Dive into the current line of code.
AI agents invoke step_into to trigger actions in DebugMCP. 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.
While step_into itself doesn't modify data directly, it is fundamentally an execution control tool that advances program state through the debugger. The severity is high because an AI agent could use this in combination with other tools (evaluate_expression, get_variables_values) to systematically explore and potentially exploit code paths, access sensitive data during execution, or trigger unintended side effects.
From the tool's definition Tool 'step_into' is part of a debugging suite that 'debug code inside VS Code by setting breakpoints, stepping through execution, inspecting variables, and evaluating expressions.' Stepping into code execution is an Execute action that controls program flow…
Documented attack patterns abuse exactly the kind of access step_into gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and DebugMCP, and nothing reaches the server without passing your rules. This is the rule we recommend for step_into:
{
"version": "1",
"default": "deny",
"tools": {
"step_into": {
"limits": [
{
"counter": "step_into_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}step_into 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.
Dive into the current line of code. It is categorised as a Execute tool in the DebugMCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Debug MCP server in PolicyLayer and add a rule for step_into: 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 DebugMCP. Nothing to install.
step_into 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 step_into 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 step_into. 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.
step_into is provided by the Debug MCP server (microsoft/debugmcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from DebugMCP, 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.
13 DebugMCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.