Type a stored credential into an input element BY REFERENCE. The actual credential value is never exposed to the model - it is resolved locally by the MCP server. Use browser_list_credentials to see available credential names.
AI agents invoke browser_type_credential to trigger actions in MCProxy. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool executes a browser action (typing credentials into an input field), which falls under the Execute category as it triggers external browser operations. The severity is high because it involves injecting stored credentials into web pages - if misused, an AI agent could authenticate into accounts, submit credentials to malicious sites, or facilitate account takeover.
From the tool's definition 'Type a stored credential into an input element BY REFERENCE' - triggers a browser action that inputs sensitive authentication credentials into a web page element, constituting an external operation with security implications
Attacks that exploit this kind of access
Type a stored credential into an input element BY REFERENCE. The actual credential value is never exposed to the model - it is resolved locally by the MCP server. Use browser_list_credentials to see available credential names. It is categorised as a Execute tool in the MCProxy MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the MCProxy MCP server in PolicyLayer and add a rule for browser_type_credential: 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 MCProxy. Nothing to install.
browser_type_credential is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the browser_type_credential 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 browser_type_credential. 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.
browser_type_credential is provided by the MCProxy MCP server (saladtechnologies/mcproxy). 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 →