[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71JE8K4XP23N1WM5CS7PPEW":3,"skill-same:01M71JE8K4XP23N1WM5CS7PPEW":70},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":42,"linkOnly":43,"markdownContent":44,"markdownBody":45,"frontMatter":46,"refName":47,"commitSha":48,"skillDirectory":5,"forksCount":49,"repositoryHtmlUrl":50,"repositoryHomepageUrl":51,"ownerHtmlUrl":52,"ownerBlogUrl":53,"ownerDescription":54,"updatedAt":55,"install":56,"resources":69},"01M71JE8K4XP23N1WM5CS7PPEW","create-skill",null,"Guides users through creating effective Agent Skills for Cursor. Use when you want to create, write, or author a new skill, or asks about skill structure, best practices, or SKILL.md format.","caezium\u002Fskills","caezium","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F113233555?v=4","01M71JB77ZPBAWEK009Y2W8BP7","Henry Zhang","01M71JB792NVXCWSYD27HQ0964",3,"MIT","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fcreate-skill\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41],"create","create skill","format","about","cursor","users","author","best practices","use","for","new skill","effective agent skills","write","structure","guides","you","when","new","2026-11-10 10:12:30",false,"---\nname: create-skill\ndescription: >-\n  Guides users through creating effective Agent Skills for Cursor. Use when you\n  want to create, write, or author a new skill, or asks about skill structure,\n  best practices, or SKILL.md format.\n---\n# Creating Skills in Cursor\n\nThis skill guides you through creating effective Agent Skills for Cursor. Skills are markdown files that teach the agent how to perform specific tasks: reviewing PRs using team standards, generating commit messages in a preferred format, querying database schemas, or any specialized workflow.\n\n## Before You Begin: Gather Requirements\n\nBefore creating a skill, gather essential information from the user about:\n\n1. **Purpose and scope**: What specific task or workflow should this skill help with?\n2. **Target location**: Should this be a personal skill (~\u002F.cursor\u002Fskills\u002F) or project skill (.cursor\u002Fskills\u002F)?\n3. **Trigger scenarios**: When should the agent automatically apply this skill?\n4. **Key domain knowledge**: What specialized information does the agent need that it wouldn't already know?\n5. **Output format preferences**: Are there specific templates, formats, or styles required?\n6. **Existing patterns**: Are there existing examples or conventions to follow?\n\n### Inferring from Context\n\nIf you have previous conversation context, infer the skill from what was discussed. You can create skills based on workflows, patterns, or domain knowledge that emerged in the conversation.\n\n### Gathering Additional Information\n\nIf you need clarification, use the AskQuestion tool when available:\n\n```\nExample AskQuestion usage:\n- \"Where should this skill be stored?\" with options like [\"Personal (~\u002F.cursor\u002Fskills\u002F)\", \"Project (.cursor\u002Fskills\u002F)\"]\n- \"Should this skill include executable scripts?\" with options like [\"Yes\", \"No\"]\n```\n\nIf the AskQuestion tool is not available, ask these questions conversationally.\n\n---\n\n## Skill File Structure\n\n### Directory Layout\n\nSkills are stored as directories containing a `SKILL.md` file:\n\n```\nskill-name\u002F\n├── SKILL.md              # Required - main instructions\n├── reference.md          # Optional - detailed documentation\n├── examples.md           # Optional - usage examples\n└── scripts\u002F              # Optional - utility scripts\n    ├── validate.py\n    └── helper.sh\n```\n\n### Storage Locations\n\n| Type | Path | Scope |\n|------|------|-------|\n| Personal | ~\u002F.cursor\u002Fskills\u002Fskill-name\u002F | Available across all your projects |\n| Project | .cursor\u002Fskills\u002Fskill-name\u002F | Shared with anyone using the repository |\n\n**IMPORTANT**: Never create skills in `~\u002F.cursor\u002Fskills-cursor\u002F`. This directory is reserved for Cursor's internal built-in skills and is managed automatically by the system.\n\n### SKILL.md Structure\n\nEvery skill requires a `SKILL.md` file with YAML frontmatter and markdown body:\n\n```markdown\n---\nname: your-skill-name\ndescription: Brief description of what this skill does and when to use it\n---\n\n# Your Skill Name\n\n## Instructions\nClear, step-by-step guidance for the agent.\n\n## Examples\nConcrete examples of using this skill.\n```\n\n### Required Metadata Fields\n\n| Field | Requirements | Purpose |\n|-------|--------------|---------|\n| `name` | Max 64 chars, lowercase letters\u002Fnumbers\u002Fhyphens only | Unique identifier for the skill |\n| `description` | Max 1024 chars, non-empty | Helps agent decide when to apply the skill |\n\n---\n\n## Writing Effective Descriptions\n\nThe description is **critical** for skill discovery. The agent uses it to decide when to apply your skill.\n\n### Description Best Practices\n\n1. **Write in third person** (the description is injected into the system prompt):\n   - ✅ Good: \"Processes Excel files and generates reports\"\n   - ❌ Avoid: \"I can help you process Excel files\"\n   - ❌ Avoid: \"You can use this to process Excel files\"\n\n2. **Be specific and include trigger terms**:\n   - ✅ Good: \"Extract text and tables from PDF files, fill forms, merge documents. Use when working with PDF files or when the user mentions PDFs, forms, or document extraction.\"\n   - ❌ Vague: \"Helps with documents\"\n\n3. **Include both WHAT and WHEN**:\n   - WHAT: What the skill does (specific capabilities)\n   - WHEN: When the agent should use it (trigger scenarios)\n\n### Description Examples\n\n```yaml\n# PDF Processing\ndescription: Extract text and tables from PDF files, fill forms, merge documents. Use when working with PDF files or when the user mentions PDFs, forms, or document extraction.\n\n# Excel Analysis\ndescription: Analyze Excel spreadsheets, create pivot tables, generate charts. Use when analyzing Excel files, spreadsheets, tabular data, or .xlsx files.\n\n# Git Commit Helper\ndescription: Generate descriptive commit messages by analyzing git diffs. Use when the user asks for help writing commit messages or reviewing staged changes.\n\n# Code Review\ndescription: Review code for quality, security, and best practices following team standards. Use when reviewing pull requests, code changes, or when the user asks for a code review.\n```\n\n---\n\n## Core Authoring Principles\n\n### 1. Concise is Key\n\nThe context window is shared with conversation history, other skills, and requests. Every token competes for space.\n\n**Default assumption**: The agent is already very smart. Only add context it doesn't already have.\n\nChallenge each piece of information:\n- \"Does the agent really need this explanation?\"\n- \"Can I assume the agent knows this?\"\n- \"Does this paragraph justify its token cost?\"\n\n**Good (concise)**:\n```markdown\n## Extract PDF text\n\nUse pdfplumber for text extraction:\n\n\\`\\`\\`python\nimport pdfplumber\n\nwith pdfplumber.open(\"file.pdf\") as pdf:\n    text = pdf.pages[0].extract_text()\n\\`\\`\\`\n```\n\n**Bad (verbose)**:\n```markdown\n## Extract PDF text\n\nPDF (Portable Document Format) files are a common file format that contains\ntext, images, and other content. To extract text from a PDF, you'll need to\nuse a library. There are many libraries available for PDF processing, but we\nrecommend pdfplumber because it's easy to use and handles most cases well...\n```\n\n### 2. Keep SKILL.md Under 500 Lines\n\nFor optimal performance, the main SKILL.md file should be concise. Use progressive disclosure for detailed content.\n\n### 3. Progressive Disclosure\n\nPut essential information in SKILL.md; detailed reference material in separate files that the agent reads only when needed.\n\n```markdown\n# PDF Processing\n\n## Quick start\n[Essential instructions here]\n\n## Additional resources\n- For complete API details, see [reference.md](reference.md)\n- For usage examples, see [examples.md](examples.md)\n```\n\n**Keep references one level deep** - link directly from SKILL.md to reference files. Deeply nested references may result in partial reads.\n\n### 4. Set Appropriate Degrees of Freedom\n\nMatch specificity to the task's fragility:\n\n| Freedom Level | When to Use | Example |\n|---------------|-------------|---------|\n| **High** (text instructions) | Multiple valid approaches, context-dependent | Code review guidelines |\n| **Medium** (pseudocode\u002Ftemplates) | Preferred pattern with acceptable variation | Report generation |\n| **Low** (specific scripts) | Fragile operations, consistency critical | Database migrations |\n\n---\n\n## Common Patterns\n\n### Template Pattern\n\nProvide output format templates:\n\n```markdown\n## Report structure\n\nUse this template:\n\n\\`\\`\\`markdown\n# [Analysis Title]\n\n## Executive summary\n[One-paragraph overview of key findings]\n\n## Key findings\n- Finding 1 with supporting data\n- Finding 2 with supporting data\n\n## Recommendations\n1. Specific actionable recommendation\n2. Specific actionable recommendation\n\\`\\`\\`\n```\n\n### Examples Pattern\n\nFor skills where output quality depends on seeing examples:\n\n```markdown\n## Commit message format\n\n**Example 1:**\nInput: Added user authentication with JWT tokens\nOutput:\n\\`\\`\\`\nfeat(auth): implement JWT-based authentication\n\nAdd login endpoint and token validation middleware\n\\`\\`\\`\n\n**Example 2:**\nInput: Fixed bug where dates displayed incorrectly\nOutput:\n\\`\\`\\`\nfix(reports): correct date formatting in timezone conversion\n\nUse UTC timestamps consistently across report generation\n\\`\\`\\`\n```\n\n### Workflow Pattern\n\nBreak complex operations into clear steps with checklists:\n\n```markdown\n## Form filling workflow\n\nCopy this checklist and track progress:\n\n\\`\\`\\`\nTask Progress:\n- [ ] Step 1: Analyze the form\n- [ ] Step 2: Create field mapping\n- [ ] Step 3: Validate mapping\n- [ ] Step 4: Fill the form\n- [ ] Step 5: Verify output\n\\`\\`\\`\n\n**Step 1: Analyze the form**\nRun: \\`python scripts\u002Fanalyze_form.py input.pdf\\`\n...\n```\n\n### Conditional Workflow Pattern\n\nGuide through decision points:\n\n```markdown\n## Document modification workflow\n\n1. Determine the modification type:\n\n   **Creating new content?** → Follow \"Creation workflow\" below\n   **Editing existing content?** → Follow \"Editing workflow\" below\n\n2. Creation workflow:\n   - Use docx-js library\n   - Build document from scratch\n   ...\n```\n\n### Feedback Loop Pattern\n\nFor quality-critical tasks, implement validation loops:\n\n```markdown\n## Document editing process\n\n1. Make your edits\n2. **Validate immediately**: \\`python scripts\u002Fvalidate.py output\u002F\\`\n3. If validation fails:\n   - Review the error message\n   - Fix the issues\n   - Run validation again\n4. **Only proceed when validation passes**\n```\n\n---\n\n## Utility Scripts\n\nPre-made scripts offer advantages over generated code:\n- More reliable than generated code\n- Save tokens (no code in context)\n- Save time (no code generation)\n- Ensure consistency across uses\n\n```markdown\n## Utility scripts\n\n**analyze_form.py**: Extract all form fields from PDF\n\\`\\`\\`bash\npython scripts\u002Fanalyze_form.py input.pdf > fields.json\n\\`\\`\\`\n\n**validate.py**: Check for errors\n\\`\\`\\`bash\npython scripts\u002Fvalidate.py fields.json\n# Returns: \"OK\" or lists conflicts\n\\`\\`\\`\n```\n\nMake clear whether the agent should **execute** the script (most common) or **read** it as reference.\n\n---\n\n## Anti-Patterns to Avoid\n\n### 1. Windows-Style Paths\n- ✅ Use: `scripts\u002Fhelper.py`\n- ❌ Avoid: `scripts\\helper.py`\n\n### 2. Too Many Options\n```markdown\n# Bad - confusing\n\"You can use pypdf, or pdfplumber, or PyMuPDF, or...\"\n\n# Good - provide a default with escape hatch\n\"Use pdfplumber for text extraction.\nFor scanned PDFs requiring OCR, use pdf2image with pytesseract instead.\"\n```\n\n### 3. Time-Sensitive Information\n```markdown\n# Bad - will become outdated\n\"If you're doing this before August 2025, use the old API.\"\n\n# Good - use an \"old patterns\" section\n## Current method\nUse the v2 API endpoint.\n\n## Old patterns (deprecated)\n\u003Cdetails>\n\u003Csummary>Legacy v1 API\u003C\u002Fsummary>\n...\n\u003C\u002Fdetails>\n```\n\n### 4. Inconsistent Terminology\nChoose one term and use it throughout:\n- ✅ Always \"API endpoint\" (not mixing \"URL\", \"route\", \"path\")\n- ✅ Always \"field\" (not mixing \"box\", \"element\", \"control\")\n\n### 5. Vague Skill Names\n- ✅ Good: `processing-pdfs`, `analyzing-spreadsheets`\n- ❌ Avoid: `helper`, `utils`, `tools`\n\n---\n\n## Skill Creation Workflow\n\nWhen helping a user create a skill, follow this process:\n\n### Phase 1: Discovery\n\nGather information about:\n1. The skill's purpose and primary use case\n2. Storage location (personal vs project)\n3. Trigger scenarios\n4. Any specific requirements or constraints\n5. Existing examples or patterns to follow\n\nIf you have access to the AskQuestion tool, use it for efficient structured gathering. Otherwise, ask conversationally.\n\n### Phase 2: Design\n\n1. Draft the skill name (lowercase, hyphens, max 64 chars)\n2. Write a specific, third-person description\n3. Outline the main sections needed\n4. Identify if supporting files or scripts are needed\n\n### Phase 3: Implementation\n\n1. Create the directory structure\n2. Write the SKILL.md file with frontmatter\n3. Create any supporting reference files\n4. Create any utility scripts if needed\n\n### Phase 4: Verification\n\n1. Verify the SKILL.md is under 500 lines\n2. Check that the description is specific and includes trigger terms\n3. Ensure consistent terminology throughout\n4. Verify all file references are one level deep\n5. Test that the skill can be discovered and applied\n\n---\n\n## Complete Example\n\nHere's a complete example of a well-structured skill:\n\n**Directory structure:**\n```\ncode-review\u002F\n├── SKILL.md\n├── STANDARDS.md\n└── examples.md\n```\n\n**SKILL.md:**\n```markdown\n---\nname: code-review\ndescription: Review code for quality, security, and maintainability following team standards. Use when reviewing pull requests, examining code changes, or when the user asks for a code review.\n---\n\n# Code Review\n\n## Quick Start\n\nWhen reviewing code:\n\n1. Check for correctness and potential bugs\n2. Verify security best practices\n3. Assess code readability and maintainability\n4. Ensure tests are adequate\n\n## Review Checklist\n\n- [ ] Logic is correct and handles edge cases\n- [ ] No security vulnerabilities (SQL injection, XSS, etc.)\n- [ ] Code follows project style conventions\n- [ ] Functions are appropriately sized and focused\n- [ ] Error handling is comprehensive\n- [ ] Tests cover the changes\n\n## Providing Feedback\n\nFormat feedback as:\n- 🔴 **Critical**: Must fix before merge\n- 🟡 **Suggestion**: Consider improving\n- 🟢 **Nice to have**: Optional enhancement\n\n## Additional Resources\n\n- For detailed coding standards, see [STANDARDS.md](STANDARDS.md)\n- For example reviews, see [examples.md](examples.md)\n```\n\n---\n\n## Summary Checklist\n\nBefore finalizing a skill, verify:\n\n### Core Quality\n- [ ] Description is specific and includes key terms\n- [ ] Description includes both WHAT and WHEN\n- [ ] Written in third person\n- [ ] SKILL.md body is under 500 lines\n- [ ] Consistent terminology throughout\n- [ ] Examples are concrete, not abstract\n\n### Structure\n- [ ] File references are one level deep\n- [ ] Progressive disclosure used appropriately\n- [ ] Workflows have clear steps\n- [ ] No time-sensitive information\n\n### If Including Scripts\n- [ ] Scripts solve problems rather than punt\n- [ ] Required packages are documented\n- [ ] Error handling is explicit and helpful\n- [ ] No Windows-style paths\n","# Creating Skills in Cursor\n\nThis skill guides you through creating effective Agent Skills for Cursor. Skills are markdown files that teach the agent how to perform specific tasks: reviewing PRs using team standards, generating commit messages in a preferred format, querying database schemas, or any specialized workflow.\n\n## Before You Begin: Gather Requirements\n\nBefore creating a skill, gather essential information from the user about:\n\n1. **Purpose and scope**: What specific task or workflow should this skill help with?\n2. **Target location**: Should this be a personal skill (~\u002F.cursor\u002Fskills\u002F) or project skill (.cursor\u002Fskills\u002F)?\n3. **Trigger scenarios**: When should the agent automatically apply this skill?\n4. **Key domain knowledge**: What specialized information does the agent need that it wouldn't already know?\n5. **Output format preferences**: Are there specific templates, formats, or styles required?\n6. **Existing patterns**: Are there existing examples or conventions to follow?\n\n### Inferring from Context\n\nIf you have previous conversation context, infer the skill from what was discussed. You can create skills based on workflows, patterns, or domain knowledge that emerged in the conversation.\n\n### Gathering Additional Information\n\nIf you need clarification, use the AskQuestion tool when available:\n\n```\nExample AskQuestion usage:\n- \"Where should this skill be stored?\" with options like [\"Personal (~\u002F.cursor\u002Fskills\u002F)\", \"Project (.cursor\u002Fskills\u002F)\"]\n- \"Should this skill include executable scripts?\" with options like [\"Yes\", \"No\"]\n```\n\nIf the AskQuestion tool is not available, ask these questions conversationally.\n\n---\n\n## Skill File Structure\n\n### Directory Layout\n\nSkills are stored as directories containing a `SKILL.md` file:\n\n```\nskill-name\u002F\n├── SKILL.md              # Required - main instructions\n├── reference.md          # Optional - detailed documentation\n├── examples.md           # Optional - usage examples\n└── scripts\u002F              # Optional - utility scripts\n    ├── validate.py\n    └── helper.sh\n```\n\n### Storage Locations\n\n| Type | Path | Scope |\n|------|------|-------|\n| Personal | ~\u002F.cursor\u002Fskills\u002Fskill-name\u002F | Available across all your projects |\n| Project | .cursor\u002Fskills\u002Fskill-name\u002F | Shared with anyone using the repository |\n\n**IMPORTANT**: Never create skills in `~\u002F.cursor\u002Fskills-cursor\u002F`. This directory is reserved for Cursor's internal built-in skills and is managed automatically by the system.\n\n### SKILL.md Structure\n\nEvery skill requires a `SKILL.md` file with YAML frontmatter and markdown body:\n\n```markdown\n---\nname: your-skill-name\ndescription: Brief description of what this skill does and when to use it\n---\n\n# Your Skill Name\n\n## Instructions\nClear, step-by-step guidance for the agent.\n\n## Examples\nConcrete examples of using this skill.\n```\n\n### Required Metadata Fields\n\n| Field | Requirements | Purpose |\n|-------|--------------|---------|\n| `name` | Max 64 chars, lowercase letters\u002Fnumbers\u002Fhyphens only | Unique identifier for the skill |\n| `description` | Max 1024 chars, non-empty | Helps agent decide when to apply the skill |\n\n---\n\n## Writing Effective Descriptions\n\nThe description is **critical** for skill discovery. The agent uses it to decide when to apply your skill.\n\n### Description Best Practices\n\n1. **Write in third person** (the description is injected into the system prompt):\n   - ✅ Good: \"Processes Excel files and generates reports\"\n   - ❌ Avoid: \"I can help you process Excel files\"\n   - ❌ Avoid: \"You can use this to process Excel files\"\n\n2. **Be specific and include trigger terms**:\n   - ✅ Good: \"Extract text and tables from PDF files, fill forms, merge documents. Use when working with PDF files or when the user mentions PDFs, forms, or document extraction.\"\n   - ❌ Vague: \"Helps with documents\"\n\n3. **Include both WHAT and WHEN**:\n   - WHAT: What the skill does (specific capabilities)\n   - WHEN: When the agent should use it (trigger scenarios)\n\n### Description Examples\n\n```yaml\n# PDF Processing\ndescription: Extract text and tables from PDF files, fill forms, merge documents. Use when working with PDF files or when the user mentions PDFs, forms, or document extraction.\n\n# Excel Analysis\ndescription: Analyze Excel spreadsheets, create pivot tables, generate charts. Use when analyzing Excel files, spreadsheets, tabular data, or .xlsx files.\n\n# Git Commit Helper\ndescription: Generate descriptive commit messages by analyzing git diffs. Use when the user asks for help writing commit messages or reviewing staged changes.\n\n# Code Review\ndescription: Review code for quality, security, and best practices following team standards. Use when reviewing pull requests, code changes, or when the user asks for a code review.\n```\n\n---\n\n## Core Authoring Principles\n\n### 1. Concise is Key\n\nThe context window is shared with conversation history, other skills, and requests. Every token competes for space.\n\n**Default assumption**: The agent is already very smart. Only add context it doesn't already have.\n\nChallenge each piece of information:\n- \"Does the agent really need this explanation?\"\n- \"Can I assume the agent knows this?\"\n- \"Does this paragraph justify its token cost?\"\n\n**Good (concise)**:\n```markdown\n## Extract PDF text\n\nUse pdfplumber for text extraction:\n\n\\`\\`\\`python\nimport pdfplumber\n\nwith pdfplumber.open(\"file.pdf\") as pdf:\n    text = pdf.pages[0].extract_text()\n\\`\\`\\`\n```\n\n**Bad (verbose)**:\n```markdown\n## Extract PDF text\n\nPDF (Portable Document Format) files are a common file format that contains\ntext, images, and other content. To extract text from a PDF, you'll need to\nuse a library. There are many libraries available for PDF processing, but we\nrecommend pdfplumber because it's easy to use and handles most cases well...\n```\n\n### 2. Keep SKILL.md Under 500 Lines\n\nFor optimal performance, the main SKILL.md file should be concise. Use progressive disclosure for detailed content.\n\n### 3. Progressive Disclosure\n\nPut essential information in SKILL.md; detailed reference material in separate files that the agent reads only when needed.\n\n```markdown\n# PDF Processing\n\n## Quick start\n[Essential instructions here]\n\n## Additional resources\n- For complete API details, see [reference.md](reference.md)\n- For usage examples, see [examples.md](examples.md)\n```\n\n**Keep references one level deep** - link directly from SKILL.md to reference files. Deeply nested references may result in partial reads.\n\n### 4. Set Appropriate Degrees of Freedom\n\nMatch specificity to the task's fragility:\n\n| Freedom Level | When to Use | Example |\n|---------------|-------------|---------|\n| **High** (text instructions) | Multiple valid approaches, context-dependent | Code review guidelines |\n| **Medium** (pseudocode\u002Ftemplates) | Preferred pattern with acceptable variation | Report generation |\n| **Low** (specific scripts) | Fragile operations, consistency critical | Database migrations |\n\n---\n\n## Common Patterns\n\n### Template Pattern\n\nProvide output format templates:\n\n```markdown\n## Report structure\n\nUse this template:\n\n\\`\\`\\`markdown\n# [Analysis Title]\n\n## Executive summary\n[One-paragraph overview of key findings]\n\n## Key findings\n- Finding 1 with supporting data\n- Finding 2 with supporting data\n\n## Recommendations\n1. Specific actionable recommendation\n2. Specific actionable recommendation\n\\`\\`\\`\n```\n\n### Examples Pattern\n\nFor skills where output quality depends on seeing examples:\n\n```markdown\n## Commit message format\n\n**Example 1:**\nInput: Added user authentication with JWT tokens\nOutput:\n\\`\\`\\`\nfeat(auth): implement JWT-based authentication\n\nAdd login endpoint and token validation middleware\n\\`\\`\\`\n\n**Example 2:**\nInput: Fixed bug where dates displayed incorrectly\nOutput:\n\\`\\`\\`\nfix(reports): correct date formatting in timezone conversion\n\nUse UTC timestamps consistently across report generation\n\\`\\`\\`\n```\n\n### Workflow Pattern\n\nBreak complex operations into clear steps with checklists:\n\n```markdown\n## Form filling workflow\n\nCopy this checklist and track progress:\n\n\\`\\`\\`\nTask Progress:\n- [ ] Step 1: Analyze the form\n- [ ] Step 2: Create field mapping\n- [ ] Step 3: Validate mapping\n- [ ] Step 4: Fill the form\n- [ ] Step 5: Verify output\n\\`\\`\\`\n\n**Step 1: Analyze the form**\nRun: \\`python scripts\u002Fanalyze_form.py input.pdf\\`\n...\n```\n\n### Conditional Workflow Pattern\n\nGuide through decision points:\n\n```markdown\n## Document modification workflow\n\n1. Determine the modification type:\n\n   **Creating new content?** → Follow \"Creation workflow\" below\n   **Editing existing content?** → Follow \"Editing workflow\" below\n\n2. Creation workflow:\n   - Use docx-js library\n   - Build document from scratch\n   ...\n```\n\n### Feedback Loop Pattern\n\nFor quality-critical tasks, implement validation loops:\n\n```markdown\n## Document editing process\n\n1. Make your edits\n2. **Validate immediately**: \\`python scripts\u002Fvalidate.py output\u002F\\`\n3. If validation fails:\n   - Review the error message\n   - Fix the issues\n   - Run validation again\n4. **Only proceed when validation passes**\n```\n\n---\n\n## Utility Scripts\n\nPre-made scripts offer advantages over generated code:\n- More reliable than generated code\n- Save tokens (no code in context)\n- Save time (no code generation)\n- Ensure consistency across uses\n\n```markdown\n## Utility scripts\n\n**analyze_form.py**: Extract all form fields from PDF\n\\`\\`\\`bash\npython scripts\u002Fanalyze_form.py input.pdf > fields.json\n\\`\\`\\`\n\n**validate.py**: Check for errors\n\\`\\`\\`bash\npython scripts\u002Fvalidate.py fields.json\n# Returns: \"OK\" or lists conflicts\n\\`\\`\\`\n```\n\nMake clear whether the agent should **execute** the script (most common) or **read** it as reference.\n\n---\n\n## Anti-Patterns to Avoid\n\n### 1. Windows-Style Paths\n- ✅ Use: `scripts\u002Fhelper.py`\n- ❌ Avoid: `scripts\\helper.py`\n\n### 2. Too Many Options\n```markdown\n# Bad - confusing\n\"You can use pypdf, or pdfplumber, or PyMuPDF, or...\"\n\n# Good - provide a default with escape hatch\n\"Use pdfplumber for text extraction.\nFor scanned PDFs requiring OCR, use pdf2image with pytesseract instead.\"\n```\n\n### 3. Time-Sensitive Information\n```markdown\n# Bad - will become outdated\n\"If you're doing this before August 2025, use the old API.\"\n\n# Good - use an \"old patterns\" section\n## Current method\nUse the v2 API endpoint.\n\n## Old patterns (deprecated)\n\u003Cdetails>\n\u003Csummary>Legacy v1 API\u003C\u002Fsummary>\n...\n\u003C\u002Fdetails>\n```\n\n### 4. Inconsistent Terminology\nChoose one term and use it throughout:\n- ✅ Always \"API endpoint\" (not mixing \"URL\", \"route\", \"path\")\n- ✅ Always \"field\" (not mixing \"box\", \"element\", \"control\")\n\n### 5. Vague Skill Names\n- ✅ Good: `processing-pdfs`, `analyzing-spreadsheets`\n- ❌ Avoid: `helper`, `utils`, `tools`\n\n---\n\n## Skill Creation Workflow\n\nWhen helping a user create a skill, follow this process:\n\n### Phase 1: Discovery\n\nGather information about:\n1. The skill's purpose and primary use case\n2. Storage location (personal vs project)\n3. Trigger scenarios\n4. Any specific requirements or constraints\n5. Existing examples or patterns to follow\n\nIf you have access to the AskQuestion tool, use it for efficient structured gathering. Otherwise, ask conversationally.\n\n### Phase 2: Design\n\n1. Draft the skill name (lowercase, hyphens, max 64 chars)\n2. Write a specific, third-person description\n3. Outline the main sections needed\n4. Identify if supporting files or scripts are needed\n\n### Phase 3: Implementation\n\n1. Create the directory structure\n2. Write the SKILL.md file with frontmatter\n3. Create any supporting reference files\n4. Create any utility scripts if needed\n\n### Phase 4: Verification\n\n1. Verify the SKILL.md is under 500 lines\n2. Check that the description is specific and includes trigger terms\n3. Ensure consistent terminology throughout\n4. Verify all file references are one level deep\n5. Test that the skill can be discovered and applied\n\n---\n\n## Complete Example\n\nHere's a complete example of a well-structured skill:\n\n**Directory structure:**\n```\ncode-review\u002F\n├── SKILL.md\n├── STANDARDS.md\n└── examples.md\n```\n\n**SKILL.md:**\n```markdown\n---\nname: code-review\ndescription: Review code for quality, security, and maintainability following team standards. Use when reviewing pull requests, examining code changes, or when the user asks for a code review.\n---\n\n# Code Review\n\n## Quick Start\n\nWhen reviewing code:\n\n1. Check for correctness and potential bugs\n2. Verify security best practices\n3. Assess code readability and maintainability\n4. Ensure tests are adequate\n\n## Review Checklist\n\n- [ ] Logic is correct and handles edge cases\n- [ ] No security vulnerabilities (SQL injection, XSS, etc.)\n- [ ] Code follows project style conventions\n- [ ] Functions are appropriately sized and focused\n- [ ] Error handling is comprehensive\n- [ ] Tests cover the changes\n\n## Providing Feedback\n\nFormat feedback as:\n- 🔴 **Critical**: Must fix before merge\n- 🟡 **Suggestion**: Consider improving\n- 🟢 **Nice to have**: Optional enhancement\n\n## Additional Resources\n\n- For detailed coding standards, see [STANDARDS.md](STANDARDS.md)\n- For example reviews, see [examples.md](examples.md)\n```\n\n---\n\n## Summary Checklist\n\nBefore finalizing a skill, verify:\n\n### Core Quality\n- [ ] Description is specific and includes key terms\n- [ ] Description includes both WHAT and WHEN\n- [ ] Written in third person\n- [ ] SKILL.md body is under 500 lines\n- [ ] Consistent terminology throughout\n- [ ] Examples are concrete, not abstract\n\n### Structure\n- [ ] File references are one level deep\n- [ ] Progressive disclosure used appropriately\n- [ ] Workflows have clear steps\n- [ ] No time-sensitive information\n\n### If Including Scripts\n- [ ] Scripts solve problems rather than punt\n- [ ] Required packages are documented\n- [ ] Error handling is explicit and helpful\n- [ ] No Windows-style paths\n",{"name":5,"description":7},"main","6ef866230c1ba3459fd08fc65141dfd37fffd41a",0,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills","https:\u002F\u002Fskills.henryzh.dev","https:\u002F\u002Fgithub.com\u002Fcaezium","henryzh.dev","quant, data science, full-stack, robotics.\r\nbuilding cool stuff","2026-08-24 19:21:44",{"source":8,"skillName":5,"commands":57,"prompt":67,"zipAvailable":68},[58,61,64],{"tool":59,"command":60},"npx","npx skills add caezium\u002Fskills --skill create-skill",{"tool":62,"command":63},"bunx","bunx skills add caezium\u002Fskills --skill create-skill",{"tool":65,"command":66},"pnpm","pnpm dlx skills add caezium\u002Fskills --skill create-skill","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fcreate-skill\u002FSKILL.md",true,[],[71,90,117,134,155,177],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":85,"publishedAt":42},"01M71JT4RYCAS18RF6003YYR1D","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwriting-for-agents\u002FSKILL.md",[77,80],{"categoryNo":78,"categoryName":79,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":81,"categoryName":82,"taxonomyType":83,"taxonomyTypeDesc":84},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[86,33,87,88,32,89,40],"writing for agents","writing","editing","documents",{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":99,"publishedAt":42},"01M71JT0GES4AZ5ADH8EADNN1Z","wrangler","Cloudflare Workers CLI for deploying, developing, and managing Workers, KV, R2, D1, Vectorize, Hyperdrive, Workers AI, Containers, Queues, Workflows, Pipelines, and Secrets Store. Load before running wrangler commands to ensure correct syntax and best practices. Biases towards retrieval from Cloudflare docs over pre-trained knowledge.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwrangler\u002FSKILL.md",[96],{"categoryNo":97,"categoryName":98,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",[92,100,101,102,103,104,105,106,107,108,109,110,111,33,112,113,114,115,31,116],"cloudflare","d1","knowledge","workflows","cli","secrets","commands","syntax","cloudflare workers","r2","queues","running","containers","correct","load","workers","store",{"skillNo":118,"name":119,"nameZh":6,"description":120,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":121,"categories":122,"keywords":123,"publishedAt":42},"01M71JSWX4R203M32K9A18AAW8","workers-best-practices","Reviews and authors Cloudflare Workers code against production best practices. Load when writing new Workers, reviewing Worker code, configuring wrangler.jsonc, or checking for common Workers anti-patterns (streaming, floating promises, global state, secrets, bindings, observability). Biases towards retrieval from Cloudflare docs over pre-trained knowledge.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fworkers-best-practices\u002FSKILL.md",[],[124,115,31,125,126,127,33,100,128,102,41,92,87,114,40,129,130,131,132,133],"workers best practices","observability","reviews","promises","worker","code","common","streaming","state","anti-patterns",{"skillNo":135,"name":136,"nameZh":6,"description":137,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":138,"categories":139,"keywords":140,"publishedAt":42},"01M71JST6P88MXGT2MWDTPJGEG","wizard","Generate an interactive bash wizard that walks a human through steps only they can perform. Use when provisioning infrastructure, setting up credentials or CI secrets, walking an unfamiliar third-party dashboard, or running a one-off migration or cutover. Don't invoke this for steps the agent can perform itself.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwizard\u002FSKILL.md",[],[136,111,141,33,142,143,144,145,32,146,40,147,148,149,150,151,152,153,154,105],"setting","up","ci","provisioning infrastructure","only","invoke","Bash","one","interactive","can","credentials","dashboard","generate","human",{"skillNo":156,"name":157,"nameZh":6,"description":158,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":159,"sourceHtmlUrl":160,"categories":161,"keywords":162,"publishedAt":42},"01M71JSQ95S7BQ61DAPN95X2W4","why","Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Discovers available evidence sources, searches each relevant category, and returns a cited read on decisions and tradeoffs. Use how for runtime behavior.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwhy\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwhy\u002FSKILL.md",[],[157,163,164,165,166,33,167,168,32,169,170,171,172,173,174,175,176],"sources","thresholds","evidence","relevant","runtime","design","decisions","work","how","behavior","read","data","rationale","returns",{"skillNo":178,"name":179,"nameZh":6,"description":180,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":181,"categories":182,"keywords":184,"publishedAt":42},"01M71JSMHFVSH8HK8GXN6YAX6J","week-prep","Generate a briefing with calendar, tasks, and attention flags. Use for \"week prep\", \"briefing\", or \"what's on this week\".","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fweek-prep\u002FSKILL.md",[183],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[185,186,187,188,189,33,190,191,192,193,32,147,194,195,196,153,197,173,198,199],"week prep","prep","deep","attention","get","search","messages","briefing","MCP","tasks","with","slack","calendar","flags","grep"]