Purpose: Perform web searches for a given objective and return results in an LLM-friendly format and with parameters tuned for LLMs. Ideal Use Cases: - For live queries that benefit from updated data from the web. - When needing to gather information from multiple sources at once or gathering de...
Single-target operation
Part of the Parallel Web Search MCP server. Enforce policies on this tool with Intercept, the open-source MCP proxy.
AI agents call web_search_preview to retrieve information from Parallel Web Search without modifying any data. This is common in research, monitoring, and reporting workflows where the agent needs context before taking action. Because read operations don't change state, they are generally safe to allow without restrictions -- but you may still want rate limits to control API costs.
Even though web_search_preview only reads data, uncontrolled read access can leak sensitive information or rack up API costs. An agent caught in a retry loop could make thousands of calls per minute. A rate limit gives you a safety net without blocking legitimate use.
Read-only tools are safe to allow by default. No rate limit needed unless you want to control costs.
tools:
web_search_preview:
rules:
- action: allowSee the full Parallel Web Search policy for all 1 tools.
Agents calling read-class tools like web_search_preview have been implicated in these attack patterns. Read the full case and prevention policy for each:
Other tools in the Read risk category across the catalogue. The same policy patterns (rate-limit, allow) apply to each.
Purpose: Perform web searches for a given objective and return results in an LLM-friendly format and with parameters tuned for LLMs. Ideal Use Cases: - For live queries that benefit from updated data from the web. - When needing to gather information from multiple sources at once or gathering detailed information from a single source. Performance Benefits: - Ability to handle complex queries more effectively. - Controllability in output size and sources considered. Examples: - Performing a broad search on a topic with multiple facets. - Enquiring about a specific detail on a topic. How to use: - For multi-hop queries, first try to use list to narrow down sources. Then use targeted with include_domains to get more detailed results. - If you have budget or multi turn capabilities, start with list searches first. . It is categorised as a Read tool in the Parallel Web Search MCP Server, which means it retrieves data without modifying state.
Add a rule in your Intercept YAML policy under the tools section for web_search_preview. You can allow, deny, rate-limit, or validate arguments. Then run Intercept as a proxy in front of the Parallel Web Search MCP server.
web_search_preview 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_search_preview rule in your Intercept 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 Intercept policy for web_search_preview. 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_search_preview is provided by the Parallel Web Search MCP server (parallel/search). Intercept sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Open source. One binary. Zero dependencies.
npx -y @policylayer/intercept