dive_preflight
Analyze a project BEFORE starting a UI dive. Scans the project directory to detect: framework (Vite, Next.js, CRA, etc.), dev scripts, required services (frontend, backend like Convex/Supabase/Firebase), port assignments, whether services are already running, route definitions from source code, a...
This record as markdown: /tools/io-github-homenshum-nodebench/dive-preflight.md
What dive_preflight does on Nodebench
AI agents call dive_preflight 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 dive_preflight is rated Low
dive_preflight is purely an information-gathering and analysis tool. It scans project files and configuration to detect framework, scripts, services, and routes, then returns structured data (a launch plan). No data is modified, no code is executed, and no external operations are triggered — it only reads and analyzes existing project state to prepare the agent for later steps.
From the tool's definition The tool 'Analyze a project BEFORE starting a UI dive. Scans the project directory to detect: framework...port assignments, whether services are already running, route definitions from source code, and environment requirements.
Attacks that exploit this kind of access
The rule that runs dive_preflight 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 dive_preflight, this is the rule to start with:
dive_preflight 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 dive_preflight call is checked against it from then on.
Questions about dive_preflight
Analyze a project BEFORE starting a UI dive. Scans the project directory to detect: framework (Vite, Next.js, CRA, etc.), dev scripts, required services (frontend, backend like Convex/Supabase/Firebase), port assignments, whether services are already running, route definitions from source code, and environment requirements. Returns a structured launch plan the agent should follow to get the app running before navigating. This is always Step 0 of a dive. 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 dive_preflight: 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.
dive_preflight 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 dive_preflight 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 dive_preflight. 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.
dive_preflight 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