Apply the approved wording table: labels stay short, explanations are one
full sentence about the user's traffic, and detail moves behind an info
icon on the first-run card, the body banners, the Settings Helper row and
the rejected card. One word per thing: traffic and flows, recorded,
truncated, domains, searches, the CLI, Retry.
Backend messages that reach the window change with them, with the tests
and harness fixtures that assert on the old text. The two working files
for the pass are removed from the branch.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
A sort key can now name a header or query parameter column (`field:header.x-cache`). The
database sorts it by the value the column shows, read from the stored headers or URL with the
name bound as a parameter, and the app's in-memory sort places live flows the same way. Custom
column headers sort on click and Option-click and their menus' sort items work; the sort is
kept with the columns and removing a column drops it from the sort.
The Add Column field completes `@` and the paths a column accepts, with no colon and no values,
opens its menu on focus, normalises a pasted term, and says why a path cannot be added.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Captured traffic is now cleared as the app launches by default, with a
"Keep between launches" choice that brings the size limit with it. An
install that had a size limit set keeps its traffic. The Proxy settings
tab is regrouped into Proxy, Decryption, Traffic and Terminal, and the
per-body limit row is gone.
The app now watches macOS proxy settings and follows them: when another
app or the user points the system proxy elsewhere, the override shows as
off and the settings saved for restoring are dropped, so quitting puts
back nothing the app didn't set.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The app held every flow in memory and redid all of its work for each new
one: the sidebar mounted every domain and path node (its tree scrolled
inside a container that never bounded it), every tree row looked up its
atoms by scanning every atom the family had made, and the layout, detail
pane and list rows re-rendered per flow. At 3,000 flows each new one cost
about 90 ms of main thread, more than the traffic left time for.
The app now holds a window of the newest 2,000 flows, applies new ones in
batches, and loads older ones in pages from the database on scrolling to
the top, through a new list_flows RPC that pages in SQL. The database
keeps every flow.
Also adds clear_flows: an RPC, POST /flows/clear on the control API, and
a Clear button in the header that Cmd+K opens, confirming first.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
