repository is the official skills collection maintained by Baidu AI Cloud (BCE). It contains skills that demonstrate what's possible with Claude's skills system, with a focus on Baidu AI Cloud ecosystem integration and Chinese language support.

Each skill is self-contained in its own folder with a SKILL.md file containing the instructions and metadata that Claude uses. Browse through these skills to get inspiration for your own skills or to understand different patterns and approaches.

Skill Sets

• ./skills: Official BCE skill collection
• ./spec: The Agent Skills specification
• ./template: Skill template for creating new skills

Skills Content

famou-artifact-generator

Description: Create the core materials for a FaMou task from an initial idea or problem statement.

Purpose:

• Clarify requirements and generate problem.md
• Generate init.py, evaluator.py, and prompt.md
• Verify that the initial solution and evaluator can run correctly

famou-data-analysis

Description: Analyze and understand datasets used in FaMou tasks or standalone data analysis work.

Purpose:

• Understand data structure and field meanings
• Identify data quality issues such as missing values, duplicates, and anomalies
• Produce analysis conclusions and data processing suggestions

famou-experiment-manager

Description: Manage the lifecycle of FaMou experiments through configuration, submission, and result retrieval.

Purpose:

• Check famou-ctl availability and API configuration
• Find config.yaml and submit experiments
• Query status, inspect logs, delete experiments, and fetch results

famou-result-visualization

Description: Convert a final Python-form FaMou solution into an interactive visualization page.

Purpose:

• Read and understand the structure of the solution
• Choose a suitable visualization style for the problem type
• Generate an HTML page to display the final solution result

baidu-search

Description: Search the web using Baidu AI Search Engine (BDSE). Use for live information, documentation, or research topics.

Purpose:

• Search the web via Baidu AI Search API for real-time information retrieval
• Support flexible query parameters including result count and time range filtering
• Requires BAIDU_API_KEY configured via Baidu AI Cloud console

medical-bill-organizer

Description: Medical bill organizer that automatically classifies, performs OCR recognition, and extracts key information from medical documents.

Purpose:

• Automatically classify medical bills into 28 categories (e.g., ID cards, medical records, invoices, prescriptions, etc.)
• Perform OCR recognition and extract key information such as hospital names, admission dates, and billing amounts
• Generate a summarized CSV file of all invoices with itemized and total amounts
• Support folder or archive (zip/rar/7z) input for batch processing

qianfanocr-document-intelligence

Description: Analyze image files, image URLs, PDF files, and PDF URLs for visual content understanding, including document parsing, layout analysis, OCR, key information extraction, chart understanding, and document VQA.

Purpose:

• Perform document parsing to extract structured Markdown from images and PDFs
• Support multiple analysis modes: document parsing, layout analysis, element recognition, general OCR, key information extraction, chart understanding, and document VQA
• Provide bundled CLI tools for image/PDF processing with flexible batch and concurrent execution
• Requires QIANFAN_TOKEN configured via Baidu Qianfan platform

Try in Claude Code

Method 1: Plugin Marketplace

You can register this repository as a Claude Code Plugin marketplace by running the following command in Claude Code:

/plugin marketplace add baidubce/skills

Then, to install a specific set of skills:

1. Select Browse and install plugins
2. Select bce-agent-skills
3. Select the skill set you want
4. Select Install now

Method 2: Manual Copy

Clone the repository and copy the skills to your Claude skills directory:

rm -rf ~/.claude/skills/famou-*                    # remove old version
git clone https://github.com/baidubce/skills.git   # download from github
cp -r skills/skills/* ~/.claude/skills/            # install to claude-code config directory
npx skills list -g                                 # view all installed skill

Creating a Basic Skill

Skills are simple to create - just a folder with a SKILL.md file containing YAML frontmatter and instructions. You can use the template in this repository as a starting point:

---
name: my-skill-name
description: A clear description of what this skill does and when to use it
---

# My Skill Name

[Add your instructions here that Claude will follow when this skill is active]

## Examples
- Example usage 1
- Example usage 2

## Guidelines
- Guideline 1
- Guideline 2

The frontmatter requires only two fields:

• name - A unique identifier for your skill (lowercase, hyphens for spaces)
• description - A complete description of what the skill does and when to use it

The markdown content below contains the instructions, examples, and guidelines that Claude will follow.

For Developers

Branch Strategy

Workflow

1. Fork this repository and create your feature branch from develop
2. Commit your changes and push to your fork
3. Open a Pull Request targeting the develop branch of baidubce/skills
4. A reviewer will review your PR and merge it into develop
5. Periodically, develop is merged into main to cut a new stable release

Note: Do not submit PRs directly to main. All contributions must go through develop for review.

Contributing

We welcome contributions from the community! To add a new skill:

1. Fork this repository
2. Create a feature branch from develop
3. Create a new folder under skills/ with your skill name
4. Add a SKILL.md file following the Agent Skills specification
5. Add a LICENSE.txt file if applicable
6. Submit a Pull Request to the develop branch

License

Skills in this repository are licensed under the Apache 2.0 License unless otherwise noted. See individual skill directories for specific license information.