Create a safety branch. Branch name must start with ai-change/.
AI agents use mcp_git_create_branch to create or update resources in Local Code — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Local Code environment.
Creating a Git branch is a reversible write operation—it modifies the repository state by adding a new branch reference, but does not delete data or execute arbitrary code. The branch can be deleted if needed. This is less severe than Destructive (which cannot be undone) or Execute (which runs arbitrary operations).
From the tool's definition Tool creates a new Git branch with the description 'Create a safety branch.' This is a write operation that creates a new reference in version control.
Attacks that exploit this kind of access
Create a safety branch. Branch name must start with ai-change/. It is categorised as a Write tool in the Local Code MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Local Code MCP server in PolicyLayer and add a rule for mcp_git_create_branch: 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 Local Code. Nothing to install.
mcp_git_create_branch 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 mcp_git_create_branch 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 mcp_git_create_branch. 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.
mcp_git_create_branch is provided by the Local Code MCP server (js190-prog/local-code-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →