Install a new MCP server in obot that runs as
AI agents use add_npx_mcp to create or update resources in Obot Admin — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Obot Admin environment.
This tool installs (creates) a new MCP server configuration in the obot gateway. It is a Write operation as it adds a new resource. Severity is high because adding a new MCP server could introduce untrusted or malicious servers into the gateway, expanding the attack surface significantly. Confidence is slightly reduced because the description is truncated.
From the tool's definition 'Install a new MCP server in obot that runs as' — creates/installs a new MCP server resource
Attacks that exploit this kind of access
Install a new MCP server in obot that runs as. It is categorised as a Write tool in the Obot Admin MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Obot Admin MCP server in PolicyLayer and add a rule for add_npx_mcp: 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 Obot Admin. Nothing to install.
add_npx_mcp 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_npx_mcp 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_npx_mcp. 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_npx_mcp is provided by the Obot Admin MCP server (kiarashedraki/obot-admin-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 →