AI agents call web_fetch to retrieve information from TOOL4LM without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves data from the web—a classic Read operation. It has no side effects on the target system or data. Severity is low because fetching public web content poses minimal risk; the main concern would be accessing sensitive URLs or overwhelming targets, but that is user-error rather than the tool's inherent design. Confidence is high because the name and server context clearly indicate data retrieval.
From the tool's definition Tool name is 'web_fetch' (alias of 'web.fetch'); server description states it 'enhances local LLMs with web search' and is for 'information retrieval'. The 'fetch' operation retrieves web content without modifying it.
Attacks that exploit this kind of access
Alias of web.fetch. It is categorised as a Read tool in the TOOL4LM MCP Server, which means it retrieves data without modifying state.
Register the TOOL4LM MCP server in PolicyLayer and add a rule for web_fetch: 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 TOOL4LM. Nothing to install.
web_fetch 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_fetch 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_fetch. 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_fetch is provided by the TOOL4LM MCP server (khanhs-234/tool4lm). 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 →