AI agents use copy-project to create or update resources in GitHub Projects MCP Server — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your GitHub Projects MCP Server environment.
The tool creates a new copy of an existing project, which is a write operation that modifies the state of the GitHub account by adding new resources. While reversible (the copy can be deleted), it commits new data structures. This is less severe than Execute (no arbitrary code) or Destructive (reversible), but more impactful than simple Read operations.
From the tool's definition copy-project duplicates a GitHub Project, creating new project resources and associated metadata. This is a reversible write operation that creates new data without deletion.
Documented attack patterns abuse exactly the kind of access copy-project gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and GitHub Projects MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for copy-project:
{
"version": "1",
"default": "deny",
"tools": {
"copy-project": {
"limits": [
{
"counter": "copy-project_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}copy-project 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.
Copy a GitHub Project. It is categorised as a Write tool in the GitHub Projects MCP Server MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the GitHub Projects MCP Server MCP server in PolicyLayer and add a rule for copy-project: 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 GitHub Projects MCP Server. Nothing to install.
copy-project 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 copy-project 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 copy-project. 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.
copy-project is provided by the GitHub Projects MCP Server MCP server (taylor-lindores-reeves/mcp-github-projects). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 21 GitHub Projects MCP Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
21 GitHub Projects MCP Server tools catalogued and risk-classified — across an index of 42,500+ MCP servers.