Search X (formerly Twitter) in real-time using Grok's X Search. Supports structured output via JSON Schema and multi-turn chaining via response IDs. Use this tool when you need: - Real-time posts, trends, or public opinion from X - Fetching a specific post and its thread/replies by URL - Structu...
Single-target operation
Part of the Grok MCP server. Enforce policies on this tool with Intercept, the open-source MCP proxy.
AI agents call x_search to retrieve information from Grok 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 x_search 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:
x_search:
rules:
- action: allowSee the full Grok policy for all 1 tools.
Agents calling read-class tools like x_search 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.
Search X (formerly Twitter) in real-time using Grok's X Search. Supports structured output via JSON Schema and multi-turn chaining via response IDs. Use this tool when you need: - Real-time posts, trends, or public opinion from X - Fetching a specific post and its thread/replies by URL - Structured extraction of X data (topics, sentiment, reactions, etc.) - Follow-up searches that build on a previous result (drill-down, filtering, summarization) Workflow for deep research: 1. First call: use output_schema to get structured data + capture response_id 2. Follow-up calls: pass previous_response_id to continue with context Workflow for thread extraction: 1. Pass the post URL as prompt with output_schema to get structured thread data (original post + top replies ranked by engagement) 2. Use the returned structured data for analysis, commentary, or conversation. It is categorised as a Read tool in the Grok MCP Server, which means it retrieves data without modifying state.
Add a rule in your Intercept YAML policy under the tools section for x_search. You can allow, deny, rate-limit, or validate arguments. Then run Intercept as a proxy in front of the Grok MCP server.
x_search 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 x_search 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 x_search. 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.
x_search is provided by the Grok MCP server (grok-mcp-server). 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