Returns process.cwd()
AI agents call getCwd to retrieve information from Mcp Registry Registry without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool only retrieves and returns information (the current working directory) without modifying, executing operations, or having destructive effects. It is a pure information retrieval operation, fitting the Read category with low severity since directory path disclosure is generally low-impact unless sensitive environment information is exposed.
From the tool's definition Tool name 'getCwd' and description 'Returns process.cwd()' indicate it retrieves the current working directory of the process—a read-only query with no side effects.
Attacks that exploit this kind of access
Returns process.cwd(). It is categorised as a Read tool in the Mcp Registry Registry MCP Server, which means it retrieves data without modifying state.
Register the Mcp Registry Registry MCP server in PolicyLayer and add a rule for getCwd: 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 Mcp Registry Registry. Nothing to install.
getCwd is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the getCwd 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 getCwd. 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.
getCwd is provided by the Mcp Registry Registry MCP server (@mastra/mcp-registry-registry). 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 →