investigate
Investigate a company, person, or topic and return a concise sourced artifact. This is the default v3 entry point: one ask in, one report-shaped output out.
This record as markdown: /tools/io-github-homenshum-nodebench/investigate.md
What investigate does on Nodebench
AI agents call investigate to retrieve information from Nodebench without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Why investigate is rated Low
The tool performs research and information gathering (investigation) and returns a formatted report. The verb 'investigate' combined with the emphasis on returning sourced artifacts indicates a read-only query operation that retrieves and synthesizes existing information without modifying data, executing code, deleting resources, or committing financial transactions.
From the tool's definition Tool description states it 'Investigate a company, person, or topic and return a concise sourced artifact' — this is a retrieval and reporting operation with 'one ask in, one report-shaped output out', indicating data querying with no modification or side…
Attacks that exploit this kind of access
The rule that runs investigate safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Nodebench, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For investigate, this is the rule to start with:
investigate 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 Nodebench, apply this rule, and every investigate call is checked against it from then on.
Questions about investigate
Investigate a company, person, or topic and return a concise sourced artifact. This is the default v3 entry point: one ask in, one report-shaped output out. It is categorised as a Read tool in the Nodebench MCP Server, which means it retrieves data without modifying state.
Register the Nodebench MCP server in PolicyLayer and add a rule for investigate: 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 Nodebench. Nothing to install.
investigate 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 investigate 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 investigate. 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.
investigate is provided by the Nodebench MCP server (nodebench-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Nodebench, and thousands of servers like it.
This server
Across the catalogue