AI agents use add_mcp_server to create or update resources in Snak — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Snak environment.
This tool modifies an agent's configuration by installing and configuring MCP servers. While reversible (as suggested by the sibling tool 'remove_mcp_server'), it represents a significant Write operation that alters agent capabilities and could expand the attack surface if a malicious MCP server is installed.
From the tool's definition Tool description states 'Add/install/configure one or multiple new MCP servers for a specific agent.' The verbs 'add', 'install', and 'configure' indicate modification of agent configuration and capabilities.
Attacks that exploit this kind of access
Add/install/configure one or multiple new MCP servers for a specific agent. Use when user wants to add MCP server capabilities to an existing agent. It is categorised as a Write tool in the Snak MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Snak MCP server in PolicyLayer and add a rule for add_mcp_server: 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 Snak. Nothing to install.
add_mcp_server 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 add_mcp_server 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 add_mcp_server. 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.
add_mcp_server is provided by the Snak MCP server (kasarlabs/snak). 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 →