Output formats

Every command renders through one output layer, so what you learn once applies everywhere. Set the format with -o/--output.

Formats

Format	Best for
list	Reading at a terminal (the default there): each record is a short section, a heading then its fields below.
table	Scanning one column down many rows: a rounded, aligned grid that shrinks to your terminal width.
markdown	A GitHub-flavored pipe table to paste into an issue, a PR, or a README.
jsonl	Pipelines; one JSON object per line (the default in a pipe).
json	A single JSON array, for tools that want one document.
csv	Spreadsheets and data tools.
tsv	Tab-separated, for cut and friends.
url	Just the URL column, one per line.
raw	The first column, unadorned (also used by --template).

The list and table views are two takes on the same rows. list puts one record in front of you at a time, which reads better when a record carries many fields; table lines many records up in a grid, which is better when you want to compare one column across rows. Reach for -o table for the latter.

auto

The default is auto: the readable list when stdout is a terminal, jsonl when it is a pipe or file. So the same command reads well by eye and composes well in a script with no extra flags:

wiki search "physics"            # a list on screen
wiki search "physics" | cat      # JSONL into the pipe

Color

On a terminal the output is colored: a bold pink heading and aligned cyan keys in list, a bold header and dimmed borders in table, and syntax-highlighted json/jsonl. Color is on for an interactive terminal and off when piped, so machine-read output stays plain. --color (auto|always|never) overrides it, and the NO_COLOR convention is honored. With color off, list emits literal GitHub-flavored markdown you can paste as-is.

Progress

A read can wait on the network before it has anything to show. When the terminal is interactive, wiki prints a small spinner to standard error while it waits and clears it the moment the first result is ready. It only ever writes to standard error, so a pipe like wiki search physics | jq and a redirect like wiki search physics > out.jsonl never see it; standard output stays clean. --quiet turns it off.

Picking columns

--fields selects and orders columns by name. It applies to the column-shaped formats (list, table, markdown, csv, tsv); the JSON formats always carry the full record:

wiki search "volcano" --fields title,description -o csv
wiki revisions "Pi" --fields revid,user,timestamp

--no-header drops the header row from csv, tsv, and markdown, and the section heading from list, which is handy when feeding another command:

wiki search "physics" -n 1 --fields title -o csv --no-header

Limiting results

-n/--limit caps how many rows come back. 0 means the API default:

wiki search "cat" -n 5
wiki category Physics -n 100

URLs

-o url prints just the URL of each row, perfect for piping into a fetcher or xargs:

wiki links "Alan Turing" -o url | head
wiki search "turing" -o url | xargs -I{} echo open {}

Templates

--template runs a Go text/template over each row's underlying value, for fully custom lines. It implies raw output:

wiki search "physics" --template '{{.Title}} => {{.URL}}'
wiki revisions "Pi" --template '{{.Timestamp}} {{.User}}'

The fields available are the JSON fields of each record; see the structured output of a command (-o json) to discover them.

Exit codes

Commands set a precise exit code so scripts can branch on the outcome. They are listed on the troubleshooting page.