[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6KG82GAEY3GG77CMGCK1C2T":3,"skill-same:01M6KG82GAEY3GG77CMGCK1C2T":74},{"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":41,"linkOnly":42,"markdownContent":43,"markdownBody":44,"frontMatter":45,"refName":46,"commitSha":47,"skillDirectory":48,"forksCount":14,"repositoryHtmlUrl":49,"repositoryHomepageUrl":6,"ownerHtmlUrl":50,"ownerBlogUrl":51,"ownerDescription":52,"updatedAt":53,"install":54,"resources":67},"01M6KG82GAEY3GG77CMGCK1C2T","fallback-doc-with-delegation",null,"Generate professional documents after tool failures using shell_agent for complex formats or write_file for simple text\u002Fmarkdown","xyliu-cs\u002FOpenSpace-GDPval","xyliu-cs","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F89236536?v=4","01M6KF8Z1V4TXS86CKSH48BY8T","xy","01M6KF8Z67443W0SHXM85JEF0B",0,"MIT","https:\u002F\u002Fgithub.com\u002Fxyliu-cs\u002FOpenSpace-GDPval\u002Fblob\u002Fmain\u002Fgdpval_bench\u002Fskills\u002Fwrite-file-fallback-report-enhanced\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],"fallback doc with delegation","delegation","doc","with","fallback","text","for","formats","professional","simple","write","documents","tool","markdown","shell","file","generate","2026-11-10 10:12:30",false,"---\nname: fallback-doc-with-delegation\ndescription: Generate professional documents after tool failures using shell_agent for complex formats or write_file for simple text\u002Fmarkdown\n---\n\n# Fallback Document Generation with Delegation\n\n## When to Use This Skill\n\nUse this workflow when attempting to generate a document or report, but **multiple primary data source tools fail simultaneously**:\n\n- `read_file` returns binary\u002Fimage data instead of text (common with PDFs)\n- `search_web` returns errors or no results\n- `execute_code_sandbox` fails unexpectedly\n- Other data retrieval tools are unavailable\n\n**Key insight**: Rather than getting stuck on failed data retrieval, pivot immediately to generating the document using the most appropriate available method—either `shell_agent` for complex formats or `write_file` for simple text.\n\n## Step-by-Step Instructions\n\n### Step 1: Detect Tool Failure Pattern\n\nRecognize when you're in a fallback scenario:\n\n```\nTOOL_FAILURE_INDICATORS = [\n    \"read_file returns binary or image data\",\n    \"search_web returns unknown error or empty results\",\n    \"execute_code_sandbox fails repeatedly\",\n    \"Multiple consecutive tool failures on data retrieval\"\n]\n```\n\n**Decision point**: If 2+ indicators are present, proceed to Step 2.\n\n### Step 2: Choose the Appropriate Fallback Method\n\n**Critical Decision**: Select between `shell_agent` and `write_file` based on document complexity:\n\n| Use `shell_agent` when... | Use `write_file` when... |\n|---------------------------|--------------------------|\n| Generating PDFs or formatted documents | Creating simple markdown\u002Ftext files |\n| Complex layout or styling needed | Plain structured content is sufficient |\n| External libraries required (reportlab, pandoc) | No special formatting libraries needed |\n| Multi-step generation process | Single-pass content generation |\n| Verification\u002Fvalidation needed | Direct file creation is adequate |\n\n**Decision Tree**:\n```\nIs the output a PDF or complex formatted document?\n├── YES → Use shell_agent (delegates to skilled agent)\n└── NO → Is it markdown or plain text?\n    ├── YES → Use write_file (direct content generation)\n    └── NO → Use shell_agent (handles uncertainty)\n```\n\n### Step 3A: Execute with shell_agent (Complex Documents)\n\nFor PDFs and complex formatting:\n\n```python\n# Delegate to shell_agent with clear task description\nshell_agent(\n    task=\"Generate a professional [document type] about [topic]. \"\n         \"Use appropriate libraries (e.g., reportlab for PDFs). \"\n         \"Include: [list key sections\u002Frequirements]. \"\n         \"Save to working directory as [filename].\"\n)\n```\n\n**Advantages of shell_agent**:\n- Agent autonomously handles library selection and usage\n- Automatically retries and fixes errors\n- Can use multiple tools internally (run_shell, execute_code_sandbox)\n- Validates output before completion\n\n**Example**:\n```python\nshell_agent(\n    task=\"Create a 1-page SBAR (Situation-Background-Assessment-Recommendation) \"\n         \"template PDF with 4 sections. Each section should have guiding points \"\n         \"and blank lined spaces for user input. Use reportlab library. \"\n         \"Save as SBAR_template.pdf in current directory.\"\n)\n```\n\n### Step 3B: Execute with write_file (Simple Documents)\n\nFor markdown and plain text:\n\n```python\n# Generate content with embedded domain knowledge\nreport_content = f\"\"\"# [Document Title]\n\n## Executive Summary\n[Brief overview of key findings\u002Fcontent]\n\n## Background\n[Context and scope - use embedded knowledge]\n\n## Main Content\n[Organized sections with headers, lists, tables as appropriate]\n\n## Limitations & Notes\n[Transparent about data source limitations if relevant]\n\n## Recommendations\u002FNext Steps\n[Actionable guidance based on available information]\n\"\"\"\n\nwrite_file(path=\"output\u002Freport.md\", content=report_content)\n```\n\n### Step 4: Structure Content Professionally\n\nRegardless of method, ensure professional organization:\n\n```markdown\n# [Document Title]\n\n## Executive Summary\n[Brief overview of key findings\u002Fcontent]\n\n## Background\n[Context and scope - use embedded knowledge]\n\n## Main Content\n[Organized sections with headers, lists, tables as appropriate]\n\n## Limitations & Notes\n[Transparent about data source limitations if relevant]\n\n## Recommendations\u002FNext Steps\n[Actionable guidance based on available information]\n```\n\n### Step 5: Leverage Embedded Domain Knowledge\n\nWhen external data is unavailable:\n\n- Use **general domain knowledge** appropriately\n- Clearly distinguish between verified facts and general guidance\n- Include **actionable frameworks** rather than specific unverified data\n- Add **placeholder notes** where specific data would enhance the document\n\nExample:\n```markdown\n> **Note**: Specific [metric\u002Fdata point] would typically be sourced from \n> [expected source]. The guidance below reflects established best practices \n> in this domain.\n```\n\n### Step 6: Validate Output\n\n```python\n# For shell_agent: verify the output exists and is valid\nlist_dir(path=\".\")  # Confirm file was created\n\n# For PDFs, optionally verify with pdfinfo via run_shell\n# For markdown, optionally read to confirm content\n\n# For write_file: confirm file exists\nlist_dir(path=\".\")  # Confirm file exists\n```\n\n## Code Example\n\n```python\n# Complete detection and pivot pattern\ndef generate_fallback_document(task_goal, output_format=\"markdown\"):\n    # After detecting tool failures:\n    \n    # Step 1: Choose method based on format\n    if output_format == \"pdf\" or \"complex\" in task_goal:\n        # Use shell_agent for complex formats\n        shell_agent(\n            task=f\"Generate a professional {output_format} document about \"\n                 f\"{task_goal}. Use appropriate libraries. Include executive \"\n                 f\"summary, main content, limitations, and recommendations. \"\n                 f\"Save to working directory.\"\n        )\n        return \"Document generated via shell_agent delegation\"\n    \n    else:\n        # Use write_file for simple formats\n        report_content = f\"\"\"# {task_goal} Report\n\n## Executive Summary\nThis report was generated using established domain knowledge due to \ntemporary unavailability of primary data sources.\n\n## Key Frameworks and Guidance\n[Structured content with headers, bullets, tables]\n\n## Limitations\n- Specific data points from [expected sources] were unavailable\n- Recommendations based on general best practices\n\n## Action Items\n1. [Concrete step 1]\n2. [Concrete step 2]\n\"\"\"\n        write_file(path=\"generated_report.md\", content=report_content)\n        return \"Report generated with write_file\"\n```\n\n## Best Practices\n\n| Do | Don't |\n|----|-------|\n| Choose method based on document complexity | Default to write_file for PDFs without consideration |\n| Delegate complex formatting to shell_agent | Keep retrying failing tools 5+ times |\n| Be transparent about limitations | Claim unverified specifics as facts |\n| Provide actionable frameworks | Leave the task incomplete |\n| Use professional document structure | Output unstructured text walls |\n| Validate the generated output | Assume success without verification |\n\n## Common Pitfalls\n\n1. **Wrong tool selection**: Using write_file for PDF generation instead of delegating to shell_agent\n2. **Over-apologizing**: Acknowledge limitations once, then deliver value\n3. **Under-delivering**: A well-structured partial report beats no report\n4. **Misrepresenting certainty**: Use appropriate hedging language\n5. **Skipping validation**: Always confirm the file was created successfully\n\n## Success Criteria\n\nThe skill is successfully applied when:\n\n- [ ] Document is generated despite tool failures\n- [ ] Appropriate method chosen (shell_agent for complex, write_file for simple)\n- [ ] Content is professionally structured (headers, sections, lists)\n- [ ] Limitations are transparently noted\n- [ ] Actionable guidance is provided\n- [ ] File is successfully created and verified\n\n## Method Comparison\n\n| Aspect | shell_agent | write_file |\n|--------|-------------|------------|\n| Best for | PDFs, complex layouts, multi-step generation | Markdown, plain text, single-pass content |\n| Error handling | Autonomous retry and fix | Manual error handling required |\n| Library access | Full library ecosystem | Content-only (no external libs) |\n| Processing time | Longer (agent reasoning) | Faster (direct write) |\n| Verification | Built-in validation | Manual verification needed |\n| Cost | Higher (agent execution) | Lower (simple operation) |\n\n## Decision Checklist\n\nBefore selecting method, ask:\n\n1. Is the output format PDF or other binary format? → **shell_agent**\n2. Does it require external libraries (reportlab, pandoc, etc.)? → **shell_agent**\n3. Is it markdown or plain text only? → **write_file**\n4. Does it need multi-step processing or validation? → **shell_agent**\n5. Is quick, simple content generation sufficient? → **write_file**\n","# Fallback Document Generation with Delegation\n\n## When to Use This Skill\n\nUse this workflow when attempting to generate a document or report, but **multiple primary data source tools fail simultaneously**:\n\n- `read_file` returns binary\u002Fimage data instead of text (common with PDFs)\n- `search_web` returns errors or no results\n- `execute_code_sandbox` fails unexpectedly\n- Other data retrieval tools are unavailable\n\n**Key insight**: Rather than getting stuck on failed data retrieval, pivot immediately to generating the document using the most appropriate available method—either `shell_agent` for complex formats or `write_file` for simple text.\n\n## Step-by-Step Instructions\n\n### Step 1: Detect Tool Failure Pattern\n\nRecognize when you're in a fallback scenario:\n\n```\nTOOL_FAILURE_INDICATORS = [\n    \"read_file returns binary or image data\",\n    \"search_web returns unknown error or empty results\",\n    \"execute_code_sandbox fails repeatedly\",\n    \"Multiple consecutive tool failures on data retrieval\"\n]\n```\n\n**Decision point**: If 2+ indicators are present, proceed to Step 2.\n\n### Step 2: Choose the Appropriate Fallback Method\n\n**Critical Decision**: Select between `shell_agent` and `write_file` based on document complexity:\n\n| Use `shell_agent` when... | Use `write_file` when... |\n|---------------------------|--------------------------|\n| Generating PDFs or formatted documents | Creating simple markdown\u002Ftext files |\n| Complex layout or styling needed | Plain structured content is sufficient |\n| External libraries required (reportlab, pandoc) | No special formatting libraries needed |\n| Multi-step generation process | Single-pass content generation |\n| Verification\u002Fvalidation needed | Direct file creation is adequate |\n\n**Decision Tree**:\n```\nIs the output a PDF or complex formatted document?\n├── YES → Use shell_agent (delegates to skilled agent)\n└── NO → Is it markdown or plain text?\n    ├── YES → Use write_file (direct content generation)\n    └── NO → Use shell_agent (handles uncertainty)\n```\n\n### Step 3A: Execute with shell_agent (Complex Documents)\n\nFor PDFs and complex formatting:\n\n```python\n# Delegate to shell_agent with clear task description\nshell_agent(\n    task=\"Generate a professional [document type] about [topic]. \"\n         \"Use appropriate libraries (e.g., reportlab for PDFs). \"\n         \"Include: [list key sections\u002Frequirements]. \"\n         \"Save to working directory as [filename].\"\n)\n```\n\n**Advantages of shell_agent**:\n- Agent autonomously handles library selection and usage\n- Automatically retries and fixes errors\n- Can use multiple tools internally (run_shell, execute_code_sandbox)\n- Validates output before completion\n\n**Example**:\n```python\nshell_agent(\n    task=\"Create a 1-page SBAR (Situation-Background-Assessment-Recommendation) \"\n         \"template PDF with 4 sections. Each section should have guiding points \"\n         \"and blank lined spaces for user input. Use reportlab library. \"\n         \"Save as SBAR_template.pdf in current directory.\"\n)\n```\n\n### Step 3B: Execute with write_file (Simple Documents)\n\nFor markdown and plain text:\n\n```python\n# Generate content with embedded domain knowledge\nreport_content = f\"\"\"# [Document Title]\n\n## Executive Summary\n[Brief overview of key findings\u002Fcontent]\n\n## Background\n[Context and scope - use embedded knowledge]\n\n## Main Content\n[Organized sections with headers, lists, tables as appropriate]\n\n## Limitations & Notes\n[Transparent about data source limitations if relevant]\n\n## Recommendations\u002FNext Steps\n[Actionable guidance based on available information]\n\"\"\"\n\nwrite_file(path=\"output\u002Freport.md\", content=report_content)\n```\n\n### Step 4: Structure Content Professionally\n\nRegardless of method, ensure professional organization:\n\n```markdown\n# [Document Title]\n\n## Executive Summary\n[Brief overview of key findings\u002Fcontent]\n\n## Background\n[Context and scope - use embedded knowledge]\n\n## Main Content\n[Organized sections with headers, lists, tables as appropriate]\n\n## Limitations & Notes\n[Transparent about data source limitations if relevant]\n\n## Recommendations\u002FNext Steps\n[Actionable guidance based on available information]\n```\n\n### Step 5: Leverage Embedded Domain Knowledge\n\nWhen external data is unavailable:\n\n- Use **general domain knowledge** appropriately\n- Clearly distinguish between verified facts and general guidance\n- Include **actionable frameworks** rather than specific unverified data\n- Add **placeholder notes** where specific data would enhance the document\n\nExample:\n```markdown\n> **Note**: Specific [metric\u002Fdata point] would typically be sourced from \n> [expected source]. The guidance below reflects established best practices \n> in this domain.\n```\n\n### Step 6: Validate Output\n\n```python\n# For shell_agent: verify the output exists and is valid\nlist_dir(path=\".\")  # Confirm file was created\n\n# For PDFs, optionally verify with pdfinfo via run_shell\n# For markdown, optionally read to confirm content\n\n# For write_file: confirm file exists\nlist_dir(path=\".\")  # Confirm file exists\n```\n\n## Code Example\n\n```python\n# Complete detection and pivot pattern\ndef generate_fallback_document(task_goal, output_format=\"markdown\"):\n    # After detecting tool failures:\n    \n    # Step 1: Choose method based on format\n    if output_format == \"pdf\" or \"complex\" in task_goal:\n        # Use shell_agent for complex formats\n        shell_agent(\n            task=f\"Generate a professional {output_format} document about \"\n                 f\"{task_goal}. Use appropriate libraries. Include executive \"\n                 f\"summary, main content, limitations, and recommendations. \"\n                 f\"Save to working directory.\"\n        )\n        return \"Document generated via shell_agent delegation\"\n    \n    else:\n        # Use write_file for simple formats\n        report_content = f\"\"\"# {task_goal} Report\n\n## Executive Summary\nThis report was generated using established domain knowledge due to \ntemporary unavailability of primary data sources.\n\n## Key Frameworks and Guidance\n[Structured content with headers, bullets, tables]\n\n## Limitations\n- Specific data points from [expected sources] were unavailable\n- Recommendations based on general best practices\n\n## Action Items\n1. [Concrete step 1]\n2. [Concrete step 2]\n\"\"\"\n        write_file(path=\"generated_report.md\", content=report_content)\n        return \"Report generated with write_file\"\n```\n\n## Best Practices\n\n| Do | Don't |\n|----|-------|\n| Choose method based on document complexity | Default to write_file for PDFs without consideration |\n| Delegate complex formatting to shell_agent | Keep retrying failing tools 5+ times |\n| Be transparent about limitations | Claim unverified specifics as facts |\n| Provide actionable frameworks | Leave the task incomplete |\n| Use professional document structure | Output unstructured text walls |\n| Validate the generated output | Assume success without verification |\n\n## Common Pitfalls\n\n1. **Wrong tool selection**: Using write_file for PDF generation instead of delegating to shell_agent\n2. **Over-apologizing**: Acknowledge limitations once, then deliver value\n3. **Under-delivering**: A well-structured partial report beats no report\n4. **Misrepresenting certainty**: Use appropriate hedging language\n5. **Skipping validation**: Always confirm the file was created successfully\n\n## Success Criteria\n\nThe skill is successfully applied when:\n\n- [ ] Document is generated despite tool failures\n- [ ] Appropriate method chosen (shell_agent for complex, write_file for simple)\n- [ ] Content is professionally structured (headers, sections, lists)\n- [ ] Limitations are transparently noted\n- [ ] Actionable guidance is provided\n- [ ] File is successfully created and verified\n\n## Method Comparison\n\n| Aspect | shell_agent | write_file |\n|--------|-------------|------------|\n| Best for | PDFs, complex layouts, multi-step generation | Markdown, plain text, single-pass content |\n| Error handling | Autonomous retry and fix | Manual error handling required |\n| Library access | Full library ecosystem | Content-only (no external libs) |\n| Processing time | Longer (agent reasoning) | Faster (direct write) |\n| Verification | Built-in validation | Manual verification needed |\n| Cost | Higher (agent execution) | Lower (simple operation) |\n\n## Decision Checklist\n\nBefore selecting method, ask:\n\n1. Is the output format PDF or other binary format? → **shell_agent**\n2. Does it require external libraries (reportlab, pandoc, etc.)? → **shell_agent**\n3. Is it markdown or plain text only? → **write_file**\n4. Does it need multi-step processing or validation? → **shell_agent**\n5. Is quick, simple content generation sufficient? → **write_file**\n",{"name":5,"description":7},"main","af1b40a42167d09c30b3a63000aa14ace4c1e229","gdpval_bench\u002Fskills\u002Fwrite-file-fallback-report-enhanced","https:\u002F\u002Fgithub.com\u002Fxyliu-cs\u002FOpenSpace-GDPval","https:\u002F\u002Fgithub.com\u002Fxyliu-cs","https:\u002F\u002Fxyliu-cs.github.io","PhD Student in CS, CUHK-Shenzhen.","2026-04-06 20:32:15",{"source":8,"skillName":5,"commands":55,"prompt":65,"zipAvailable":66},[56,59,62],{"tool":57,"command":58},"npx","npx skills add xyliu-cs\u002FOpenSpace-GDPval --skill fallback-doc-with-delegation",{"tool":60,"command":61},"bunx","bunx skills add xyliu-cs\u002FOpenSpace-GDPval --skill fallback-doc-with-delegation",{"tool":63,"command":64},"pnpm","pnpm dlx skills add xyliu-cs\u002FOpenSpace-GDPval --skill fallback-doc-with-delegation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fxyliu-cs\u002FOpenSpace-GDPval\u002Fblob\u002Fmain\u002Fgdpval_bench\u002Fskills\u002Fwrite-file-fallback-report-enhanced\u002FSKILL.md",true,[68],{"path":69,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":72,"sourceHtmlUrl":73},".skill_id","OTHER","其他",49,"https:\u002F\u002Fgithub.com\u002Fxyliu-cs\u002FOpenSpace-GDPval\u002Fblob\u002Fmain\u002Fgdpval_bench\u002Fskills\u002Fwrite-file-fallback-report-enhanced\u002F.skill_id",[75,88,96,111,131,146],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":41},"01M6KG8MF1S664SWBDENYRSCK8","zip-export-deliverable","Ensures codebase deliverables are properly exported as ZIP archives through a dedicated finalization step","https:\u002F\u002Fgithub.com\u002Fxyliu-cs\u002FOpenSpace-GDPval\u002Fblob\u002Fmain\u002Fgdpval_bench\u002Fskills\u002Fzip-export-deliverable\u002FSKILL.md",[],[82,83,84,85,86,87],"zip export deliverable","zip","export","codebase","deliverables","as",{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":41},"01M6KG8GAZWHHEK7VNJDW0ME21","zip-deliverable-finalization","Ensure codebase deliverables are properly packaged as ZIP exports in a dedicated final step","https:\u002F\u002Fgithub.com\u002Fxyliu-cs\u002FOpenSpace-GDPval\u002Fblob\u002Fmain\u002Fgdpval_bench\u002Fskills\u002Fzip-deliverable-finalization\u002FSKILL.md",[],[83,95,87,85,86],"zip deliverable finalization",{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":41},"01M6KG8CMQQ1PYMNTSRZ9VS1MD","zero-iteration-failure-analysis","Identify and handle agent failures with 0 iterations as pre-execution system issues","https:\u002F\u002Fgithub.com\u002Fxyliu-cs\u002FOpenSpace-GDPval\u002Fblob\u002Fmain\u002Fgdpval_bench\u002Fskills\u002Fzero-iteration-failure-analysis\u002FSKILL.md",[],[103,104,105,106,107,108,109,110,87,27],"ZeRO","analysis","iteration","failure analysis","zero iteration failure analysis","execution","issues","system",{"skillNo":112,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":122,"publishedAt":41},"01M6KG86W3PA07FT9W58YHHERM","write-file-fallback-report","Generate professional documents using write_file when primary data sources fail, leveraging embedded domain knowledge instead of external retrieval","https:\u002F\u002Fgithub.com\u002Fxyliu-cs\u002FOpenSpace-GDPval\u002Fblob\u002Fmain\u002Fgdpval_bench\u002Fskills\u002Fwrite-file-fallback-report\u002FSKILL.md",[117],{"categoryNo":118,"categoryName":119,"taxonomyType":120,"taxonomyTypeDesc":121},"01JBSEEDCAT000000000000034","行政与办公人员","PROFESSION","适用职业",[123,28,34,39,124,125,126,127,35,40,128,129,130,32],"write file fallback report","report","retrieval","knowledge","sources","data","embedded","domain",{"skillNo":132,"name":133,"nameZh":6,"description":134,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":135,"categories":136,"keywords":138,"publishedAt":41},"01M6KG7WT7GX5NYCGEFHW22NV7","context-anchored-fallback-report","Generate documents with write_file when retrieval tools fail, with explicit guardrails against task context drift","https:\u002F\u002Fgithub.com\u002Fxyliu-cs\u002FOpenSpace-GDPval\u002Fblob\u002Fmain\u002Fgdpval_bench\u002Fskills\u002Fwrite-file-fallback-report-enhanced-4bc5c2\u002FSKILL.md",[137],{"categoryNo":118,"categoryName":119,"taxonomyType":120,"taxonomyTypeDesc":121},[139,28,140,124,34,35,40,39,141,142,143,144,125,27,145],"context anchored fallback report","context","Guardrails","Tools","task","drift","context drift",{"skillNo":147,"name":148,"nameZh":6,"description":149,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":150,"categories":151,"keywords":152,"publishedAt":41},"01M6KG7R5SVB1ZAKZQADKAQWYB","working-directory-resolution","Resolve file access failures by changing to the correct working directory before command execution","https:\u002F\u002Fgithub.com\u002Fxyliu-cs\u002FOpenSpace-GDPval\u002Fblob\u002Fmain\u002Fgdpval_bench\u002Fskills\u002Fworking-directory-resolution\u002FSKILL.md",[],[153,154,155,156,39,108,157,158,159],"directory","working directory resolution","resolution","file access","access","correct","command"]