Corpus
The corpus holds 111 measured terminal screens of 33 agents. Each screen is a pair of files: the .screen file holds the exact bytes, and the .yaml file holds the metadata and the expected result.
Read the format of a screen for each field and each state.
- Antigravity CLI (26 screens)
- Aider (2 screens)
- Amp (1 screen)
- Auggie (1 screen)
- Claude Code (35 screens)
- Cline CLI (1 screen)
- CodeBuddy Code (1 screen)
- Codebuff (1 screen)
- Codex CLI (17 screens)
- Continue CLI (1 screen)
- Copilot CLI (1 screen)
- Crush (1 screen)
- Cursor CLI (1 screen)
- Factory Droid (1 screen)
- Every Code (1 screen)
- Gemini CLI (2 screens)
- Goose (1 screen)
- gptme (1 screen)
- Grok CLI (1 screen)
- iFlow CLI (1 screen)
- Open Interpreter (1 screen)
- Junie CLI (1 screen)
- Kilo Code CLI (2 screens)
- Kimi Code CLI (1 screen)
- Kiro CLI (1 screen)
- Letta Code (1 screen)
- Neovate Code (1 screen)
- opencode (1 screen)
- opencode 2 (1 screen)
- OpenHands CLI (1 screen)
- Qoder CLI (1 screen)
- Qwen Code (1 screen)
- Mistral Vibe (1 screen)
Antigravity CLI (corpus/agy)
Verified versions
- 1.4.5, 2026-10-24
fresh capture of 8 screens in a private tmux server, 120x40
Covers the folder-trust dialog, idle, typed, busy, a command permission dialog and the AskQuestion dialog. Not checked: the file permission dialog, the model picker, the command menu, the shortcuts help.
- 1.4.7, 2026-10-30
fresh capture of 6 screens in a private tmux server, 120x40, by scenarios/agy/deny-and-tools.sh and scenarios/agy/continue.sh
Covers idle, busy, the file permission dialog, the screen after "2. No, deny creation", the screen after Esc on the file dialog, and agy --continue in a folder with no conversation. Not checked on 1.4.7: the trust dialog, typed text, the command permission dialog, the question dialog, the model picker, the command menu, the shortcuts help.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
after-deny-v1-4-7 |
IDLE |
none |
1.4.7 | measured-edited | 41 |
after-escape-v1-4-7 |
IDLE |
none |
1.4.7 | measured-edited | 41 |
after-question-v1-4-5 |
IDLE |
none |
1.4.5 | measured-edited | none |
busy-v1-4-5 |
BUSY |
none |
1.4.5 | measured-edited | none |
busy-v1-4-7 |
BUSY |
none |
1.4.7 | measured-edited | 17, 45 |
busy |
BUSY |
none |
1.4.5 | measured-edited | 17 |
color-scheme |
UNKNOWN |
none given | 1.4.5 | measured-edited | 17 |
command-menu |
UNKNOWN |
none given | 1.4.5 | measured-edited | 18 |
continue-empty-v1-4-7 |
IDLE |
none |
1.4.7 | measured-edited | 43 |
file-permission-v1-4-7 |
PERMISSION |
permission |
1.4.7 | measured-edited | 4, 18, 41, 42 |
file-permission |
PERMISSION |
permission |
1.4.5 | measured-edited | 4, 18 |
idle-v1-4-5 |
IDLE |
none |
1.4.5 | measured-edited | none |
idle-v1-4-7 |
IDLE |
none |
1.4.7 | measured-edited | 17 |
idle |
IDLE |
none |
1.4.5 | measured-edited | 17 |
model-picker |
UNKNOWN |
none given | 1.4.5 | measured-edited | 18 |
permission-v1-4-5 |
PERMISSION |
permission |
1.4.5 | measured-edited | none |
permission |
PERMISSION |
permission |
1.4.5 | measured-edited | 4, 17 |
question-v1-4-5 |
PERMISSION |
question |
1.4.5 | measured-edited | none |
question |
PERMISSION |
question |
1.4.5 | measured-edited | 4, 18 |
running-command-v1-4-5 |
BUSY |
none |
1.4.5 | measured-edited | none |
shortcuts |
UNKNOWN |
none given | 1.4.5 | measured-edited | 18 |
terms |
UNKNOWN |
none given | 1.4.5 | measured-edited | 17 |
trust-v1-4-5 |
TRUST |
trust |
1.4.5 | measured-edited | none |
trust |
TRUST |
trust |
1.4.5 | measured-edited | 10, 17 |
typed-v1-4-5 |
IDLE |
none |
1.4.5 | measured-edited | none |
typed |
IDLE |
none |
1.4.5 | measured-edited | 17 |
Aider (corpus/aider)
Verified versions
- 0.86.2, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runners ubuntu-24.04 (x64) and ubuntu-22.04-arm (arm64)
Covers the git question of the first start (git-repo-question-v0-86-2), and the analytics question that ubuntu-22.04-arm gave in its place (analytics-question-v0-86-2). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
analytics-question-v0-86-2 |
NO_BOX |
none |
0.86.2 | measured | 30, 49 |
git-repo-question-v0-86-2 |
NO_BOX |
none |
0.86.2 | measured | 30 |
Amp (corpus/amp)
Verified versions
- 0.0.1793390435-g08e9ec, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (login-v0-0-1793390435-g08e9ec). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
login-v0-0-1793390435-g08e9ec |
NO_BOX |
none |
0.0.1793390435-g08e9ec | measured | 46 |
Auggie (corpus/auggie)
Verified versions
- 0.36.0, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (login-v0-36-0). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
login-v0-36-0 |
NO_BOX |
none |
0.36.0 | measured | 46 |
Claude Code (corpus/claude)
Verified versions
- 2.1.309, 2026-10-24
fresh capture of 14 screens in a private tmux server, 120x40
Covers idle, typed, busy, Bash and Write permission dialogs, the AskUserQuestion wizard (two tabs and the review), and a background shell. Not checked: the folder-trust dialog, Rewind, plan mode, long pasted text.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
after-ask-v2-1-309 |
IDLE |
none |
2.1.309 | measured-edited | none |
after-deny-v2-1-309 |
IDLE |
none |
2.1.309 | measured-edited | none |
ask-q1-v2-1-309 |
PERMISSION |
question |
2.1.309 | measured-edited | none |
ask-q2-v2-1-309 |
PERMISSION |
question |
2.1.309 | measured-edited | none |
ask-review-v2-1-309 |
PERMISSION |
review |
2.1.309 | measured-edited | none |
ask |
PERMISSION |
question |
2.1.263 | measured-edited | 3, 4 |
background-shell-v2-1-309 |
IDLE |
none |
2.1.309 | measured-edited | none |
busy-v2-1-309 |
BUSY |
none |
2.1.309 | measured-edited | none |
busy |
BUSY |
none |
2.1.226 | derived | 1 |
dialog-over-box |
DIALOG_OVER_BOX |
none given | 2.1.283 | measured-edited | 2 |
idle-after-turn-v2-1-309 |
IDLE |
none |
2.1.309 | measured-edited | none |
idle-default-v2-1-309 |
IDLE |
none |
2.1.309 | measured-edited | none |
idle-placeholder-auto-v2-1-309 |
IDLE |
none |
2.1.309 | measured-edited | none |
idle-shell-done-v2-1-309 |
IDLE |
none |
2.1.309 | measured-edited | none |
idle |
IDLE |
none |
2.1.226 | derived | 1 |
overlay-footer-without-options |
IDLE |
none |
2.1.283 | derived | 2 |
overlay-plain-list |
IDLE |
none |
2.1.283 | derived | 2 |
overlay-without-footer |
IDLE |
none |
2.1.283 | derived | 2 |
paste-and-text |
IDLE |
none given | 2.1.309 | derived | 13 |
paste-lines |
IDLE |
none given | 2.1.309 | derived | 13 |
paste |
IDLE |
none |
2.1.309 | measured-edited | 13 |
permission-bash-v2-1-309 |
PERMISSION |
permission |
2.1.309 | measured-edited | none |
permission-write-v2-1-309 |
PERMISSION |
permission |
2.1.309 | measured-edited | none |
permission |
PERMISSION |
permission |
2.1.226 | measured-edited | 1, 4 |
quoted-dialog |
DIALOG_OVER_BOX known false positive |
none given | 2.1.283 | derived | 2 |
review |
PERMISSION |
review |
2.1.263 | measured-edited | 3, 4 |
rewind |
UNKNOWN |
none given | 2.1.226 | measured-edited | 1 |
styled-placeholder |
IDLE |
none given | 2.1.309 | derived | 5 |
styled-suggestion |
IDLE |
none |
2.1.309 | measured-edited | 5 |
styled-typed-same |
IDLE |
none |
2.1.309 | derived | 5 |
suggestion |
IDLE |
none |
2.1.226 | derived | 5 |
trust-accept-selected |
TRUST |
trust |
2.1.251 | derived | 10 |
trust |
TRUST |
trust |
2.1.251 | measured-edited | 1, 10 |
typed-list |
IDLE |
none |
2.1.263 | derived | 1, 3 |
typed-v2-1-309 |
IDLE |
none |
2.1.309 | measured-edited | none |
Cline CLI (corpus/cline)
Verified versions
- 3.0.74, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (provider-v3-0-74). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
provider-v3-0-74 |
NO_BOX |
none |
3.0.74 | measured | 46 |
CodeBuddy Code (corpus/codebuddy)
Verified versions
- 2.179.0, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (trust-v2-179-0). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
trust-v2-179-0 |
TRUST |
trust |
2.179.0 | measured-edited | 61 |
Codebuff (corpus/codebuff)
Verified versions
- 1.0.688, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (login-v1-0-688). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
login-v1-0-688 |
NO_BOX |
none |
1.0.688 | measured | 61 |
Codex CLI (corpus/codex)
Verified versions
- 0.163.0, 2026-10-24
capture-pane -p of codex --no-daemon in a private tmux server, 120x40, from the Codex measurement report of the tg-bridge project (9 screens, empty lines removed by the report), and one aw doctor capture of the approval dialog
Covers idle, typed, busy, a busy turn with a background command, the end of a turn, the command approval dialog, the end of a cancelled approval, the folder-trust dialog and the hooks review dialog. Not checked: the file-change approval dialog, styled captures, the slash command menu, a pane narrower than 120 columns, a TUI that uses the shared app-server daemon, and the keys of the approval dialog other than Esc.
- 0.164.0, 2026-10-30
capture-pane -p of codex in the default daemon mode (no --no-daemon), 120x40, in a private tmux server with a throwaway HOME, by the scenario scenarios/codex/screens.sh
Covers idle, busy, the F2 Warnings view and the screen after Esc on it. The busy footer still shows "? for shortcuts". Not checked: styled captures, and a pane narrower than 120 columns.
- 0.164.0, 2026-10-30
capture-pane -p of codex in the default daemon mode, 120x40, in a private tmux server with a throwaway HOME, by the scenarios scenarios/codex/deny.sh and scenarios/codex/interrupt.sh
Covers the command approval dialog, the screen after the digit 3 ("No, and tell Codex what to do differently") on it, and the dialog "Task is still running" that Ctrl+C opens during a turn. Not checked: the file-change approval dialog, the hooks review dialog of this version (fact 38 gives its text), styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
after-cancel-v0-163-0 |
IDLE |
none |
0.163.0 | measured-edited | 20 |
after-deny-v0-164-0 |
IDLE |
none |
0.164.0 | measured-edited | 20, 36 |
after-turn-v0-163-0 |
IDLE |
none |
0.163.0 | measured-edited | 20 |
after-warnings-v0-164-0 |
IDLE |
none |
0.164.0 | measured-edited | 20 |
approval-full-v0-163-0 |
PERMISSION |
permission |
0.163.0 | measured-edited | 19, 20 |
approval-v0-163-0 |
PERMISSION |
permission |
0.163.0 | measured-edited | 19, 20 |
approval-v0-164-0 |
PERMISSION |
permission |
0.164.0 | measured | 20, 36 |
busy-background-v0-163-0 |
BUSY |
none |
0.163.0 | measured-edited | 20 |
busy-v0-163-0 |
BUSY |
none |
0.163.0 | measured-edited | 20 |
busy-v0-164-0 |
BUSY |
none |
0.164.0 | measured-edited | 20 |
hooks-review-v0-163-0 |
UNKNOWN |
none given | 0.163.0 | measured-edited | 20 |
idle-v0-163-0 |
IDLE |
none |
0.163.0 | measured-edited | 20 |
idle-v0-164-0 |
IDLE |
none |
0.164.0 | measured-edited | 20 |
task-running-v0-164-0 |
PERMISSION |
question |
0.164.0 | measured-edited | 37 |
trust-v0-163-0 |
TRUST |
trust |
0.163.0 | measured-edited | 20 |
typed-v0-163-0 |
UNKNOWN known false positive |
none |
0.163.0 | measured-edited | 20 |
warnings-v0-164-0 |
UNKNOWN |
none |
0.164.0 | measured-edited | none |
Continue CLI (corpus/continue)
Verified versions
- 1.5.47, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (api-key-v1-5-47). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
api-key-v1-5-47 |
NO_BOX |
none |
1.5.47 | measured | 46 |
Copilot CLI (corpus/copilot)
Verified versions
- 1.0.105, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (trust-v1-0-105). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
trust-v1-0-105 |
TRUST |
trust |
1.0.105 | measured-edited | 30 |
Crush (corpus/crush)
Verified versions
- 0.99.0, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (provider-v0-99-0). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
provider-v0-99-0 |
NO_BOX |
none |
0.99.0 | measured | 46 |
Cursor CLI (corpus/cursor)
Verified versions
- 2026.10.22-27dcf5f, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (login-v2026-10-22-27dcf5f). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
login-v2026-10-22-27dcf5f |
NO_BOX |
none |
2026.10.22-27dcf5f | measured | 29 |
Factory Droid (corpus/droid)
Verified versions
- 1.0.0, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (login-v1-0-0). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
login-v1-0-0 |
NO_BOX |
none |
1.0.0 | measured | 46 |
Every Code (corpus/everycode)
Verified versions
- 0.6.196, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-22.04 (x64)
Covers the first screen with no login (login-v0-6-196). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
login-v0-6-196 |
NO_BOX |
none |
0.6.196 | measured | 61 |
Gemini CLI (corpus/gemini)
Verified versions
- 0.66.0, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runners ubuntu-24.04 (x64) and ubuntu-22.04 (x64)
Covers the first screen with no login (trust-v0-66-0), and the same screen with the color warning that ubuntu-22.04 gave (trust-color-warning-v0-66-0). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
trust-color-warning-v0-66-0 |
TRUST |
trust |
0.66.0 | measured | 30, 56 |
trust-v0-66-0 |
TRUST |
trust |
0.66.0 | measured | 30 |
Goose (corpus/goose)
Verified versions
- 1.57.0, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (setup-v1-57-0). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
setup-v1-57-0 |
NO_BOX |
none |
1.57.0 | measured | 46 |
gptme (corpus/gptme)
Verified versions
- 0.35.1, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (conversation-v0-35-1). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
conversation-v0-35-1 |
NO_BOX |
none |
0.35.1 | measured | 61 |
Grok CLI (corpus/grok)
Verified versions
- 1.1.7, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (host-mode-v1-1-7). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
host-mode-v1-1-7 |
NO_BOX |
none |
1.1.7 | measured-edited | 61 |
iFlow CLI (corpus/iflow)
Verified versions
- 0.5.19, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (base-url-v0-5-19). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
base-url-v0-5-19 |
NO_BOX |
none |
0.5.19 | measured-edited | 61 |
Open Interpreter (corpus/interpreter)
Verified versions
- 0.0.56, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (login-v0-0-56). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
login-v0-0-56 |
NO_BOX |
none |
0.0.56 | measured | 61, 65 |
Junie CLI (corpus/junie)
Verified versions
- 26.10.27, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (statistics-v26-10-27). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
statistics-v26-10-27 |
NO_BOX |
none |
26.10.27 | measured | 61 |
Kilo Code CLI (corpus/kilo)
Verified versions
- 7.8.18, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runners ubuntu-24.04 (x64) and macos-26-intel (x64)
Covers the input box of the first start (idle-v7-8-18), and the empty pane that macos-26-intel gave at the same step (blank-v7-8-18). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
blank-v7-8-18 |
NO_BOX |
none |
7.8.18 | measured | 46, 57 |
idle-v7-8-18 |
IDLE |
none |
7.8.18 | measured-edited | 46 |
Kimi Code CLI (corpus/kimi)
Verified versions
- 2.4.1, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (trust-v2-4-1). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
trust-v2-4-1 |
TRUST |
trust |
2.4.1 | measured-edited | 46 |
Kiro CLI (corpus/kiro)
Verified versions
- 2.37.0, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (login-v2-37-0). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
login-v2-37-0 |
NO_BOX |
none |
2.37.0 | measured | 61 |
Letta Code (corpus/letta)
Verified versions
- 0.34.35, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (backend-v0-34-35). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
backend-v0-34-35 |
NO_BOX |
none |
0.34.35 | measured | 61 |
Neovate Code (corpus/neovate)
Verified versions
- 0.28.5, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (no-model-v0-28-5). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
no-model-v0-28-5 |
IDLE |
none |
0.28.5 | measured-edited | 61 |
opencode (corpus/opencode)
Verified versions
- 1.19.5, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (idle-v1-19-5). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
idle-v1-19-5 |
IDLE |
none |
1.19.5 | measured-edited | 29 |
opencode 2 (corpus/opencode2)
Verified versions
- 2.2.5, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (idle-v2-2-5). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
idle-v2-2-5 |
IDLE |
none |
2.2.5 | measured-edited | 63 |
OpenHands CLI (corpus/openhands)
Verified versions
- 1.16.0, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (settings-v1-16-0). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
settings-v1-16-0 |
UNKNOWN |
none |
1.16.0 | measured | 46 |
Qoder CLI (corpus/qoder)
Verified versions
- 1.1.81, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (login-v1-1-81). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
login-v1-1-81 |
NO_BOX |
none |
1.1.81 | measured | 61 |
Qwen Code (corpus/qwen)
Verified versions
- 0.26.4, 2026-10-30
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41142220844, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (provider-v0-26-4). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
provider-v0-26-4 |
NO_BOX |
none |
0.26.4 | measured | 30 |
Mistral Vibe (corpus/vibe)
Verified versions
- 3.0.4, 2026-10-31
capture-pane -p, 120x40, 20 s after the first start with no login, by scenarios/first-run.sh in the scenarios workflow run 41151440753, on the CI runner ubuntu-24.04 (x64)
Covers the first screen with no login (welcome-v3-0-4). Not checked: the screens after a login, any turn, a permission dialog, styled captures, and a pane narrower than 120 columns.
Screens
| Screen | State | Dialog | Version | Provenance | Facts |
|---|---|---|---|---|---|
welcome-v3-0-4 |
NO_BOX |
none |
3.0.4 | measured | 61 |