sandbox-failure-recovery
Recover from execute_code_sandbox failures by writing code to file and executing via run_shell
- License 许可
- MIT
- 所属仓库
- HKUDS/AgentPilot
- 最近更新
- 2026-11-12
- Stars
- ★ 8.0k
- Forks
- 944
01 前言 Front Matter
- name
- sandbox-failure-recovery
- description
- Recover from execute_code_sandbox failures by writing code to file and executing via run_shell
02 资源文件 1
- 其他.skill_id38 B
03 SKILL.md 正文
Sandbox Failure Recovery Workflow
When execute_code_sandbox fails (often due to infrastructure issues, timeouts, or complex dependencies), use this recovery pattern to achieve identical results by writing the code to a file and executing it directly.
When to Use This Skill
execute_code_sandboxreturns an error or times out- The task involves spreadsheet generation (Excel, CSV) or file output
- You have working Python code that needs to be executed
- The sandbox environment appears unstable
Step-by-Step Instructions
Step 1: Preserve the Failed Code
When execute_code_sandbox fails, capture the Python code that was attempted. If the code was generated but not saved, reconstruct it from the execution attempt.
Step 2: Write Code to File
Use write_file to save the Python script to a .py file:
write_file with:
- path: "script_name.py" (e.g., "generate_report.py", "process_data.py")
- content: <the full Python code>
Example:
# Content for write_file
import pandas as pd
from openpyxl import Workbook
# Your spreadsheet generation logic here
df = pd.DataFrame({'Revenue': [100, 200, 300]})
df.to_excel('output.xlsx', index=False)
Step 3: Execute via run_shell
Run the saved script using run_shell with Python 3:
run_shell with:
- command: "python3 script_name.py"
- timeout: 60 (or higher for complex operations)
Step 4: Verify Output
Check that the expected output files were created:
list_dir with:
- path: "."
Or read the generated file to confirm correctness:
read_file with:
- file_path: "output.xlsx"
- filetype: "xlsx"
Complete Example
Scenario: Generate an Excel P&L report after sandbox failure.
# Step 1: Write the Python script
write_file:
path: "generate_pnl_report.py"
content: |
import pandas as pd
from openpyxl import Workbook
# Create revenue data
data = {
'Tour Stop': ['London', 'Paris', 'Berlin'],
'Revenue': [50000, 45000, 38000],
'Withholding Tax': [5000, 4500, 3800],
'Expenses': [12000, 11000, 9500]
}
df = pd.DataFrame(data)
df['Net Income'] = df['Revenue'] - df['Withholding Tax'] - df['Expenses']
# Export to Excel
df.to_excel('pnl_report.xlsx', index=False)
print("Report generated successfully")
# Step 2: Execute the script
run_shell:
command: "python3 generate_pnl_report.py"
timeout: 60
# Step 3: Verify
list_dir:
path: "."
Best Practices
-
Use descriptive filenames - Name files after their purpose (e.g.,
generate_report.py,process_spreadsheet.py) -
Set appropriate timeouts - Complex operations may need 60+ seconds
-
Include error handling in your Python code - Add try/except blocks to capture and report issues:
try: # Your code here except Exception as e: print(f"Error: {e}") raise -
Clean up temporary files - After successful execution, you may remove the
.pyfile if not needed -
Check dependencies - Ensure required packages (pandas, openpyxl, etc.) are available in the shell environment
Why This Works
run_shellexecutes in a more stable environment than the sandbox- File-based execution avoids sandbox memory/resource constraints
- The Python interpreter in the shell has full access to installed packages
- Output files are written directly to the workspace
<THIS_SKILL_EVOLVED>
04 同仓库更多 Skills
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.
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.
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.
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.
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.
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 /tmp helper scripts.