feat(cli): recognise Shift+Enter as a newline key

Closes #5346.

Most terminals send the same byte sequence for `Enter` and `Shift+Enter`
by default, so the application can't tell them apart — this is a terminal
protocol limitation, not something Hermes can paper over. But terminals
that implement the Kitty keyboard protocol (Kitty / foot / WezTerm /
Ghostty by default; iTerm2 / Alacritty / VS Code terminal / Warp once the
protocol is enabled) DO emit a distinct sequence for `Shift+Enter`:

  - `\x1b[13;2u`     — Kitty / CSI-u, modifier=2
  - `\x1b[27;2;13~`  — xterm modifyOtherKeys=2

Stock prompt_toolkit doesn't have the CSI-u sequence in its
`ANSI_SEQUENCES` table at all, and it maps the modifyOtherKeys variant to
plain `Keys.ControlM` (Enter) — i.e. it strips the Shift modifier, which
is the bug users actually hit on iTerm2 and friends.

This PR adds `hermes_cli/pt_input_extras.install_shift_enter_alias()`,
called once at CLI startup from `cli.py`, which inserts/overwrites those
sequences in `ANSI_SEQUENCES` so they decode to `(Keys.Escape, Keys.ControlM)`
— the same key tuple `Alt+Enter` produces. The existing Alt+Enter newline
handler (`@kb.add('escape', 'enter')` in `cli.py`) then fires unchanged,
so there is no new keybinding to register and no behavioral change for
terminals that don't emit the distinct sequences.

Files
=====

* `hermes_cli/pt_input_extras.py` — new module hosting the helper. Lives
  outside `cli.py` so it's importable in tests without dragging in the
  full CLI runtime (which depends on `fire`, `rich`, etc.).
* `cli.py` — calls `install_shift_enter_alias()` once at module import.
  Wrapped in try/except so prompt_toolkit version drift can't break CLI
  startup.
* `tests/cli/test_cli_shift_enter_newline.py` — 6 tests:
  - registration of all three byte sequences
  - overwrite of stock prompt_toolkit's broken modifyOtherKeys mapping
  - idempotency
  - parser equivalence: CSI-u Shift+Enter == Alt+Enter
  - parser equivalence: modifyOtherKeys Shift+Enter == Alt+Enter
  - plain Enter remains a single key (submit), distinct from the two-key
    Alt+Enter / Shift+Enter tuple
* `website/docs/user-guide/cli.md` — keybinding table updated; new
  "Shift+Enter compatibility" subsection with a per-terminal status table
  noting macOS Terminal / stock Windows Terminal cannot distinguish the
  keystroke at the protocol level.
* `website/docs/getting-started/quickstart.md`,
  `website/docs/guides/tips.md` — short mention pointing readers at the
  full compatibility note in `cli.md`.

Tested
======

  pytest tests/cli/test_cli_shift_enter_newline.py        # 6 passed

Live-tested by triggering `\x1b[13;2u` against the running Vt100Parser
(see test). Not exercised in a real terminal end-to-end because that
requires a Kitty-protocol-capable host; the test exercises the parser
path that drives the live terminal too.
This commit is contained in:
Syed Abdur Rehman Ali 2026-05-08 04:03:45 +05:30 committed by Teknium
parent cacb984732
commit f5b635f6ab
6 changed files with 163 additions and 5 deletions

7
cli.py
View file

@ -70,6 +70,13 @@ try:
_STEADY_CURSOR = CursorShape.BLOCK # Non-blinking block cursor
except (ImportError, AttributeError):
_STEADY_CURSOR = None
try:
from hermes_cli.pt_input_extras import install_shift_enter_alias
install_shift_enter_alias()
del install_shift_enter_alias
except Exception:
pass
import threading
import queue

View file

@ -0,0 +1,51 @@
"""Augmentations to prompt_toolkit's input-parsing tables.
Imported once at CLI startup. Each helper installs a small mapping into
prompt_toolkit's `ANSI_SEQUENCES` so byte sequences emitted by modern
keyboard protocols (Kitty / xterm `modifyOtherKeys`) decode to existing
key tuples Hermes already binds.
Kept in a standalone module separate from `cli.py` so the registrations
can be unit-tested without importing the whole CLI runtime.
"""
from __future__ import annotations
def install_shift_enter_alias() -> int:
"""Map Shift+Enter byte sequences to the (Escape, ControlM) key tuple
that Alt+Enter produces, so the existing Alt+Enter newline handler
fires for terminals that emit a distinct Shift+Enter.
Sequences mapped:
- "\\x1b[13;2u" Kitty keyboard protocol / CSI-u, modifier=2 (Shift)
- "\\x1b[27;2;13~" xterm modifyOtherKeys=2, modifier=2 (Shift)
- "\\x1b[27;2;13u" alternate ordering some emitters use
The CSI-u sequence is not in stock prompt_toolkit. The modifyOtherKeys
variant `\\x1b[27;2;13~` IS in stock prompt_toolkit but mapped to plain
`Keys.ControlM` i.e. Shift+Enter behaves identically to Enter, which
is the very bug this helper exists to fix. We therefore overwrite
those two specific keys (and `\\x1b[27;2;13u`) unconditionally; other
`\\x1b[27;...;13~` sequences (Ctrl+Enter, Alt+Enter via modifyOtherKeys
variants 5/6/etc.) are left untouched.
Default macOS Terminal and stock Windows Terminal still send the same
byte for Enter and Shift+Enter, so there is no fix for those terminals
at the application layer the sequences above never reach Hermes.
Returns the number of sequences whose mapping was changed.
"""
try:
from prompt_toolkit.input.ansi_escape_sequences import ANSI_SEQUENCES
from prompt_toolkit.keys import Keys
except Exception:
return 0
alt_enter = (Keys.Escape, Keys.ControlM)
changed = 0
for seq in ("\x1b[13;2u", "\x1b[27;2;13~", "\x1b[27;2;13u"):
if ANSI_SEQUENCES.get(seq) != alt_enter:
ANSI_SEQUENCES[seq] = alt_enter
changed += 1
return changed

View file

@ -0,0 +1,88 @@
"""Verify Shift+Enter byte sequences parse to the same key tuple Alt+Enter
produces, so the existing Alt+Enter newline handler in `cli.py` fires for
terminals that emit a distinct Shift+Enter under the Kitty keyboard protocol
or xterm modifyOtherKeys mode.
"""
from __future__ import annotations
import pytest
from prompt_toolkit.input.ansi_escape_sequences import ANSI_SEQUENCES
from prompt_toolkit.input.vt100_parser import Vt100Parser
from prompt_toolkit.keys import Keys
from hermes_cli.pt_input_extras import install_shift_enter_alias
SHIFT_ENTER_SEQUENCES = (
"\x1b[13;2u", # Kitty / CSI-u, modifier=2 (Shift)
"\x1b[27;2;13~", # xterm modifyOtherKeys=2
"\x1b[27;2;13u",
)
@pytest.fixture(autouse=True)
def _ensure_alias_installed():
"""Make every test idempotent — install the alias once per test run."""
install_shift_enter_alias()
def _parse(byte_seq: str):
out = []
parser = Vt100Parser(out.append)
for ch in byte_seq:
parser.feed(ch)
parser.flush()
return [kp.key for kp in out]
def test_install_registers_all_three_sequences():
for seq in SHIFT_ENTER_SEQUENCES:
assert seq in ANSI_SEQUENCES, f"missing mapping for {seq!r}"
assert ANSI_SEQUENCES[seq] == (Keys.Escape, Keys.ControlM)
def test_install_overwrites_stock_modifyotherkeys_shift_enter():
"""Stock prompt_toolkit maps `\\x1b[27;2;13~` to plain Keys.ControlM —
i.e. it drops the Shift modifier and treats Shift+Enter like Enter,
which is the bug this helper exists to fix. The install must overwrite
that entry."""
seq = "\x1b[27;2;13~"
ANSI_SEQUENCES[seq] = Keys.ControlM
install_shift_enter_alias()
assert ANSI_SEQUENCES[seq] == (Keys.Escape, Keys.ControlM)
def test_install_returns_zero_when_already_correct():
"""Idempotency — running install twice should not report a second change."""
install_shift_enter_alias()
assert install_shift_enter_alias() == 0
def test_csi_u_shift_enter_parses_as_alt_enter():
"""Kitty keyboard protocol Shift+Enter must parse to the same key tuple
Alt+Enter produces, so the existing handler is reused."""
alt_enter = _parse("\x1b\r")
shift_enter = _parse("\x1b[13;2u")
assert shift_enter == alt_enter, (
f"Shift+Enter via CSI-u should parse identically to Alt+Enter; "
f"got {shift_enter!r} vs {alt_enter!r}"
)
def test_modify_other_keys_shift_enter_parses_as_alt_enter():
"""xterm modifyOtherKeys=2 Shift+Enter must parse identically to Alt+Enter."""
alt_enter = _parse("\x1b\r")
shift_enter = _parse("\x1b[27;2;13~")
assert shift_enter == alt_enter
def test_plain_enter_remains_distinct_from_alt_enter():
"""Plain Enter must keep emitting a single key (submit), not a two-key
Alt+Enter tuple otherwise we would have broken submit."""
enter = _parse("\r")
alt_enter = _parse("\x1b\r")
assert enter != alt_enter
assert len(enter) == 1
assert len(alt_enter) == 2

View file

@ -204,7 +204,7 @@ Type `/` to see an autocomplete dropdown of all commands:
### Multi-line input
Press `Alt+Enter` or `Ctrl+J` to add a new line. Great for pasting code or writing detailed prompts.
Press `Alt+Enter`, `Ctrl+J`, or `Shift+Enter` to add a new line. `Shift+Enter` requires a terminal that sends it as a distinct sequence (Kitty / foot / WezTerm / Ghostty by default; iTerm2 / Alacritty / VS Code terminal once the Kitty keyboard protocol is enabled). `Alt+Enter` and `Ctrl+J` work in every terminal.
### Interrupt the agent

View file

@ -36,7 +36,7 @@ Before writing a long prompt explaining how to do something, check if there's al
### Multi-Line Input
Press **Alt+Enter** (or **Ctrl+J**) to insert a newline without sending. This lets you compose multi-line prompts, paste code blocks, or structure complex requests before hitting Enter to send.
Press **Alt+Enter**, **Ctrl+J**, or **Shift+Enter** to insert a newline without sending. `Shift+Enter` only works when the terminal sends it as a distinct keystroke (Kitty / foot / WezTerm / Ghostty by default; iTerm2 / Alacritty / VS Code terminal once the Kitty keyboard protocol is enabled). The other two work in every terminal.
### Paste Detection

View file

@ -92,7 +92,7 @@ When resuming a previous session (`hermes -c` or `hermes --resume <id>`), a "Pre
| Key | Action |
|-----|--------|
| `Enter` | Send message |
| `Alt+Enter` or `Ctrl+J` | New line (multi-line input) |
| `Alt+Enter`, `Ctrl+J`, or `Shift+Enter` | New line (multi-line input). `Shift+Enter` requires a terminal that distinguishes it from `Enter` — see below. |
| `Alt+V` | Paste an image from the clipboard when supported by the terminal |
| `Ctrl+V` | Paste text and opportunistically attach clipboard images |
| `Ctrl+B` | Start/stop voice recording when voice mode is enabled (`voice.record_key`, default: `ctrl+b`) |
@ -204,7 +204,7 @@ personalities:
There are two ways to enter multi-line messages:
1. **`Alt+Enter` or `Ctrl+J`** — inserts a new line
1. **`Alt+Enter`, `Ctrl+J`, or `Shift+Enter`** — inserts a new line
2. **Backslash continuation** — end a line with `\` to continue:
```
@ -214,9 +214,21 @@ There are two ways to enter multi-line messages:
```
:::info
Pasting multi-line text is supported — use `Alt+Enter` or `Ctrl+J` to insert newlines, or simply paste content directly.
Pasting multi-line text is supported — use any of the newline keys above, or simply paste content directly.
:::
### Shift+Enter compatibility
Most terminals send the same byte sequence for `Enter` and `Shift+Enter` by default, so applications cannot distinguish them. Hermes recognises `Shift+Enter` only when the terminal sends a distinct sequence via the [Kitty keyboard protocol](https://sw.kovidgoyal.net/kitty/keyboard-protocol/) or xterm's `modifyOtherKeys` mode.
| Terminal | Status |
|---|---|
| Kitty, foot, WezTerm, Ghostty | Distinct `Shift+Enter` enabled by default |
| iTerm2 (recent), Alacritty, VS Code terminal, Warp | Supported once the Kitty protocol is enabled in settings |
| macOS Terminal.app, stock Windows Terminal | Not supported — `Shift+Enter` is indistinguishable from `Enter` |
Where the terminal cannot distinguish them, `Alt+Enter` and `Ctrl+J` continue to work everywhere.
## Interrupting the Agent
You can interrupt the agent at any point: