[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6HXF7VE1F78VGVFEBNRBKQN":3,"skill-same:01M6HXF7VE1F78VGVFEBNRBKQN":66},{"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":33,"linkOnly":34,"markdownContent":35,"markdownBody":36,"frontMatter":37,"refName":38,"commitSha":39,"skillDirectory":40,"forksCount":41,"repositoryHtmlUrl":42,"repositoryHomepageUrl":6,"ownerHtmlUrl":43,"ownerBlogUrl":44,"ownerDescription":6,"updatedAt":45,"install":46,"resources":59},"01M6HXF7VE1F78VGVFEBNRBKQN","code-execution-fallback",null,"Handle code execution failures with fallback strategies and anchored workspace paths","HKUDS\u002FAgentPilot","HKUDS","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F118165258?v=4","01M6HW8VRNZP64G0P4MVJW2KKQ","✨Data Intelligence Lab@HKU✨","01M6HW8VWZVHBRHYH3T330PFWV",8029,"MIT","https:\u002F\u002Fgithub.com\u002FHKUDS\u002FAgentPilot\u002Fblob\u002Fmain\u002Fopenspace\u002Flegacy\u002Fv1\u002Fgdpval_bench\u002Fskills\u002Fcode-execution-fallback\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[24,25,26,27,28,29,30,31,32],"code","execution","fallback","code execution","code execution fallback","workspace","strategies","paths","with","2026-11-10 10:12:30",false,"---\nname: code-execution-fallback\ndescription: Handle code execution failures with fallback strategies and anchored workspace paths\n---\n\n# Code Execution Fallback & Workspace Anchoring\n\nThis skill provides a robust pattern for executing code when the primary method fails, combined with proper workspace path management to prevent file location errors.\n\n## Core Techniques\n\n### 1. Workspace Path Anchoring\n\nAlways establish and verify your working directory at the start of any task:\n\n```python\n# At the beginning of any code execution\nimport os\nworkspace_path = os.getcwd()\nprint(f\"Working directory: {workspace_path}\")\n```\n\n```bash\n# In shell scripts\npwd\necho \"Current directory: $(pwd)\"\n```\n\n**Why:** Prevents files from being written to unexpected locations when agents switch between tools.\n\n### 2. Execution Fallback Ladder\n\nWhen `execute_code_sandbox` fails, follow this escalation pattern:\n\n#### Level 1: Retry with Simpler Code\n- Simplify the code structure\n- Remove complex dependencies\n- Add explicit error handling\n\n#### Level 2: Use `run_shell` with Heredoc\nWhen sandbox execution repeatedly fails, switch to shell execution:\n\n```bash\npython3 \u003C\u003C 'EOF'\nimport os\nimport pandas as pd\n\n# Your code here\ndata = {\"col1\": [1, 2, 3], \"col2\": [\"a\", \"b\", \"c\"]}\ndf = pd.DataFrame(data)\ndf.to_csv(\"output.csv\", index=False)\nprint(\"File written successfully\")\nEOF\n```\n\n**Key points:**\n- Use `\u003C\u003C 'EOF'` (quoted) to prevent variable expansion\n- Include all imports and dependencies inline\n- Add explicit success\u002Ffailure messages\n\n#### Level 3: Delegate to `shell_agent`\nFor complex multi-step tasks with error recovery needs:\n\n```\nTask: Create a data processing pipeline that reads CSV, transforms data, and outputs Excel\nRequirements:\n- Handle missing values\n- Apply transformations\n- Write to .\u002Foutput\u002F directory\n- Retry on transient errors\n```\n\n### 3. Explicit Path Management\n\nAlways use absolute or explicitly relative paths:\n\n```python\n# BAD - relies on implicit working directory\ndf.to_csv(\"output\u002Fdata.csv\")\n\n# GOOD - explicit path anchoring\nimport os\nbase_path = os.getcwd()\noutput_dir = os.path.join(base_path, \"output\")\nos.makedirs(output_dir, exist_ok=True)\ndf.to_csv(os.path.join(output_dir, \"data.csv\"))\n```\n\n```bash\n# BAD\ncd some_dir && python script.py\n\n# GOOD\nSCRIPT_DIR=\"$(cd \"$(dirname \"${BASH_SOURCE[0]}\")\" && pwd)\"\ncd \"$SCRIPT_DIR\"\npython script.py\n```\n\n## Decision Tree\n\n```\nexecute_code_sandbox fails?\n├── Yes, with syntax\u002Fimport errors → Fix code, retry Level 1\n├── Yes, with timeout\u002Fresource errors → Use Level 2 (run_shell heredoc)\n├── Yes, with unknown\u002Funclear errors → Use Level 3 (shell_agent)\n└── No, success → Verify output file exists at expected path\n```\n\n## Common Failure Scenarios & Solutions\n\n| Error Type | Likely Cause | Recommended Fallback |\n|------------|--------------|---------------------|\n| ModuleNotFoundError | Missing packages | run_shell with pip install first |\n| Timeout | Long-running operation | shell_agent with progress tracking |\n| PermissionError | Wrong directory | Verify workspace path, use explicit paths |\n| Unknown error | Sandbox limitations | run_shell or shell_agent |\n\n## Example: Robust File Generation\n\n```python\n# Step 1: Anchor workspace\nimport os\nworkspace = os.getcwd()\nprint(f\"Workspace: {workspace}\")\n\n# Step 2: Create output directory explicitly\noutput_path = os.path.join(workspace, \"deliverables\")\nos.makedirs(output_path, exist_ok=True)\n\n# Step 3: Generate content with error handling\ntry:\n    # Your generation logic here\n    with open(os.path.join(output_path, \"report.txt\"), \"w\") as f:\n        f.write(\"Content here\")\n    print(f\"Success: File written to {output_path}\")\nexcept Exception as e:\n    print(f\"Error: {e}\")\n    # Signal to escalate to run_shell or shell_agent\n    raise\n```\n\n## Anti-Patterns to Avoid\n\n- ❌ Assuming current directory without verification\n- ❌ Using relative paths like `..\u002Foutput\u002Ffile.txt` without context\n- ❌ Repeatedly retrying failed `execute_code_sandbox` without changing approach\n- ❌ Not checking if output files exist after generation\n- ❌ Mixing implicit and explicit path styles in same task\n\n## Verification Checklist\n\nAfter any code execution:\n- [ ] Confirm working directory was verified at start\n- [ ] Confirm output files exist at expected paths\n- [ ] Confirm file contents are non-empty and valid\n- [ ] If execution failed, escalate to next fallback level within 2 retries","# Code Execution Fallback & Workspace Anchoring\n\nThis skill provides a robust pattern for executing code when the primary method fails, combined with proper workspace path management to prevent file location errors.\n\n## Core Techniques\n\n### 1. Workspace Path Anchoring\n\nAlways establish and verify your working directory at the start of any task:\n\n```python\n# At the beginning of any code execution\nimport os\nworkspace_path = os.getcwd()\nprint(f\"Working directory: {workspace_path}\")\n```\n\n```bash\n# In shell scripts\npwd\necho \"Current directory: $(pwd)\"\n```\n\n**Why:** Prevents files from being written to unexpected locations when agents switch between tools.\n\n### 2. Execution Fallback Ladder\n\nWhen `execute_code_sandbox` fails, follow this escalation pattern:\n\n#### Level 1: Retry with Simpler Code\n- Simplify the code structure\n- Remove complex dependencies\n- Add explicit error handling\n\n#### Level 2: Use `run_shell` with Heredoc\nWhen sandbox execution repeatedly fails, switch to shell execution:\n\n```bash\npython3 \u003C\u003C 'EOF'\nimport os\nimport pandas as pd\n\n# Your code here\ndata = {\"col1\": [1, 2, 3], \"col2\": [\"a\", \"b\", \"c\"]}\ndf = pd.DataFrame(data)\ndf.to_csv(\"output.csv\", index=False)\nprint(\"File written successfully\")\nEOF\n```\n\n**Key points:**\n- Use `\u003C\u003C 'EOF'` (quoted) to prevent variable expansion\n- Include all imports and dependencies inline\n- Add explicit success\u002Ffailure messages\n\n#### Level 3: Delegate to `shell_agent`\nFor complex multi-step tasks with error recovery needs:\n\n```\nTask: Create a data processing pipeline that reads CSV, transforms data, and outputs Excel\nRequirements:\n- Handle missing values\n- Apply transformations\n- Write to .\u002Foutput\u002F directory\n- Retry on transient errors\n```\n\n### 3. Explicit Path Management\n\nAlways use absolute or explicitly relative paths:\n\n```python\n# BAD - relies on implicit working directory\ndf.to_csv(\"output\u002Fdata.csv\")\n\n# GOOD - explicit path anchoring\nimport os\nbase_path = os.getcwd()\noutput_dir = os.path.join(base_path, \"output\")\nos.makedirs(output_dir, exist_ok=True)\ndf.to_csv(os.path.join(output_dir, \"data.csv\"))\n```\n\n```bash\n# BAD\ncd some_dir && python script.py\n\n# GOOD\nSCRIPT_DIR=\"$(cd \"$(dirname \"${BASH_SOURCE[0]}\")\" && pwd)\"\ncd \"$SCRIPT_DIR\"\npython script.py\n```\n\n## Decision Tree\n\n```\nexecute_code_sandbox fails?\n├── Yes, with syntax\u002Fimport errors → Fix code, retry Level 1\n├── Yes, with timeout\u002Fresource errors → Use Level 2 (run_shell heredoc)\n├── Yes, with unknown\u002Funclear errors → Use Level 3 (shell_agent)\n└── No, success → Verify output file exists at expected path\n```\n\n## Common Failure Scenarios & Solutions\n\n| Error Type | Likely Cause | Recommended Fallback |\n|------------|--------------|---------------------|\n| ModuleNotFoundError | Missing packages | run_shell with pip install first |\n| Timeout | Long-running operation | shell_agent with progress tracking |\n| PermissionError | Wrong directory | Verify workspace path, use explicit paths |\n| Unknown error | Sandbox limitations | run_shell or shell_agent |\n\n## Example: Robust File Generation\n\n```python\n# Step 1: Anchor workspace\nimport os\nworkspace = os.getcwd()\nprint(f\"Workspace: {workspace}\")\n\n# Step 2: Create output directory explicitly\noutput_path = os.path.join(workspace, \"deliverables\")\nos.makedirs(output_path, exist_ok=True)\n\n# Step 3: Generate content with error handling\ntry:\n    # Your generation logic here\n    with open(os.path.join(output_path, \"report.txt\"), \"w\") as f:\n        f.write(\"Content here\")\n    print(f\"Success: File written to {output_path}\")\nexcept Exception as e:\n    print(f\"Error: {e}\")\n    # Signal to escalate to run_shell or shell_agent\n    raise\n```\n\n## Anti-Patterns to Avoid\n\n- ❌ Assuming current directory without verification\n- ❌ Using relative paths like `..\u002Foutput\u002Ffile.txt` without context\n- ❌ Repeatedly retrying failed `execute_code_sandbox` without changing approach\n- ❌ Not checking if output files exist after generation\n- ❌ Mixing implicit and explicit path styles in same task\n\n## Verification Checklist\n\nAfter any code execution:\n- [ ] Confirm working directory was verified at start\n- [ ] Confirm output files exist at expected paths\n- [ ] Confirm file contents are non-empty and valid\n- [ ] If execution failed, escalate to next fallback level within 2 retries",{"name":5,"description":7},"main","6b84b0689ca0d9e7b1e4061188ff6b60758e0f8a","openspace\u002Flegacy\u002Fv1\u002Fgdpval_bench\u002Fskills\u002Fcode-execution-fallback",944,"https:\u002F\u002Fgithub.com\u002FHKUDS\u002FAgentPilot","https:\u002F\u002Fgithub.com\u002FHKUDS","https:\u002F\u002Fsites.google.com\u002Fview\u002Fchaoh","2026-11-12 22:26:39",{"source":8,"skillName":5,"commands":47,"prompt":57,"zipAvailable":58},[48,51,54],{"tool":49,"command":50},"npx","npx skills add HKUDS\u002FAgentPilot --skill code-execution-fallback",{"tool":52,"command":53},"bunx","bunx skills add HKUDS\u002FAgentPilot --skill code-execution-fallback",{"tool":55,"command":56},"pnpm","pnpm dlx skills add HKUDS\u002FAgentPilot --skill code-execution-fallback","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FHKUDS\u002FAgentPilot\u002Fblob\u002Fmain\u002Fopenspace\u002Flegacy\u002Fv1\u002Fgdpval_bench\u002Fskills\u002Fcode-execution-fallback\u002FSKILL.md",true,[60],{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},".skill_id","OTHER","其他",37,"https:\u002F\u002Fgithub.com\u002FHKUDS\u002FAgentPilot\u002Fblob\u002Fmain\u002Fopenspace\u002Flegacy\u002Fv1\u002Fgdpval_bench\u002Fskills\u002Fcode-execution-fallback\u002F.skill_id",[67,86,105,130,154,177],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":33},"01M6HYB297NDABRERJRVG0K8R0","agentpilot","Set up AgentPilot and use its single entry to prepare missing inputs and run configured RSI experiments. Use when the user requests AgentPilot installation or work with AgentPilot.","https:\u002F\u002Fgithub.com\u002FHKUDS\u002FAgentPilot\u002Fblob\u002Fmain\u002Fplugins\u002Fagentpilot\u002Fskills\u002Fagentpilot\u002FSKILL.md",[],[69,74,75,32,76,77,78,79,80,81,82,83,84,85],"prepare","requests","missing","installation","rsi","up","use","user","work","SET","run","experiments",{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":33},"01M6HYAYWXTTE1J7DJM0MFNQ4H","remember","Review auto-memory entries and propose promotions to OPENSPACE.md, OPENSPACE.local.md, or retained auto-memory topics; detect duplicates, stale facts, conflicts, and ambiguous entries before making changes.","https:\u002F\u002Fgithub.com\u002FHKUDS\u002FAgentPilot\u002Fblob\u002Fmain\u002Fopenspace\u002Fopenspace\u002Fskills\u002Fremember\u002FSKILL.md",[],[88,93,94,95,96,97,98,99,100,101,102,103,104],"memory","review","auto","openspace","changes","facts","detect","entries","local","conflicts","stale","duplicates",{"skillNo":106,"name":107,"nameZh":6,"description":108,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":116,"publishedAt":33},"01M6HYAVMPM4SBDV60R2B5HPV8","skill-discovery","Search reusable OpenSpace skills locally, and browse cloud packages step by step when a cloud skill may help. Reusing proven skills saves tokens, improves reliability, and extends your capabilities beyond built-in tools.","https:\u002F\u002Fgithub.com\u002FHKUDS\u002FAgentPilot\u002Fblob\u002Fmain\u002Fopenspace\u002Fopenspace\u002Fhost_skills\u002Fskill-discovery\u002FSKILL.md",[111],{"categoryNo":112,"categoryName":113,"taxonomyType":114,"taxonomyTypeDesc":115},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[117,118,119,96,120,121,122,123,124,125,126,127,128,129],"skill discovery","discovery","reusable","step by step","tokens","packages","help","cloud","reliability","capabilities","browse","search","Tools",{"skillNo":131,"name":132,"nameZh":6,"description":133,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":134,"categories":135,"keywords":137,"publishedAt":33},"01M6HYAR811B3RB0M50RBEDGN6","delegate-task","Delegate tasks to OpenSpace — a full-stack autonomous worker for coding, DevOps, web research, and desktop automation, backed by an extensive MCP tool and skill library. Skills auto-improve through use, reducing token consumption over time. A cloud community lets agents share and collectively evolve reusable skills.","https:\u002F\u002Fgithub.com\u002FHKUDS\u002FAgentPilot\u002Fblob\u002Fmain\u002Fopenspace\u002Fopenspace\u002Fhost_skills\u002Fdelegate-task\u002FSKILL.md",[136],{"categoryNo":112,"categoryName":113,"taxonomyType":114,"taxonomyTypeDesc":115},[138,139,140,96,141,142,143,144,145,119,95,80,146,147,148,149,150,151,152,153],"delegate","task","delegate task","time","improve","desktop automation","worker","community","autonomous","tool","stack","for","share","tasks","automation","MCP",{"skillNo":155,"name":156,"nameZh":6,"description":157,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":158,"categories":159,"keywords":160,"publishedAt":33},"01M6HYAM9FSGV6E52AYENDQXYH","write-special-chars-ts-check","Workaround for write_file failures caused by special characters (apostrophes, backticks, template literals) using a Python heredoc script, plus a reliable TypeScript type-checking step that filters output to only the newly-written files to avoid being misled by pre-existing errors.","https:\u002F\u002Fgithub.com\u002FHKUDS\u002FAgentPilot\u002Fblob\u002Fmain\u002Fopenspace\u002Flegacy\u002Fv1\u002Fshowcase\u002Fskills\u002Fwrite-special-chars-ts-check\u002FSKILL.md",[],[161,162,163,164,165,166,167,168,169,170,171,172,173,174,149,175,176],"write special chars ts check","write","check","type","files","filters","only","template","file","python","typescript","output","Existing","type-checking","errors","script",{"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":183,"publishedAt":33},"01M6HYAH05HGV7SVWSVR8PJRQD","write-special-chars-readback","Workaround for write_file failures caused by special characters (apostrophes, backticks, template literals) using a Python heredoc script, with mandatory post-write verification via read_file and cleanup of \u002Ftmp helper scripts.","https:\u002F\u002Fgithub.com\u002FHKUDS\u002FAgentPilot\u002Fblob\u002Fmain\u002Fopenspace\u002Flegacy\u002Fv1\u002Fshowcase\u002Fskills\u002Fwrite-special-chars-readback\u002FSKILL.md",[],[184,185,162,32,149,176,186,187,188,189,190,168,169,191,192,170],"write special chars readback","readback","helper","post","via","scripts","read","verification","cleanup"]