hubcap network - Stream network requests and responses

When to use

Monitor network requests and responses in real time as NDJSON. Use har for standard HAR format output suitable for analysis tools. Use waitrequest or waitresponse to block until a specific network event occurs.

Usage

hubcap network [--duration <duration>]

Arguments

None.

Flags

Flag	Type	Default	Description
`--duration`	duration	`0`	How long to capture; 0 = until interrupted

Output

NDJSON stream written to stdout. Each line is a JSON object representing either a request or a response.

Field	Type	Description
`type`	string	`"request"` or `"response"`
`requestId`	string	Unique identifier for the request/response pair
`url`	string	Request URL
`method`	string	HTTP method (request lines only)
`status`	int	HTTP status code (response lines only)
`mimeType`	string	Response MIME type (response lines only)

Request line:

{"type":"request","requestId":"1000.1","method":"GET","url":"https://example.com/api/data"}

Response line:

{"type":"response","requestId":"1000.1","status":200,"mimeType":"application/json"}

Errors

Condition	Exit code	Stderr
Chrome not connected	2	`error: connecting to Chrome: ...`
Invalid flag value	1	`invalid value "..." for flag -duration: ...`

Examples

Stream all network activity until Ctrl-C:

hubcap network

Capture network activity for 10 seconds:

hubcap network --duration 10s

Filter for API requests using jq:

hubcap network --duration 30s | jq 'select(.type == "request" and (.url | contains("/api/")))'

Get the response body for a captured request:

hubcap network --duration 5s | jq -r 'select(.type == "response" and .status == 200) | .requestId' | head -1 | xargs hubcap responsebody