Extract visible text content from the current Safari page, or from a specific CSS selector. Returns clean text stripped of scripts/styles. Useful for reading article content, scraping data, or verifying page state without needing complex safari_js code. Use timeout param with selector to wait for...
AI agents call get_page_text to retrieve information from MacWright without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs read-only operations on web page content. It retrieves and formats text from the DOM without modifying any state, executing code, or causing side effects beyond passive observation. The mention of 'scraping data' is a retrieval use case, not a write or destructive action.
From the tool's definition Tool description states it 'Extract[s] visible text content' and is 'Useful for reading article content, scraping data, or verifying page state'.
Attacks that exploit this kind of access
Extract visible text content from the current Safari page, or from a specific CSS selector. Returns clean text stripped of scripts/styles. Useful for reading article content, scraping data, or verifying page state without needing complex safari_js code. Use timeout param with selector to wait for dynamically-loaded content (combines wait_for_element + get_page_text in one call). Use section:true to extract an entire section from a heading to the next heading of the same level — great for Wikipedia, MDN, and other documentation pages. Use viewport:true to get only the text visible on screen right now (great for page-by-page reading after scroll_page). It is categorised as a Read tool in the MacWright MCP Server, which means it retrieves data without modifying state.
Register the MacWright MCP server in PolicyLayer and add a rule for get_page_text: 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 MacWright. Nothing to install.
get_page_text is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_page_text 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 get_page_text. 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.
get_page_text is provided by the MacWright MCP server (ruchit-p/macwright). 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 →