AGENT SKILL

python-shell-fallback

Fallback to shell execution when sandbox fails for library-dependent Python code

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
python-shell-fallback
description
Fallback to shell execution when sandbox fails for library-dependent Python code

02 资源文件 1

03 SKILL.md 正文

Python Shell Fallback

When execute_code_sandbox fails for Python code that depends on external libraries, use run_shell with a heredoc Python script as a reliable fallback. The shell environment typically has better access to installed packages and system configuration than the sandbox.

When to Use This Pattern

Apply this skill when:

  • execute_code_sandbox returns errors related to missing libraries (e.g., ModuleNotFoundError)
  • The code requires packages that may not be installed in the sandbox environment
  • You need PDF generation, data visualization, or other library-heavy operations
  • Previous sandbox execution attempts have failed with import errors

How to Implement

Step 1: Identify the Failure

Check if the sandbox error indicates a library/dependency issue:

  • ModuleNotFoundError: No module named 'xxx'
  • ImportError: cannot import name 'xxx'
  • Similar package-related errors

Step 2: Convert to Shell Execution

Rewrite the code execution using run_shell with a heredoc:

python3 << 'EOF'
# Your Python code here
import library_name
# ... rest of code
EOF

Step 3: Execute via run_shell

Call run_shell with the heredoc Python script:

command: python3 << 'EOF'
import reportlab
from reportlab.lib.pagesizes import letter
from reportlab.pdfgen import canvas

# Your library-dependent code here
c = canvas.Canvas("output.pdf", pagesize=letter)
c.drawString(100, 750, "Hello World")
c.save()
EOF

Example: PDF Generation Fallback

Sandbox attempt (fails):

# execute_code_sandbox call fails with:
# ModuleNotFoundError: No module named 'reportlab'

Shell fallback (works):

python3 << 'EOF'
from reportlab.lib.pagesizes import letter
from reportlab.pdfgen import canvas

c = canvas.Canvas("legal_memo.pdf", pagesize=letter)
c.setTitle("Legal Memorandum")
c.drawString(100, 750, "CONFIDENTIAL LEGAL MEMORANDUM")
c.save()
print("PDF created successfully")
EOF

Best Practices

  1. Use quoted heredoc (<< 'EOF') - Prevents variable expansion, keeping Python code intact
  2. Verify output - Check stdout/stderr from run_shell to confirm success
  3. Handle file paths - Ensure output files are written to accessible directories
  4. Test incrementally - For complex scripts, test in smaller chunks first
  5. Include error handling - Add try/except blocks to capture issues gracefully

Alternative: Inline Python Command

For simple one-liners or short scripts:

python3 -c "import reportlab; print(reportlab.__version__)"

Troubleshooting

If shell execution also fails:

  • Check if Python3 is available: which python3
  • Verify package installation: pip3 list | grep package_name
  • Try python instead of python3 (some systems differ)
  • Consider installing the package: pip3 install package_name

04 同仓库更多 Skills

HKUDS
HKUDS/AgentPilot

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.

★ 8.0k
HKUDS
HKUDS/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.

★ 8.0k
HKUDS
HKUDS/AgentPilot

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.

Skill与智能体开发
★ 8.0k
HKUDS
HKUDS/AgentPilot

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.

Skill与智能体开发
★ 8.0k
HKUDS
HKUDS/AgentPilot

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.

★ 8.0k
HKUDS
HKUDS/AgentPilot

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.

★ 8.0k