AI agents use edit_code to create or update resources in Claude Code — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Claude Code environment.
This tool creates or modifies code reversibly based on user instructions. While it modifies code (potentially important files), the changes are reversible through version control or undo operations, and the tool itself does not execute the code or delete it irreversibly. Therefore it falls under Write rather than Execute or Destructive.
From the tool's definition Tool name is 'edit_code' and description states it 'Edits the given code based on instructions.' The verb 'edits' indicates modification of code artifacts.
Documented attack patterns abuse exactly the kind of access edit_code gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Claude Code, and nothing reaches the server without passing your rules. This is the rule we recommend for edit_code:
{
"version": "1",
"default": "deny",
"tools": {
"edit_code": {
"limits": [
{
"counter": "edit_code_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}edit_code stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Edits the given code based on instructions. It is categorised as a Write tool in the Claude Code MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Claude Code MCP server in PolicyLayer and add a rule for edit_code: 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 Claude Code. Nothing to install.
edit_code is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the edit_code 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 edit_code. 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.
edit_code is provided by the Claude Code MCP server (kunihiros/claude-code-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Claude Code, 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.
7 Claude Code tools catalogued and risk-classified — across an index of 43,000+ MCP servers.