AI agents call meshes_get_connection to retrieve information from Mesheshq without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves connection details (configuration, status, mappings) without modifying state or executing external operations. The absence of credential exposure further limits its risk. It falls clearly into the Read category with low severity since connection metadata is typically non-sensitive configuration data.
From the tool's definition Tool name 'meshes_get_connection' and description 'Get details of a specific connection' indicate a retrieval operation. The explicit note 'Does not expose raw credentials' confirms it returns metadata without sensitive secrets.
Attacks that exploit this kind of access
Get details of a specific connection. Does not expose raw credentials. It is categorised as a Read tool in the Mesheshq MCP Server, which means it retrieves data without modifying state.
Register the Mesheshq MCP server in PolicyLayer and add a rule for meshes_get_connection: 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 Mesheshq. Nothing to install.
meshes_get_connection 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 meshes_get_connection 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 meshes_get_connection. 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.
meshes_get_connection is provided by the Mesheshq MCP server (mesheshq/meshes-mcp-server). 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 →