agentgui 1.0.1017 → 1.0.1019

package/AGENTS.md

@@ -1,5 +1,9 @@
 # AgentGUI — Agent Notes
 
+## Mobile sessions-drawer sliver fix (2026-07-03) — twenty-ninth run
+
+Scoped negative-space/attention follow-up (not a full agent fan-out) to the explicit "looks very diy" feedback. Live-browser screenshot + DOM measurement at 420px found a real regression the prior two runs' screenshot audits missed: the closed `.ws-sessions` drawer left a ~35px sliver visibly overlapping the chat content on mobile. Root cause in `../design` `app-shell.css`: the `<=1100px` breakpoint's closed-state `transform: translateX(-110%)` assumes the drawer's `left` is `0`; the `<=900px` breakpoint then shifts the drawer's `left` to `var(--ws-rail-w-collapsed)` (60px) so it sits right of the icon-only rail, but never adjusted the closed-state translate to match — `-110%` of a 248px-wide drawer is only -272.8px, undershooting the needed -308px (left + width) by ~35px. Fixed by overriding the closed-state transform inside the 900px block: `translateX(calc(-100% - var(--ws-rail-w-collapsed)))`, which scales correctly with both the drawer width and the rail offset. Verified via `getBoundingClientRect` before (x:-212.8, rightEdge:35.2 — visible) and after (x:-248, rightEdge:0 — fully clear) plus before/after screenshots. **Standing rule: any drawer/overlay whose `left`/`inset` changes across breakpoints must re-derive its closed-state transform from that breakpoint's own offset — a transform tuned for one breakpoint's geometry silently breaks at the next if the offset shifts underneath it.** Kit pushed `76b2bac`, agentgui pushed `4bb9f3f`.
+
 ## Visual-polish pass — shell depth, composer weight, chat gutter (2026-07-03) — twenty-eighth run
 
 User feedback: the shipped GUI "looks very diy". A live-browser screenshot audit (not a full agent fan-out) named 3 concrete, high-leverage causes and fixed all 3 in the kit. (1) `.ws-sessions` used flat `--bg`, identical to the main content column — `.ws-rail` and `.ws-pane` both already use `--bg-2` (a recessed plane framing the elevated content column) for exactly this reason; sessions was the one column breaking that pattern, so the conversation list and chat thread read as ONE continuous surface with only a 1px hairline between them — the single flattest tell in the shell. Now consistent. (2) `.chat-composer` — the box the user types into every turn — had zero elevation at rest, a flat hairline box on a flat background; added the kit's own already-defined-but-unused `--shadow-1` token (the existing focus-within ring still overrides it on focus). (3) The chat tab opts out of `.ws-main`'s padding via `mainFlush` (so the thread can center at `--measure`), but that left the controls row, heading, empty-state, and composer all flush against the column edge on both sides with zero gutter — the crumb bar above kept its own padding, so everything below visibly touched the edge the crumb text did not. Added a uniform inline gutter to `.agentchat` itself; `--measure` centering is unaffected (it centers within the now-padded box). **Standing rule: `.ws-rail`/`.ws-sessions`/`.ws-pane` must all three use `--bg-2` for consistent shell depth — a future column added flat `--bg` would repeat this exact regression.** Also landed inherited uncommitted community-app reply/delete-message work sitting dirty in the kit tree. Browser-witnessed at 1280px and 420px before/after: visible depth/weight/gutter improvement, no wrap/overlap regression, 0 console errors. Kit pushed `anentrypoint-design` `53c8464`, agentgui pushed `42708ea`. `agentgui`'s Deploy GH Pages flaked once (the known "Deployment cancelled" pattern — see the 26th-run note) and succeeded on rerun after patient wait.
@@ -24,7 +28,7 @@ Landed the highest-severity finding deferred from the 24th run (gui-completion #
 
 ## GUI perceived-performance sweep (2026-07-02) — twenty-third run
 
-Small, targeted run landing the 22nd-run's deferred perceived-perf findings. `index.html` now `modulepreload`s the marked/dompurify ESM CDN modules (was preconnect-only) so the fetch starts in parallel with the kit/app module graph instead of waiting for `markdown.js`'s dynamic import on the first chat response's escaped-text-to-rendered-markdown swap. Settings server panel shows "checking…" instead of "unknown" for version/uptime/clients/projects-folder during the very first health probe (new `state.healthChecking` flag around `recheckHealth`'s await — "unknown" reads as a permanent broken-server fact, not in-flight loading). Settings agents panel title shows "agents · loading…" instead of the literal "agents · 0/0 installed" while the initial fetch is in flight. Still deferred (its own PRD row, re-scoped not dropped): first-render network-gating — the shell paints nothing between the boot-splash and health-probe completion, needing a render-architecture change (skeleton/optimistic shell) bigger than this run's slice. No kit changes this run. agentgui pushed `6a045ba`, 28/28 tests, 0 console errors.
+Landed the 22nd-run's deferred perceived-perf findings (modulepreload for markdown CDN modules, "checking…"/"loading…" in-flight states instead of "unknown"/"0/0"); first-render network-gating re-scoped, not dropped. Full detail in rs-learn (recall "agentgui 23rd run perceived-performance sweep"). agentgui pushed `6a045ba`.
 
 ## GUI ux-craft sweep (2026-07-02) — twenty-second run
 

package/CHANGELOG.md

@@ -1,3 +1,8 @@
+## [Unreleased] - Mobile sessions-drawer sliver fix (kit 76b2bac, app 4bb9f3f)
+
+- kit: fixed the closed sessions drawer's `translateX(-110%)` at the <=900px breakpoint, which was sized for a `left:0` drawer and undershot once the drawer's left offset shifted to the collapsed rail width, leaving a ~35px sliver of the closed drawer visibly overlapping content on mobile. Why: user-reported "looks very diy" prompted a live-browser negative-space review that caught the regression via DOM measurement (`getBoundingClientRect` showed the drawer's right edge at +35px instead of 0/off-screen).
+- app: re-vendored the fixed kit build into `site/app/vendor/anentrypoint-design/`.
+
 ## [Unreleased] - GUI cohesion sweep: one-product chat-agent feel (kit 9ef409a, app 15b626c)
 
 - new workflow `.claude/workflows/gui-cohesion.js` (6 cohesion lenses -> adversarial verify -> ordered plan; 61 agents -> 40 confirmed gaps, `PUNCHLIST-COHESION.md`) — hunts the product-cohesion/feel gap vs claude.ai/code / cowork / Claude-Desktop, distinct from the audit/overhaul/predictability-polish workflows. Why: the shipped surfaces still read as a kit of parts.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "agentgui",
-  "version": "1.0.1017",
+  "version": "1.0.1019",
   "description": "Multi-agent ACP client with real-time communication",
   "type": "module",
   "main": "electron/main.js",

package/site/app/js/app.js

@@ -815,7 +815,16 @@ function sessionsColumn() {
       onInput: (v) => { state.searchQ = v; if (v.trim().length >= 2) debouncedSearch(); else { state.searchHits = null; } render(); },
     },
     onNew: () => { navTo('chat'); newChat(); },
-    onSelect: (s) => { if (state.tab === 'chat') resumeInChat({ sid: s.sid }); else loadSession(s.sid); },
+    // On files/live/settings the rail still renders (keepSessionsTrack) and is
+    // captioned 'Browse a conversation's events', but loadSession() alone only
+    // populates state - nothing on those tabs renders events, so the click had
+    // zero visible effect. Land on history first so the caption's promise
+    // (browsing events) is kept from every tab the rail appears on.
+    onSelect: (s) => {
+      if (state.tab === 'chat') { resumeInChat({ sid: s.sid }); return; }
+      if (state.tab !== 'history') navTo('history', { writeHash: false });
+      loadSession(s.sid);
+    },
     loading: state.tab === 'history' && !state.sessions.length && !state.historyError,
     loadingText: state.historySlow ? 'Indexing your Claude history — the first load can take a minute…' : undefined,
     error: state.historyError,