AUTOMATICALLY send urgent notifications when user input, decision, or immediate attention is required. Claude should proactively use this for: file conflicts, permission requests, critical errors requiring user choice, or when waiting for user response before proceeding. Essential for good UX - n...
AI agents use send_user_action_needed_notification to create or update resources in Claude Code Notification Hooks — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Claude Code Notification Hooks environment.
This tool writes notification data to the user's system—each invocation creates a new notification object in the OS notification queue. While notifications are reversible (can be dismissed) and have no persistent data modification, the action itself modifies the user's notification state and system UI.
From the tool's definition Tool AUTOMATICALLY sends notifications to the user system. The description states it will 'send urgent notifications' and 'proactively use this for' various conditions, indicating it writes/creates notification objects to the user's desktop notification…
Attacks that exploit this kind of access
AUTOMATICALLY send urgent notifications when user input, decision, or immediate attention is required. Claude should proactively use this for: file conflicts, permission requests, critical errors requiring user choice, or when waiting for user response before proceeding. Essential for good UX - notify users immediately when blocked. It is categorised as a Write tool in the Claude Code Notification Hooks MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Claude Code Notification Hooks MCP server in PolicyLayer and add a rule for send_user_action_needed_notification: 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 Claude Code Notification Hooks. Nothing to install.
send_user_action_needed_notification 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 send_user_action_needed_notification 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 send_user_action_needed_notification. 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.
send_user_action_needed_notification is provided by the Claude Code Notification Hooks MCP server (nkyy/claude-code-notify-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 →