v0.3.1
What changed
docs: say each thing about the README once
Windows behaviour, the default-profile fallback rules, and opencode's XDG
layout were each explained in two or three places, which is two or three
places to keep true. Each now has one home that the others link to, and
the jq recipes that were copies of AGENTS.md give way to a pointer at it.
feat(indicator): keep the status line a profile already has
Claude Code draws one status line, so a profile that had its own could
either keep it or show the profile. Ditto chose theirs and said so, which
was the safe answer and never the one anybody wanted.
indicator --keep-mine now installs Ditto in front of it instead: the
command it writes carries theirs as an argument, runs it with the payload
Claude Code sent, and prints the profile ahead of every line it draws. A
command that fails or says nothing leaves the profile marker on its own,
--off reads the original back out and puts it in place untouched, and a
launch still never changes an arrangement it was not asked to change.
Two things around it that could take a status line away:
- Recognition was a command containing
ditto-cliand ending in
statusline, which claimed a script somebody called
ditto-cli-statuslineand then deleted it on--off. What Ditto writes
is the binary and the subcommand as separate words, so that is what is
recognised. - A status line set for a project or by an administrator outranks the
profile's, leaving Ditto's installed and never drawn.indicatorsays
shadowedrather than reporting it as on.
feat(settings): carry your own Claude Code settings into a profile
Snapshot of the uncommitted work from the other session, committed here so
the two lines of work can be merged rather than interleaved on disk.
Install
brew install reyanshgupta/tap/ditto-cli # or: brew upgrade ditto-cli
npm install -g @reyanshgupta/ditto-cli@0.3.1
cargo binstall ditto-cli # or: cargo install ditto-cliThe archives below are the same binaries every channel installs; SHA256SUMS covers them.
Full Changelog: v0.3.0...v0.3.1