ai_sandbox

Returns stdout, stderr & exit code from running Python / JavaScript / Bash code in an isolated sandbox (prefix the code Cost: $0.005–$0.05 USDC on Base per call.

SERVEROmnicall — Agent Tools Gateway SOURCEhttps://omnicall.gocreativeai.com/mcp

Low RISK CLASS

Category Read

Parameters 11 required

Recommended Allowedsee the rule below

Registry record Grade C, identity unverified Pull the record →

This record as markdown: /tools/com-gocreativeai-omnicall-agent-tools/ai-sandbox.md

WHAT IT DOES

What ai_sandbox does on Omnicall — Agent Tools Gateway

AI agents call ai_sandbox to retrieve information from Omnicall — Agent Tools Gateway without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.

Parameter	Type	Required	Description
`arg`	string	Yes

Parameters from the server's own tool schema.

RISK

Why ai_sandbox is rated Low

Even though ai_sandbox only reads data, uncontrolled read access leaks sensitive information and racks up API costs: an agent caught in a retry loop can make thousands of calls a minute without anyone noticing.

Attacks that exploit this kind of access

RECOMMENDED RULE

The rule that runs ai_sandbox safely

PolicyLayer is an MCP gateway: it sits between your AI agents and Omnicall — Agent Tools Gateway, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For ai_sandbox, this is the rule to start with:

ai_sandbox Allowed

ai_sandbox is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "ai_sandbox": {}
  }
}

ALLOW ONLY THIS TOOL → Instant setup, no code required.

The button opens the PolicyLayer dashboard: create your workspace, connect Omnicall — Agent Tools Gateway, apply this rule, and every ai_sandbox call is checked against it from then on.

FAQ

Questions about ai_sandbox

What does the ai_sandbox tool do? +

Returns stdout, stderr & exit code from running Python / JavaScript / Bash code in an isolated sandbox (prefix the code Cost: $0.005–$0.05 USDC on Base per call. It is categorised as a Read tool in the Omnicall — Agent Tools Gateway MCP Server, which means it retrieves data without modifying state.

What parameters does ai_sandbox accept? +

ai_sandbox accepts 1 parameter: arg. Required: arg. The full parameter table on this page comes from the server's own tool schema.

How do I enforce a policy on ai_sandbox? +

Register the Omnicall — Agent Tools Gateway MCP server in PolicyLayer and add a rule for ai_sandbox: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Omnicall — Agent Tools Gateway. Nothing to install.

What risk level is ai_sandbox? +

ai_sandbox is a Read tool with low risk. Read-only tools are generally safe to allow by default.

Can I rate-limit ai_sandbox? +

Yes. Add a rate_limit block to the ai_sandbox rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block ai_sandbox completely? +

Set action: deny in the PolicyLayer policy for ai_sandbox. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides ai_sandbox? +

ai_sandbox is provided by the Omnicall — Agent Tools Gateway MCP server (https://omnicall.gocreativeai.com/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

KEEP EXPLORING

More on Omnicall — Agent Tools Gateway, and thousands of servers like it.

This server

All 79 Omnicall — Agent Tools Gateway tools →

Across the catalogue

The MCP Attack Database →

Guides

Roll out MCP under one policy →Data exfiltration →MCP token cost →