Installs a Chrome extension from the given path.
AI agents invoke install_extension to trigger actions in Chrome Devtools. 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.
Installing a browser extension executes arbitrary code with broad browser permissions (access to all tabs, network requests, DOM, cookies, etc.). An attacker could install a malicious extension that exfiltrates credentials, intercepts traffic, or performs actions across all sites. This is an external operation with significant blast radius, making it Execute at critical severity.
From the tool's definition Installs a Chrome extension from the given path
Attacks that exploit this kind of access
Installs a Chrome extension from the given path. It is categorised as a Execute tool in the Chrome Devtools MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Chrome Devtools MCP server in PolicyLayer and add a rule for install_extension: 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 Chrome Devtools. Nothing to install.
install_extension 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 install_extension 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 install_extension. 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.
install_extension is provided by the Chrome Devtools MCP server (shivamprasad99/chrome-devtools-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 →