Skip to content

T27 OS: matte glass everywhere in a window, black in full screen, panels narrow the terminal, KEYS foot rows, terminal clicks stay in the terminal - #11266

Merged
gHashTag merged 4 commits into
masterfrom
claude/t27os-glass-all
Oct 20, 2026
Merged

gHashTag merged 4 commits into
masterfrom
claude/t27os-glass-all

Conversation

@gHashTag

Copy link
Copy Markdown
Owner

Closes #11262

T27 OS: matte glass everywhere in a window, black in full screen, panels narrow the terminal, KEYS foot rows, terminal clicks stay in the terminal

Gates run on the merged tree (t27c from master): 2 spec(s) with tests clean and vacuous passes 0, seal --verify MATCH, tri types ratchet CLEAN, dupe_scan failure set a subset of master's (inherited red).

TEST specs/t27os/editors.t27 tests=[30] fail=[0] vacuous=[0]
TEST specs/t27os/term_pane.t27 tests=[10] fail=[0] vacuous=[0]
SEAL specs/t27os/editors.t27 :: all hashes MATCH
SEAL specs/t27os/term_pane.t27 :: all hashes MATCH
TYPES RATCHET: CLEAN
DUPE failing=4 extra-vs-master=[]

🤖 Generated with Claude Code

claude and others added 4 commits October 20, 2026 14:25
…k in full screen; the terminal ends at an open panel; the KEYS foot has a row per line

surface_alpha(kind, full) decides the fill of the sidebar, content, terminal and panel
surfaces: a darker tint over the behind-window glass in a window (0.45, 0.35, 0.15 over the
content, 0.58), opaque #000000 in full screen. The terminal's default background is the
terminal tint, not opaque black; the honeycomb cards are not drawn under a terminal pane.
The Queen, DOCTOR and KEYS panels share one ground (wiped to the glass and tinted), the
button plates are the same glass, and the sidebar foot is the sidebar's glass again.
The terminal grid is laid out to the open panel's left edge (content_w), so the pty gets the
narrower size through the existing resize path and nothing of it is under the panel.
The KEYS foot is the status row, the OPEN IN row and the chips in rows of three, each in its
own band computed from the chip count (editors.t27 foot_*); chip rows are 14 pt so six chips
fit under nine providers at 960 x 540.

Refs #10782

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
…al or control under it answers

While a Claude Code pane or the editor view covers the content area, the click handler
still asked card_at, so a click on the terminal's input row landed on a hidden honeycomb
cell and opened or switched an agent chat. term_up / on_term / term_click decide that a click
inside the terminal's rectangle belongs to the pane (its cursor goes solid) and the chain
below never sees it; card_live answers no card while a pane is up, in the click and under
the cursor. Every other control of the chain (title bar, header toggle, Close Dashboard,
"+", the compact dashboard bar, the sidebar rows and foot links, a right-hand panel) lies
outside the rectangle: controls_under_term samples it in six layouts and counts none.

Refs #10782

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
…) into claude/t27os-glass-all

glass_window.t27 keeps both: the CHAT petal and the home tabs of #11261, and the glass
surfaces, the panel-narrowed terminal, the KEYS foot rows and the terminal-click guard.
Routed through the glass rules: draw_home's content fill is surface_alpha(SURF_CONTENT, full)
(glass in a window, opaque black in full screen). The home tabs have no hit test that could lie
under a terminal: home_card_at only lights a hover, and a terminal pane is up only on the
chat screen (term_up takes `screen`, which is now chat-only).

Refs #11262

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
…els narrow the terminal, KEYS foot rows, terminal clicks stay in the terminal (Closes #11262)

Merge of master plus sealing, gates run on the merged tree with t27c from master: tests clean with vacuous passes 0, seal --verify MATCH, tri types ratchet CLEAN, dupe_scan failure set a subset of master's.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@gHashTag
gHashTag enabled auto-merge (squash) October 20, 2026 07:42
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-20 07:42:54 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 47
PRs with All Checks Green 3
READY 2
FAILING 47
PENDING 0
NO CHECKS YET 0

These columns do not partition: 2 + 47 + 0 + 0 = 49, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=5aec047d42ec != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

@gHashTag
gHashTag merged commit 10eeeb5 into master Oct 20, 2026
31 of 36 checks passed
gHashTag pushed a commit that referenced this pull request Oct 20, 2026
Master took the glass PR (matte glass, panels narrow the terminal, KEYS foot
rows, the chip click test following those rows). It merges without a
conflict: this branch changes the KEYS row text (status text, add/full word,
profile line) and the click and key handlers; the glass lane changed the
painting and layout, including the foot rows below the key rows.

Refs #11265

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

T27 OS: matte glass on every surface in a window, black in full screen; panels narrow the terminal; KEYS foot rows; terminal clicks stay in the terminal

2 participants