Menu
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/jeremylongshore/oss-agent-lab --skill sandboxThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Showing SKILL.md
Based on SOC occupation classification
| name | sandbox |
| display_name | Sandbox Specialist |
| description | Safe multi-language code execution via alibaba/OpenSandbox |
| version | 0.1.0 |
| source_repo | alibaba/OpenSandbox |
| license | Apache-2.0 |
| tier | experimental |
| capabilities | ["execute","code_execution","sandbox","multi_language"] |
| allowed_tools | ["execute_code","validate_code","list_runtimes"] |
| output_formats | ["python_api","cli","mcp_server","agent_skill","rest_api"] |
Wraps alibaba/OpenSandbox to provide isolated, resource-limited execution of arbitrary code snippets inside OSS Agent Lab. Each execution runs in a container-backed sandbox with seccomp syscall filters, memory limits, and configurable timeouts — no persistent side effects leak between runs.
Supported runtimes: Python, JavaScript, TypeScript, Bash, Ruby, Go, Rust, Java, C, C++.
| Tool | Description | Side Effects |
|---|---|---|
execute_code | Execute a code snippet in the sandbox; returns captured output | Subprocess spawn (sandboxed) |
validate_code | Static analysis without execution; returns errors and warnings | None |
list_runtimes | Enumerate all registered runtimes with version and availability | None |
from agents.specialists.sandbox.agent import SandboxSpecialist
from oss_agent_lab.contracts import Intent, Query, SpecialistRequest
specialist = SandboxSpecialist()
request = SpecialistRequest(
intent=Intent(
action="execute",
domain="code",
confidence=0.95,
parameters={"code": "print('hello, sandbox!')", "language": "python"},
),
query=Query(user_input="print('hello, sandbox!')"),
specialist_name="sandbox",
)
result = await specialist.execute(request)
(result.result[])
oss-lab run sandbox "print('hello, sandbox!')"
request = SpecialistRequest(
intent=Intent(
action="execute",
domain="code",
confidence=0.95,
parameters={
"code": "console.log('hello from node');",
"language": "javascript",
"timeout": 10,
},
),
query=Query(user_input="console.log('hello from node');"),
specialist_name="sandbox",
)
request = SpecialistRequest(
intent=Intent(
action="validate",
domain="code",
confidence=0.9,
parameters={"code": "import os; os.system('ls')", "language": "python"},
),
query=Query(user_input="import os; os.system('ls')"),
specialist_name="sandbox",
)
result = await specialist.execute(request)
print(result.result["valid"]) # False (warnings present)
print(result.result["warnings"]) # ['os.system() executes shell commands...']
request = SpecialistRequest(
intent=Intent(action="list_runtimes", domain="code", confidence=1.0),
query=Query(user_input=""),
specialist_name="sandbox",
)
result = await specialist.execute(request)
for rt in result.result["runtimes"]:
print(rt["name"], rt["version"], rt["available"])
{
"stdout": "[sandbox:python] Execution started\n[sandbox:python] 1 line(s) processed\n[sandbox:python] Execution complete",
"stderr": "",
"exit_code": 0,
"execution_time_ms": 12.4,
"language": "python",
"execution_id": "3f2504e0-4f89-11d3-9a0c-0305e82c3301"
}
{
"valid": true,
"errors": [],
"warnings": ["eval() executes arbitrary code strings."],
"language": "python"
}
{
"runtimes": [
{"name": "python", "version": "3.12.3", "available": true},
{"name": "javascript", "version": "node 22.2.0", "available": true}
],
"total_count": 12,
"available_count": 10
}
Wraps alibaba/OpenSandbox.