Skip to main content
Glama

inspect_networkStart

Idempotent

Start capturing network packets to monitor traffic, troubleshoot connectivity issues, and analyze data flows in real-time.

Instructions

Start network capture.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Implementation Reference

  • Main handler for 'inspect.networkStart' tool. Attaches debugger, initializes network capture state, enables CDP Network domain, and clears detach timer to keep debugger alive during capture.
    case 'inspect.networkStart': {
      // Start capturing network requests via CDP Network domain
      await ensureDebugger(tabId)
      const capture = { entries: [], listening: true }
      networkCaptures.set(tabId, capture)
      await chrome.debugger.sendCommand({ tabId }, 'Network.enable')
      // Clear detach timer — keep debugger alive for capture
      const session = debuggerSessions.get(tabId)
      if (session?.detachTimer) { clearTimeout(session.detachTimer); session.detachTimer = null }
      return {}
    }
  • CDP Network Event Listener that captures network responses when inspect.networkStart is active. Filters for API-like requests (JSON/XML) and stores response metadata in networkCaptures map.
    // --- CDP Network Event Listener (for inspect.networkStart/networkDump) ---
    
    chrome.debugger.onEvent.addListener((source, method, params) => {
      const tabId = source.tabId
      const capture = networkCaptures.get(tabId)
      if (!capture?.listening) return
    
      if (method === 'Network.responseReceived') {
        const { response, requestId } = params
        if (!response?.url) return
        // Only capture API-like requests (JSON, XHR, fetch)
        const ct = response.headers?.['content-type'] || response.headers?.['Content-Type'] || ''
        const isApi = ct.includes('json') || ct.includes('xml') || response.mimeType?.includes('json')
        const isDoc = ct.includes('html') || ct.includes('css') || ct.includes('javascript') || ct.includes('image') || ct.includes('font')
        if (isDoc && !isApi) return
        capture.entries.push({
          url: response.url,
          method: params.type || 'GET',
          status: response.status,
          type: response.mimeType || ct.split(';')[0],
          requestId,
        })
      }
    })
  • Tool registration in the capabilities response. Lists 'inspect.networkStart' as a supported tool alongside related inspect tools (inspect.page, inspect.networkDump, inspect.networkStop).
    'inspect.page', 'inspect.networkStart', 'inspect.networkDump', 'inspect.networkStop',
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

While annotations correctly declare the operation as non-destructive, non-read-only, idempotent, and open-world, the description adds no behavioral context about what format the capture uses (HAR, JSON), what traffic is included (HTTP, WebSocket, all tabs), or storage limitations. It relies entirely on structured fields.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is extremely brief (three words) with no filler or redundancy, earning high marks for efficiency. However, the brevity verges on under-specification given the tool's behavioral complexity, which penalizes it from a perfect score.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a stateful, open-world tool that initiates a capture session likely intended for later retrieval, the description is inadequate. It fails to mention the paired retrieval tool (inspect_networkDump), capture scope, or duration constraints, leaving critical operational gaps.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With zero parameters, this dimension receives the baseline score of 4. The empty schema requires no additional semantic explanation from the description.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the verb (Start) and resource (network capture), making the basic purpose understandable. However, it fails to distinguish from the sibling tool 'inspect_networkDump', which likely retrieves the captured data, leaving the agent unaware of the two-step workflow.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

There is no guidance on when to use this tool versus its sibling 'inspect_networkDump', nor any mention of prerequisites or the fact that capture continues until explicitly stopped or dumped. The agent receives no workflow context.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/LeonTing1010/tap'

If you have feedback or need assistance with the MCP directory API, please join our Discord server