Search content

grep

Group: Files and search
Turned on by: Its own switch on the Tools page of Conversation settings.
Approval: Not reviewed: runs without asking, within the limits on this page.
Profile key: tool.grep.description

Searches file contents with a regular expression and returns each matching line with its file and line number.

Matching ignores case by default and works one line at a time, so a pattern cannot span lines. The syntax has no look-around and no backreferences.

Limits

Files over 2 MiB, binary files and links are skipped. Each reported line is cut at 500 characters.
A result over 20,000 characters is saved to a file in Mework's app-data folder. The model gets the first 2,000 characters and the file's path.

It searches what Git would show: tracked files, and untracked files Git does not ignore. Outside a Git repository, dependency and build folders such as node_modules and target are skipped. A single file named as path is always searched.
You can change a recorded grep call in the timeline and run it again with Rerun.

Parameter	Type	Details
`pattern` required Search pattern	string	Regular expression to match against each line.
`case_sensitive` Case sensitive	boolean	Match case-sensitively. Default `false`
`limit` Limit	integer	Most matching lines to return. Default `250`
`offset` Offset	integer	Matching lines to skip first, to fetch the next page. Default `0`
`path` Path	string	File or directory to search, relative to the workspace. Default `"."`