Run code in a sandboxed subprocess.${RB} Languages: ${PB}. Think-in-Code \u2014 the core philosophy: the bytes your code processes never enter your conversation memory; only what you console.log() does. Reading a 700 KB log directly means 700 KB of your remaining reasoning capacity gets spent on ...
AI agents invoke ctx_execute to trigger actions in Context Mode. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool executes arbitrary code in a subprocess across multiple programming languages. Although the execution occurs in a sandbox (which mitigates some risk), the tool still permits running code whose effects depend on the arguments provided—the defining characteristic of Execute category.
From the tool's definition Tool name is 'ctx_execute' with description 'Run code in a sandboxed subprocess' and explicit mention of executing code in multiple programming languages. The tool enables code execution in a sandboxed environment as its core function.
Documented attack patterns abuse exactly the kind of access ctx_execute gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Context Mode, and nothing reaches the server without passing your rules. This is the rule we recommend for ctx_execute:
{
"version": "1",
"default": "deny",
"tools": {
"ctx_execute": {
"limits": [
{
"counter": "ctx_execute_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}ctx_execute stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Run code in a sandboxed subprocess.${RB} Languages: ${PB}. Think-in-Code \u2014 the core philosophy: the bytes your code processes never enter your conversation memory; only what you console.log() does. Reading a 700 KB log directly means 700 KB of your remaining reasoning capacity gets spent on raw bytes. Running code over that same log in this sandbox and printing a 3 KB summary leaves you with 697 KB of capacity for the actual work. Concrete shape \u2014 analyze 47 source files without reading any of them: ctx_execute(language:. It is categorised as a Execute tool in the Context Mode MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Context Mode MCP server in PolicyLayer and add a rule for ctx_execute: 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 Context Mode. Nothing to install.
ctx_execute is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the ctx_execute 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.
Set action: deny in the PolicyLayer policy for ctx_execute. 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.
ctx_execute is provided by the Context Mode MCP server (mksglu/context-mode). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Context Mode, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
11 Context Mode tools catalogued and risk-classified — across an index of 43,000+ MCP servers.