web_extract
Scrapes structured data from the current page. Pass selectors = an object mapping field names to CSS selectors (e.g. {"title":"h1","price":".price"}); returns each field's first-match text/href, null when absent.
This record as markdown: /tools/com-local-mcp-local-mcp/web-extract.md
What web_extract does on Local
AI agents call web_extract to retrieve information from Local without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
| Parameter | Type | Required | Description |
|---|---|---|---|
session | string | — | Session name (default 'default'). |
selectors | object | Yes | Field name → CSS selector map. |
Parameters from the server's own tool schema.
Why web_extract is rated Low
The web_extract tool performs data retrieval and extraction from web pages without modifying, deleting, or executing operations. It maps to the Read category (retrieves or queries data; no side effects). Severity is low because data extraction alone poses minimal risk—the blast radius depends entirely on what the extracted data is used for downstream, which is outside this tool's scope.
From the tool's definition Tool description states it 'Scrapes structured data from the current page' and 'returns each field's first-match text/href' — a read-only operation with no side effects. It uses CSS selectors to extract data, which is a passive querying mechanism.
Attacks that exploit this kind of access
The rule that runs web_extract safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Local, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For web_extract, this is the rule to start with:
web_extract is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Local, apply this rule, and every web_extract call is checked against it from then on.
Questions about web_extract
Scrapes structured data from the current page. Pass selectors = an object mapping field names to CSS selectors (e.g. {"title":"h1","price":".price"}); returns each field's first-match text/href, null when absent. It is categorised as a Read tool in the Local MCP Server, which means it retrieves data without modifying state.
web_extract accepts 2 parameters: session, selectors. Required: selectors. The full parameter table on this page comes from the server's own tool schema.
Register the Local MCP server in PolicyLayer and add a rule for web_extract: 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 Local. Nothing to install.
web_extract 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 web_extract 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 web_extract. 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.
web_extract is provided by the Local MCP server (local-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Local, and thousands of servers like it.
This server
Across the catalogue