fix: add numbered approval and clarify shortcuts (#956 )

fix: repair auxiliary client syntax corruption
2026-04-22 10:44:09 -04:00 · 2026-04-22 10:44:00 -04:00
7 changed files with 345 additions and 763 deletions
--- a/agent/account_usage.py
+++ b/agent/account_usage.py
@@ -1,326 +0,0 @@
-from __future__ import annotations
-
-from dataclasses import dataclass
-from datetime import datetime, timezone
-from typing import Any, Optional
-
-import httpx
-
-from agent.anthropic_adapter import _is_oauth_token, resolve_anthropic_token
-from hermes_cli.auth import _read_codex_tokens, resolve_codex_runtime_credentials
-from hermes_cli.runtime_provider import resolve_runtime_provider
-
-
-def _utc_now() -> datetime:
-    return datetime.now(timezone.utc)
-
-
-@dataclass(frozen=True)
-class AccountUsageWindow:
-    label: str
-    used_percent: Optional[float] = None
-    reset_at: Optional[datetime] = None
-    detail: Optional[str] = None
-
-
-@dataclass(frozen=True)
-class AccountUsageSnapshot:
-    provider: str
-    source: str
-    fetched_at: datetime
-    title: str = "Account limits"
-    plan: Optional[str] = None
-    windows: tuple[AccountUsageWindow, ...] = ()
-    details: tuple[str, ...] = ()
-    unavailable_reason: Optional[str] = None
-
-    @property
-    def available(self) -> bool:
-        return bool(self.windows or self.details) and not self.unavailable_reason
-
-
-def _title_case_slug(value: Optional[str]) -> Optional[str]:
-    cleaned = str(value or "").strip()
-    if not cleaned:
-        return None
-    return cleaned.replace("_", " ").replace("-", " ").title()
-
-
-def _parse_dt(value: Any) -> Optional[datetime]:
-    if value in (None, ""):
-        return None
-    if isinstance(value, (int, float)):
-        return datetime.fromtimestamp(float(value), tz=timezone.utc)
-    if isinstance(value, str):
-        text = value.strip()
-        if not text:
-            return None
-        if text.endswith("Z"):
-            text = text[:-1] + "+00:00"
-        try:
-            dt = datetime.fromisoformat(text)
-            return dt if dt.tzinfo else dt.replace(tzinfo=timezone.utc)
-        except ValueError:
-            return None
-    return None
-
-
-def _format_reset(dt: Optional[datetime]) -> str:
-    if not dt:
-        return "unknown"
-    local_dt = dt.astimezone()
-    delta = dt - _utc_now()
-    total_seconds = int(delta.total_seconds())
-    if total_seconds <= 0:
-        return f"now ({local_dt.strftime('%Y-%m-%d %H:%M %Z')})"
-    hours, rem = divmod(total_seconds, 3600)
-    minutes = rem // 60
-    if hours >= 24:
-        days, hours = divmod(hours, 24)
-        rel = f"in {days}d {hours}h"
-    elif hours > 0:
-        rel = f"in {hours}h {minutes}m"
-    else:
-        rel = f"in {minutes}m"
-    return f"{rel} ({local_dt.strftime('%Y-%m-%d %H:%M %Z')})"
-
-
-def render_account_usage_lines(snapshot: Optional[AccountUsageSnapshot], *, markdown: bool = False) -> list[str]:
-    if not snapshot:
-        return []
-    header = f"📈 {'**' if markdown else ''}{snapshot.title}{'**' if markdown else ''}"
-    lines = [header]
-    if snapshot.plan:
-        lines.append(f"Provider: {snapshot.provider} ({snapshot.plan})")
-    else:
-        lines.append(f"Provider: {snapshot.provider}")
-    for window in snapshot.windows:
-        if window.used_percent is None:
-            base = f"{window.label}: unavailable"
-        else:
-            remaining = max(0, round(100 - float(window.used_percent)))
-            used = max(0, round(float(window.used_percent)))
-            base = f"{window.label}: {remaining}% remaining ({used}% used)"
-        if window.reset_at:
-            base += f" • resets {_format_reset(window.reset_at)}"
-        elif window.detail:
-            base += f" • {window.detail}"
-        lines.append(base)
-    for detail in snapshot.details:
-        lines.append(detail)
-    if snapshot.unavailable_reason:
-        lines.append(f"Unavailable: {snapshot.unavailable_reason}")
-    return lines
-
-
-def _resolve_codex_usage_url(base_url: str) -> str:
-    normalized = (base_url or "").strip().rstrip("/")
-    if not normalized:
-        normalized = "https://chatgpt.com/backend-api/codex"
-    if normalized.endswith("/codex"):
-        normalized = normalized[: -len("/codex")]
-    if "/backend-api" in normalized:
-        return normalized + "/wham/usage"
-    return normalized + "/api/codex/usage"
-
-
-def _fetch_codex_account_usage() -> Optional[AccountUsageSnapshot]:
-    creds = resolve_codex_runtime_credentials(refresh_if_expiring=True)
-    token_data = _read_codex_tokens()
-    tokens = token_data.get("tokens") or {}
-    account_id = str(tokens.get("account_id", "") or "").strip() or None
-    headers = {
-        "Authorization": f"Bearer {creds['api_key']}",
-        "Accept": "application/json",
-        "User-Agent": "codex-cli",
-    }
-    if account_id:
-        headers["ChatGPT-Account-Id"] = account_id
-    with httpx.Client(timeout=15.0) as client:
-        response = client.get(_resolve_codex_usage_url(creds.get("base_url", "")), headers=headers)
-        response.raise_for_status()
-    payload = response.json() or {}
-    rate_limit = payload.get("rate_limit") or {}
-    windows: list[AccountUsageWindow] = []
-    for key, label in (("primary_window", "Session"), ("secondary_window", "Weekly")):
-        window = rate_limit.get(key) or {}
-        used = window.get("used_percent")
-        if used is None:
-            continue
-        windows.append(
-            AccountUsageWindow(
-                label=label,
-                used_percent=float(used),
-                reset_at=_parse_dt(window.get("reset_at")),
-            )
-        )
-    details: list[str] = []
-    credits = payload.get("credits") or {}
-    if credits.get("has_credits"):
-        balance = credits.get("balance")
-        if isinstance(balance, (int, float)):
-            details.append(f"Credits balance: ${float(balance):.2f}")
-        elif credits.get("unlimited"):
-            details.append("Credits balance: unlimited")
-    return AccountUsageSnapshot(
-        provider="openai-codex",
-        source="usage_api",
-        fetched_at=_utc_now(),
-        plan=_title_case_slug(payload.get("plan_type")),
-        windows=tuple(windows),
-        details=tuple(details),
-    )
-
-
-def _fetch_anthropic_account_usage() -> Optional[AccountUsageSnapshot]:
-    token = (resolve_anthropic_token() or "").strip()
-    if not token:
-        return None
-    if not _is_oauth_token(token):
-        return AccountUsageSnapshot(
-            provider="anthropic",
-            source="oauth_usage_api",
-            fetched_at=_utc_now(),
-            unavailable_reason="Anthropic account limits are only available for OAuth-backed Claude accounts.",
-        )
-    headers = {
-        "Authorization": f"Bearer {token}",
-        "Accept": "application/json",
-        "Content-Type": "application/json",
-        "anthropic-beta": "oauth-2025-04-20",
-        "User-Agent": "claude-code/2.1.0",
-    }
-    with httpx.Client(timeout=15.0) as client:
-        response = client.get("https://api.anthropic.com/api/oauth/usage", headers=headers)
-        response.raise_for_status()
-    payload = response.json() or {}
-    windows: list[AccountUsageWindow] = []
-    mapping = (
-        ("five_hour", "Current session"),
-        ("seven_day", "Current week"),
-        ("seven_day_opus", "Opus week"),
-        ("seven_day_sonnet", "Sonnet week"),
-    )
-    for key, label in mapping:
-        window = payload.get(key) or {}
-        util = window.get("utilization")
-        if util is None:
-            continue
-        used = float(util) * 100 if float(util) <= 1 else float(util)
-        windows.append(
-            AccountUsageWindow(
-                label=label,
-                used_percent=used,
-                reset_at=_parse_dt(window.get("resets_at")),
-            )
-        )
-    details: list[str] = []
-    extra = payload.get("extra_usage") or {}
-    if extra.get("is_enabled"):
-        used_credits = extra.get("used_credits")
-        monthly_limit = extra.get("monthly_limit")
-        currency = extra.get("currency") or "USD"
-        if isinstance(used_credits, (int, float)) and isinstance(monthly_limit, (int, float)):
-            details.append(
-                f"Extra usage: {used_credits:.2f} / {monthly_limit:.2f} {currency}"
-            )
-    return AccountUsageSnapshot(
-        provider="anthropic",
-        source="oauth_usage_api",
-        fetched_at=_utc_now(),
-        windows=tuple(windows),
-        details=tuple(details),
-    )
-
-
-def _fetch_openrouter_account_usage(base_url: Optional[str], api_key: Optional[str]) -> Optional[AccountUsageSnapshot]:
-    runtime = resolve_runtime_provider(
-        requested="openrouter",
-        explicit_base_url=base_url,
-        explicit_api_key=api_key,
-    )
-    token = str(runtime.get("api_key", "") or "").strip()
-    if not token:
-        return None
-    normalized = str(runtime.get("base_url", "") or "").rstrip("/")
-    credits_url = f"{normalized}/credits"
-    key_url = f"{normalized}/key"
-    headers = {
-        "Authorization": f"Bearer {token}",
-        "Accept": "application/json",
-    }
-    with httpx.Client(timeout=10.0) as client:
-        credits_resp = client.get(credits_url, headers=headers)
-        credits_resp.raise_for_status()
-        credits = (credits_resp.json() or {}).get("data") or {}
-        try:
-            key_resp = client.get(key_url, headers=headers)
-            key_resp.raise_for_status()
-            key_data = (key_resp.json() or {}).get("data") or {}
-        except Exception:
-            key_data = {}
-    total_credits = float(credits.get("total_credits") or 0.0)
-    total_usage = float(credits.get("total_usage") or 0.0)
-    details = [f"Credits balance: ${max(0.0, total_credits - total_usage):.2f}"]
-    windows: list[AccountUsageWindow] = []
-    limit = key_data.get("limit")
-    limit_remaining = key_data.get("limit_remaining")
-    limit_reset = str(key_data.get("limit_reset") or "").strip()
-    usage = key_data.get("usage")
-    if (
-        isinstance(limit, (int, float))
-        and float(limit) > 0
-        and isinstance(limit_remaining, (int, float))
-        and 0 <= float(limit_remaining) <= float(limit)
-    ):
-        limit_value = float(limit)
-        remaining_value = float(limit_remaining)
-        used_percent = ((limit_value - remaining_value) / limit_value) * 100
-        detail_parts = [f"${remaining_value:.2f} of ${limit_value:.2f} remaining"]
-        if limit_reset:
-            detail_parts.append(f"resets {limit_reset}")
-        windows.append(
-            AccountUsageWindow(
-                label="API key quota",
-                used_percent=used_percent,
-                detail=" • ".join(detail_parts),
-            )
-        )
-    if isinstance(usage, (int, float)):
-        usage_parts = [f"API key usage: ${float(usage):.2f} total"]
-        for value, label in (
-            (key_data.get("usage_daily"), "today"),
-            (key_data.get("usage_weekly"), "this week"),
-            (key_data.get("usage_monthly"), "this month"),
-        ):
-            if isinstance(value, (int, float)) and float(value) > 0:
-                usage_parts.append(f"${float(value):.2f} {label}")
-        details.append(" • ".join(usage_parts))
-    return AccountUsageSnapshot(
-        provider="openrouter",
-        source="credits_api",
-        fetched_at=_utc_now(),
-        windows=tuple(windows),
-        details=tuple(details),
-    )
-
-
-def fetch_account_usage(
-    provider: Optional[str],
-    *,
-    base_url: Optional[str] = None,
-    api_key: Optional[str] = None,
-) -> Optional[AccountUsageSnapshot]:
-    normalized = str(provider or "").strip().lower()
-    if normalized in {"", "auto", "custom"}:
-        return None
-    try:
-        if normalized == "openai-codex":
-            return _fetch_codex_account_usage()
-        if normalized == "anthropic":
-            return _fetch_anthropic_account_usage()
-        if normalized == "openrouter":
-            return _fetch_openrouter_account_usage(base_url, api_key)
-    except Exception:
-        return None
-    return None
--- a/agent/auxiliary_client.py
+++ b/agent/auxiliary_client.py
@@ -1,4 +1,4 @@
-from agent.telemetry_logger import log_token_usage\n"""Shared auxiliary client router for side tasks.
+"""Shared auxiliary client router for side tasks.

 Provides a single resolution chain so every consumer (context compression,
 session search, web extraction, vision analysis, browser vision) picks up
@@ -34,6 +34,8 @@ Payment / credit exhaustion fallback:
  their OpenRouter balance but has Codex OAuth or another provider available.
 """

+from agent.telemetry_logger import log_token_usage
+
 import json
 import logging
 import os
@@ -396,7 +398,8 @@ class _CodexCompletionsAdapter:
                    prompt_tokens=getattr(resp_usage, "input_tokens", 0),
                    completion_tokens=getattr(resp_usage, "output_tokens", 0),
                    total_tokens=getattr(resp_usage, "total_tokens", 0),
-                )\n        log_token_usage(usage.prompt_tokens, usage.completion_tokens, model)
+                )
+                log_token_usage(usage.prompt_tokens, usage.completion_tokens, model)
        except Exception as exc:
            logger.debug("Codex auxiliary Responses API call failed: %s", exc)
            raise
@@ -529,7 +532,8 @@ class _AnthropicCompletionsAdapter:
                prompt_tokens=prompt_tokens,
                completion_tokens=completion_tokens,
                total_tokens=total_tokens,
-            )\n    log_token_usage(usage.prompt_tokens, usage.completion_tokens, model)
+            )
+            log_token_usage(usage.prompt_tokens, usage.completion_tokens, model)

        choice = SimpleNamespace(
            index=0,
--- a/cli.py
+++ b/cli.py
@@ -13,7 +13,6 @@ Usage:
    python cli.py --list-tools             # List available tools and exit
 """

-import concurrent.futures
 import logging
 import os
 import shutil
@@ -64,7 +63,6 @@ from agent.usage_pricing import (
    format_duration_compact,
    format_token_count_compact,
 )
-from agent.account_usage import fetch_account_usage, render_account_usage_lines
 from hermes_cli.banner import _format_context_length, format_banner_version_label

 _COMMAND_SPINNER_FRAMES = ("⠋", "⠙", "⠹", "⠸", "⠼", "⠴", "⠦", "⠧", "⠇", "⠏")
@@ -6473,29 +6471,6 @@ class HermesCLI:
        if cost_result.status == "unknown":
            print(f"  Note:             Pricing unknown for {agent.model}")

-        # Account limits -- fetched off-thread with a hard timeout so slow
-        # provider APIs don't hang the prompt.
-        provider = getattr(agent, "provider", None) or getattr(self, "provider", None)
-        base_url = getattr(agent, "base_url", None) or getattr(self, "base_url", None)
-        api_key = getattr(agent, "api_key", None) or getattr(self, "api_key", None)
-        account_snapshot = None
-        if provider:
-            with concurrent.futures.ThreadPoolExecutor(max_workers=1) as _pool:
-                try:
-                    account_snapshot = _pool.submit(
-                        fetch_account_usage,
-                        provider,
-                        base_url=base_url,
-                        api_key=api_key,
-                    ).result(timeout=10.0)
-                except (concurrent.futures.TimeoutError, Exception):
-                    account_snapshot = None
-        account_lines = [f"  {line}" for line in render_account_usage_lines(account_snapshot)]
-        if account_lines:
-            print()
-            for line in account_lines:
-                print(line)
-
        if self.verbose:
            logging.getLogger().setLevel(logging.DEBUG)
            for noisy in ('openai', 'openai._base_client', 'httpx', 'httpcore', 'asyncio', 'hpack', 'grpc', 'modal'):
@@ -7279,6 +7254,40 @@ class HermesCLI:
            "Use your best judgement to make the choice and proceed."
        )

+    def _handle_clarify_selection(self) -> None:
+        """Process the currently selected clarify choice."""
+        state = self._clarify_state
+        if not state or self._clarify_freetext:
+            return
+
+        selected = state.get("selected", 0)
+        choices = state.get("choices") or []
+        if selected < len(choices):
+            state["response_queue"].put(choices[selected])
+            self._clarify_state = None
+            self._clarify_freetext = False
+            self._invalidate()
+            return
+
+        if selected == len(choices):
+            self._clarify_freetext = True
+            self._invalidate()
+
+    def _handle_clarify_number_shortcut(self, number: int) -> bool:
+        """Select a clarify option by number key."""
+        state = self._clarify_state
+        if not state or self._clarify_freetext:
+            return False
+
+        choices = state.get("choices") or []
+        max_option = len(choices) + 1
+        if number < 1 or number > max_option:
+            return False
+
+        state["selected"] = number - 1
+        self._handle_clarify_selection()
+        return True
+
    def _sudo_password_callback(self) -> str:
        """
        Prompt for sudo password through the prompt_toolkit UI.
@@ -7387,6 +7396,20 @@ class HermesCLI:
            choices.append("view")
        return choices

+    def _handle_approval_number_shortcut(self, number: int) -> bool:
+        """Select an approval option by number key."""
+        state = self._approval_state
+        if not state:
+            return False
+
+        choices = state.get("choices") or []
+        if number < 1 or number > len(choices):
+            return False
+
+        state["selected"] = number - 1
+        self._handle_approval_selection()
+        return True
+
    def _handle_approval_selection(self) -> None:
        """Process the currently selected dangerous-command approval choice."""
        state = self._approval_state
@@ -7462,8 +7485,9 @@ class HermesCLI:
        preview_lines.extend(_wrap_panel_text(cmd_display, 60))
        for i, choice in enumerate(choices):
            prefix = '❯ ' if i == selected else '  '
+            label = f"{i + 1}. {choice_labels.get(choice, choice)}"
            preview_lines.extend(_wrap_panel_text(
-                f"{prefix}{choice_labels.get(choice, choice)}",
+                f"{prefix}{label}",
                60,
                subsequent_indent="  ",
            ))
@@ -7481,7 +7505,7 @@ class HermesCLI:
            _append_panel_line(lines, 'class:approval-border', 'class:approval-cmd', wrapped, box_width)
        _append_blank_panel_line(lines, 'class:approval-border', box_width)
        for i, choice in enumerate(choices):
-            label = choice_labels.get(choice, choice)
+            label = f"{i + 1}. {choice_labels.get(choice, choice)}"
            style = 'class:approval-selected' if i == selected else 'class:approval-choice'
            prefix = '❯ ' if i == selected else '  '
            for wrapped in _wrap_panel_text(f"{prefix}{label}", inner_text_width, subsequent_indent="  "):
@@ -7490,6 +7514,97 @@ class HermesCLI:
        lines.append(('class:approval-border', '╰' + ('─' * box_width) + '╯\n'))
        return lines

+    def _get_clarify_display_fragments(self):
+        """Render the clarify panel for the prompt_toolkit UI."""
+        state = self._clarify_state
+        if not state:
+            return []
+
+        def _panel_box_width(title: str, content_lines: list[str], min_width: int = 46, max_width: int = 76) -> int:
+            term_cols = shutil.get_terminal_size((100, 20)).columns
+            longest = max([len(title)] + [len(line) for line in content_lines] + [min_width - 4])
+            inner = min(max(longest + 4, min_width - 2), max_width - 2, max(24, term_cols - 6))
+            return inner + 2
+
+        def _wrap_panel_text(text: str, width: int, subsequent_indent: str = "") -> list[str]:
+            wrapped = textwrap.wrap(
+                text,
+                width=max(8, width),
+                break_long_words=False,
+                break_on_hyphens=False,
+                subsequent_indent=subsequent_indent,
+            )
+            return wrapped or [""]
+
+        def _append_panel_line(lines, border_style: str, content_style: str, text: str, box_width: int) -> None:
+            inner_width = max(0, box_width - 2)
+            lines.append((border_style, "│ "))
+            lines.append((content_style, text.ljust(inner_width)))
+            lines.append((border_style, " │\n"))
+
+        def _append_blank_panel_line(lines, border_style: str, box_width: int) -> None:
+            lines.append((border_style, "│" + (" " * box_width) + "│\n"))
+
+        question = state["question"]
+        choices = state.get("choices") or []
+        selected = state.get("selected", 0)
+        preview_lines = _wrap_panel_text(question, 60)
+        for i, choice in enumerate(choices):
+            prefix = "❯ " if i == selected and not self._clarify_freetext else "  "
+            label = f"{i + 1}. {choice}"
+            preview_lines.extend(_wrap_panel_text(f"{prefix}{label}", 60, subsequent_indent="  "))
+        other_number = len(choices) + 1
+        other_label = (
+            f"❯ {other_number}. Other (type below)" if self._clarify_freetext
+            else f"❯ {other_number}. Other (type your answer)" if selected == len(choices)
+            else f"  {other_number}. Other (type your answer)"
+        )
+        preview_lines.extend(_wrap_panel_text(other_label, 60, subsequent_indent="  "))
+        box_width = _panel_box_width("Hermes needs your input", preview_lines)
+        inner_text_width = max(8, box_width - 2)
+
+        lines = []
+        lines.append(('class:clarify-border', '╭─ '))
+        lines.append(('class:clarify-title', 'Hermes needs your input'))
+        lines.append(('class:clarify-border', ' ' + ('─' * max(0, box_width - len("Hermes needs your input") - 3)) + '╮\n'))
+        _append_blank_panel_line(lines, 'class:clarify-border', box_width)
+
+        for wrapped in _wrap_panel_text(question, inner_text_width):
+            _append_panel_line(lines, 'class:clarify-border', 'class:clarify-question', wrapped, box_width)
+        _append_blank_panel_line(lines, 'class:clarify-border', box_width)
+
+        if self._clarify_freetext and not choices:
+            guidance = "Type your answer in the prompt below, then press Enter."
+            for wrapped in _wrap_panel_text(guidance, inner_text_width):
+                _append_panel_line(lines, 'class:clarify-border', 'class:clarify-choice', wrapped, box_width)
+            _append_blank_panel_line(lines, 'class:clarify-border', box_width)
+
+        if choices:
+            for i, choice in enumerate(choices):
+                style = 'class:clarify-selected' if i == selected and not self._clarify_freetext else 'class:clarify-choice'
+                prefix = '❯ ' if i == selected and not self._clarify_freetext else '  '
+                label = f"{i + 1}. {choice}"
+                wrapped_lines = _wrap_panel_text(f"{prefix}{label}", inner_text_width, subsequent_indent="  ")
+                for wrapped in wrapped_lines:
+                    _append_panel_line(lines, 'class:clarify-border', style, wrapped, box_width)
+
+            other_idx = len(choices)
+            if selected == other_idx and not self._clarify_freetext:
+                other_style = 'class:clarify-selected'
+                other_label = f'❯ {other_number}. Other (type your answer)'
+            elif self._clarify_freetext:
+                other_style = 'class:clarify-active-other'
+                other_label = f'❯ {other_number}. Other (type below)'
+            else:
+                other_style = 'class:clarify-choice'
+                other_label = f'  {other_number}. Other (type your answer)'
+            for wrapped in _wrap_panel_text(other_label, inner_text_width, subsequent_indent="  "):
+                _append_panel_line(lines, 'class:clarify-border', other_style, wrapped, box_width)
+
+        _append_blank_panel_line(lines, 'class:clarify-border', box_width)
+        lines.append(('class:clarify-border', '╰' + ('─' * box_width) + '╯\n'))
+        return lines
+
    def _secret_capture_callback(self, var_name: str, prompt: str, metadata=None) -> dict:
        return prompt_for_secret(self, var_name, prompt, metadata)

@@ -8396,17 +8511,8 @@ class HermesCLI:

            # --- Clarify choice mode: confirm the highlighted selection ---
            if self._clarify_state and not self._clarify_freetext:
-                state = self._clarify_state
-                selected = state["selected"]
-                choices = state.get("choices") or []
-                if selected < len(choices):
-                    state["response_queue"].put(choices[selected])
-                    self._clarify_state = None
-                    event.app.invalidate()
-                else:
-                    # "Other" selected → switch to freetext
-                    self._clarify_freetext = True
-                    event.app.invalidate()
+                self._handle_clarify_selection()
+                event.app.invalidate()
                return

            # --- Normal input routing ---
@@ -8526,6 +8632,19 @@ class HermesCLI:
                self._approval_state["selected"] = min(max_idx, self._approval_state["selected"] + 1)
                event.app.invalidate()

+        # --- Numbered shortcuts for clarify / approval modal prompts ---
+        for _digit in '123456789':
+            @kb.add(_digit, filter=Condition(lambda: bool(self._approval_state) or (bool(self._clarify_state) and not self._clarify_freetext)))
+            def _handle_modal_number(event, digit=_digit):
+                number = int(digit)
+                handled = False
+                if self._approval_state:
+                    handled = self._handle_approval_number_shortcut(number)
+                elif self._clarify_state and not self._clarify_freetext:
+                    handled = self._handle_clarify_number_shortcut(number)
+                if handled:
+                    event.app.invalidate()
+
        # --- /model picker: arrow-key navigation ---
        @kb.add('up', filter=Condition(lambda: bool(self._model_picker_state)))
        def model_picker_up(event):
@@ -9020,7 +9139,7 @@ class HermesCLI:
            if cli_ref._approval_state:
                remaining = max(0, int(cli_ref._approval_deadline - _time.monotonic()))
                return [
-                    ('class:hint', '  ↑/↓ to select, Enter to confirm'),
+                    ('class:hint', '  1-9 or ↑/↓ to select, Enter to confirm'),
                    ('class:clarify-countdown', f'  ({remaining}s)'),
                ]

@@ -9033,7 +9152,7 @@ class HermesCLI:
                        ('class:clarify-countdown', countdown),
                    ]
                return [
-                    ('class:hint', '  ↑/↓ to select, Enter to confirm'),
+                    ('class:hint', '  1-9 or ↑/↓ to select, Enter to confirm'),
                    ('class:clarify-countdown', countdown),
                ]

@@ -9111,71 +9230,7 @@ class HermesCLI:
            lines.append((border_style, "│" + (" " * box_width) + "│\n"))

        def _get_clarify_display():
-            """Build styled text for the clarify question/choices panel."""
-            state = cli_ref._clarify_state
-            if not state:
-                return []
-
-            question = state["question"]
-            choices = state.get("choices") or []
-            selected = state.get("selected", 0)
-            preview_lines = _wrap_panel_text(question, 60)
-            for i, choice in enumerate(choices):
-                prefix = "❯ " if i == selected and not cli_ref._clarify_freetext else "  "
-                preview_lines.extend(_wrap_panel_text(f"{prefix}{choice}", 60, subsequent_indent="  "))
-            other_label = (
-                "❯ Other (type below)" if cli_ref._clarify_freetext
-                else "❯ Other (type your answer)" if selected == len(choices)
-                else "  Other (type your answer)"
-            )
-            preview_lines.extend(_wrap_panel_text(other_label, 60, subsequent_indent="  "))
-            box_width = _panel_box_width("Hermes needs your input", preview_lines)
-            inner_text_width = max(8, box_width - 2)
-
-            lines = []
-            # Box top border
-            lines.append(('class:clarify-border', '╭─ '))
-            lines.append(('class:clarify-title', 'Hermes needs your input'))
-            lines.append(('class:clarify-border', ' ' + ('─' * max(0, box_width - len("Hermes needs your input") - 3)) + '╮\n'))
-            _append_blank_panel_line(lines, 'class:clarify-border', box_width)
-
-            # Question text
-            for wrapped in _wrap_panel_text(question, inner_text_width):
-                _append_panel_line(lines, 'class:clarify-border', 'class:clarify-question', wrapped, box_width)
-            _append_blank_panel_line(lines, 'class:clarify-border', box_width)
-
-            if cli_ref._clarify_freetext and not choices:
-                guidance = "Type your answer in the prompt below, then press Enter."
-                for wrapped in _wrap_panel_text(guidance, inner_text_width):
-                    _append_panel_line(lines, 'class:clarify-border', 'class:clarify-choice', wrapped, box_width)
-                _append_blank_panel_line(lines, 'class:clarify-border', box_width)
-
-            if choices:
-                # Multiple-choice mode: show selectable options
-                for i, choice in enumerate(choices):
-                    style = 'class:clarify-selected' if i == selected and not cli_ref._clarify_freetext else 'class:clarify-choice'
-                    prefix = '❯ ' if i == selected and not cli_ref._clarify_freetext else '  '
-                    wrapped_lines = _wrap_panel_text(f"{prefix}{choice}", inner_text_width, subsequent_indent="  ")
-                    for wrapped in wrapped_lines:
-                        _append_panel_line(lines, 'class:clarify-border', style, wrapped, box_width)
-
-                # "Other" option (5th line, only shown when choices exist)
-                other_idx = len(choices)
-                if selected == other_idx and not cli_ref._clarify_freetext:
-                    other_style = 'class:clarify-selected'
-                    other_label = '❯ Other (type your answer)'
-                elif cli_ref._clarify_freetext:
-                    other_style = 'class:clarify-active-other'
-                    other_label = '❯ Other (type below)'
-                else:
-                    other_style = 'class:clarify-choice'
-                    other_label = '  Other (type your answer)'
-                for wrapped in _wrap_panel_text(other_label, inner_text_width, subsequent_indent="  "):
-                    _append_panel_line(lines, 'class:clarify-border', other_style, wrapped, box_width)
-
-            _append_blank_panel_line(lines, 'class:clarify-border', box_width)
-            lines.append(('class:clarify-border', '╰' + ('─' * box_width) + '╯\n'))
-            return lines
+            return cli_ref._get_clarify_display_fragments()

        clarify_widget = ConditionalContainer(
            Window(
--- a/gateway/run.py
+++ b/gateway/run.py
@@ -28,8 +28,6 @@ from pathlib import Path
 from datetime import datetime
 from typing import Dict, Optional, Any, List

-from agent.account_usage import fetch_account_usage, render_account_usage_lines
-
 # ---------------------------------------------------------------------------
 # SSL certificate auto-detection for NixOS and other non-standard systems.
 # Must run BEFORE any HTTP library (discord, aiohttp, etc.) is imported.
@@ -6483,38 +6481,6 @@ class GatewayRunner:
                    if cached:
                        agent = cached[0]

-        # Resolve provider/base_url/api_key for the account-usage fetch.
-        # Prefer the live agent; fall back to persisted billing data on the
-        # SessionDB row so `/usage` still returns account info between turns
-        # when no agent is resident.
-        provider = getattr(agent, "provider", None) if agent and agent is not _AGENT_PENDING_SENTINEL else None
-        base_url = getattr(agent, "base_url", None) if agent and agent is not _AGENT_PENDING_SENTINEL else None
-        api_key = getattr(agent, "api_key", None) if agent and agent is not _AGENT_PENDING_SENTINEL else None
-        if not provider and getattr(self, "_session_db", None) is not None:
-            try:
-                _entry_for_billing = self.session_store.get_or_create_session(source)
-                persisted = self._session_db.get_session(_entry_for_billing.session_id) or {}
-            except Exception:
-                persisted = {}
-            provider = provider or persisted.get("billing_provider")
-            base_url = base_url or persisted.get("billing_base_url")
-
-        # Fetch account usage off the event loop so slow provider APIs don't
-        # block the gateway. Failures are non-fatal -- account_lines stays [].
-        account_lines: list[str] = []
-        if provider:
-            try:
-                account_snapshot = await asyncio.to_thread(
-                    fetch_account_usage,
-                    provider,
-                    base_url=base_url,
-                    api_key=api_key,
-                )
-            except Exception:
-                account_snapshot = None
-            if account_snapshot:
-                account_lines = render_account_usage_lines(account_snapshot, markdown=True)
-
        if agent and hasattr(agent, "session_total_tokens") and agent.session_api_calls > 0:
            lines = []

@@ -6572,10 +6538,6 @@ class GatewayRunner:
            if ctx.compression_count:
                lines.append(f"Compressions: {ctx.compression_count}")

-            if account_lines:
-                lines.append("")
-                lines.extend(account_lines)
-
            return "\n".join(lines)

        # No agent at all -- check session history for a rough count
@@ -6585,18 +6547,12 @@ class GatewayRunner:
            from agent.model_metadata import estimate_messages_tokens_rough
            msgs = [m for m in history if m.get("role") in ("user", "assistant") and m.get("content")]
            approx = estimate_messages_tokens_rough(msgs)
-            lines = [
-                "📊 **Session Info**",
-                f"Messages: {len(msgs)}",
-                f"Estimated context: ~{approx:,} tokens",
-                "_(Detailed usage available after the first agent response)_",
-            ]
-            if account_lines:
-                lines.append("")
-                lines.extend(account_lines)
-            return "\n".join(lines)
-        if account_lines:
-            return "\n".join(account_lines)
+            return (
+                f"📊 **Session Info**\n"
+                f"Messages: {len(msgs)}\n"
+                f"Estimated context: ~{approx:,} tokens\n"
+                f"_(Detailed usage available after the first agent response)_"
+            )
        return "No usage data available for this session."

    async def _handle_insights_command(self, event: MessageEvent) -> str:
--- a/tests/cli/test_cli_numbered_prompt_shortcuts.py
+++ b/tests/cli/test_cli_numbered_prompt_shortcuts.py
@@ -0,0 +1,172 @@
+import queue
+import threading
+from types import SimpleNamespace
+from unittest.mock import MagicMock
+
+from cli import HermesCLI
+
+
+class _FakeBuffer:
+    def __init__(self, text="", cursor_position=None):
+        self.text = text
+        self.cursor_position = len(text) if cursor_position is None else cursor_position
+
+    def reset(self, append_to_history=False):
+        self.text = ""
+        self.cursor_position = 0
+
+
+def _make_cli_stub():
+    cli = HermesCLI.__new__(HermesCLI)
+    cli._approval_state = None
+    cli._approval_deadline = 0
+    cli._approval_lock = threading.Lock()
+    cli._clarify_state = None
+    cli._clarify_freetext = False
+    cli._clarify_deadline = 0
+    cli._sudo_state = None
+    cli._sudo_deadline = 0
+    cli._secret_state = None
+    cli._secret_deadline = 0
+    cli._modal_input_snapshot = None
+    cli._invalidate = MagicMock()
+    cli._app = SimpleNamespace(invalidate=MagicMock(), current_buffer=_FakeBuffer())
+    return cli
+
+
+def test_approval_display_numbers_choices():
+    cli = _make_cli_stub()
+    cli._approval_state = {
+        "command": "sudo rm -rf /tmp/example",
+        "description": "dangerous command",
+        "choices": ["once", "session", "always", "deny"],
+        "selected": 0,
+        "response_queue": queue.Queue(),
+    }
+
+    rendered = "".join(text for _style, text in cli._get_approval_display_fragments())
+
+    assert "❯ 1. Allow once" in rendered
+    assert "2. Allow for this session" in rendered
+    assert "3. Add to permanent allowlist" in rendered
+    assert "4. Deny" in rendered
+
+
+def test_approval_number_shortcut_submits_choice():
+    cli = _make_cli_stub()
+    response_queue = queue.Queue()
+    cli._approval_state = {
+        "command": "sudo rm -rf /tmp/example",
+        "description": "dangerous command",
+        "choices": ["once", "session", "always", "deny"],
+        "selected": 0,
+        "response_queue": response_queue,
+    }
+
+    assert cli._handle_approval_number_shortcut(2) is True
+    assert response_queue.get_nowait() == "session"
+    assert cli._approval_state is None
+
+
+def test_approval_selection_still_submits_selected_choice():
+    cli = _make_cli_stub()
+    response_queue = queue.Queue()
+    cli._approval_state = {
+        "command": "sudo rm -rf /tmp/example",
+        "description": "dangerous command",
+        "choices": ["once", "session", "always", "deny"],
+        "selected": 1,
+        "response_queue": response_queue,
+    }
+
+    cli._handle_approval_selection()
+
+    assert response_queue.get_nowait() == "session"
+    assert cli._approval_state is None
+
+
+def test_approval_number_shortcut_handles_view_in_place():
+    cli = _make_cli_stub()
+    response_queue = queue.Queue()
+    cli._approval_state = {
+        "command": "sudo dd if=/tmp/in of=/usr/share/keyrings/githubcli-archive-keyring.gpg bs=4M status=progress",
+        "description": "disk copy",
+        "choices": ["once", "session", "always", "deny", "view"],
+        "selected": 0,
+        "response_queue": response_queue,
+    }
+
+    assert cli._handle_approval_number_shortcut(5) is True
+    assert cli._approval_state is not None
+    assert cli._approval_state["show_full"] is True
+    assert "view" not in cli._approval_state["choices"]
+    assert cli._approval_state["selected"] == 3
+    assert response_queue.empty()
+
+
+def test_clarify_display_numbers_choices_and_other():
+    cli = _make_cli_stub()
+    cli._clarify_state = {
+        "question": "Pick the best option",
+        "choices": ["Alpha", "Beta", "Gamma", "Delta"],
+        "selected": 1,
+        "response_queue": queue.Queue(),
+    }
+
+    rendered = "".join(text for _style, text in cli._get_clarify_display_fragments())
+
+    assert "1. Alpha" in rendered
+    assert "❯ 2. Beta" in rendered
+    assert "3. Gamma" in rendered
+    assert "4. Delta" in rendered
+    assert "5. Other (type your answer)" in rendered
+
+
+def test_clarify_number_shortcut_submits_choice():
+    cli = _make_cli_stub()
+    response_queue = queue.Queue()
+    cli._clarify_state = {
+        "question": "Pick the best option",
+        "choices": ["Alpha", "Beta", "Gamma"],
+        "selected": 0,
+        "response_queue": response_queue,
+    }
+
+    assert cli._handle_clarify_number_shortcut(3) is True
+    assert response_queue.get_nowait() == "Gamma"
+    assert cli._clarify_state is None
+    assert cli._clarify_freetext is False
+
+
+def test_clarify_selection_still_submits_selected_choice():
+    cli = _make_cli_stub()
+    response_queue = queue.Queue()
+    cli._clarify_state = {
+        "question": "Pick the best option",
+        "choices": ["Alpha", "Beta", "Gamma"],
+        "selected": 1,
+        "response_queue": response_queue,
+    }
+
+    cli._handle_clarify_selection()
+
+    assert response_queue.get_nowait() == "Beta"
+    assert cli._clarify_state is None
+    assert cli._clarify_freetext is False
+
+
+def test_clarify_number_shortcut_activates_other_freetext():
+    cli = _make_cli_stub()
+    response_queue = queue.Queue()
+    cli._clarify_state = {
+        "question": "Pick the best option",
+        "choices": ["Alpha", "Beta", "Gamma"],
+        "selected": 0,
+        "response_queue": response_queue,
+    }
+
+    assert cli._handle_clarify_number_shortcut(4) is True
+    assert cli._clarify_state is not None
+    assert cli._clarify_state["selected"] == 3
+    assert cli._clarify_freetext is True
+    assert response_queue.empty()
--- a/tests/gateway/test_usage_command.py
+++ b/tests/gateway/test_usage_command.py
@@ -175,79 +175,3 @@ class TestUsageCachedAgent:
            result = await runner._handle_usage_command(event)

        assert "Cost: included" in result
-
-
-class TestUsageAccountSection:
-    """Account-limits section appended to /usage output."""
-
-    @pytest.mark.asyncio
-    async def test_usage_command_includes_account_section(self, monkeypatch):
-        agent = _make_mock_agent(provider="openai-codex")
-        agent.base_url = "https://chatgpt.com/backend-api/codex"
-        agent.api_key = "unused"
-        runner = _make_runner(SK, cached_agent=agent)
-        event = MagicMock()
-
-        monkeypatch.setattr(
-            "gateway.run.fetch_account_usage",
-            lambda provider, base_url=None, api_key=None: object(),
-        )
-        monkeypatch.setattr(
-            "gateway.run.render_account_usage_lines",
-            lambda snapshot, markdown=False: [
-                "📈 **Account limits**",
-                "Provider: openai-codex (Pro)",
-                "Session: 85% remaining (15% used)",
-            ],
-        )
-        with patch("agent.rate_limit_tracker.format_rate_limit_compact", return_value="RPM: 50/60"), \
-             patch("agent.usage_pricing.estimate_usage_cost") as mock_cost:
-            mock_cost.return_value = MagicMock(amount_usd=None, status="included")
-            result = await runner._handle_usage_command(event)
-
-        assert "📊 **Session Token Usage**" in result
-        assert "📈 **Account limits**" in result
-        assert "Provider: openai-codex (Pro)" in result
-
-    @pytest.mark.asyncio
-    async def test_usage_command_uses_persisted_provider_when_agent_not_running(self, monkeypatch):
-        runner = _make_runner(SK)
-        runner._session_db = MagicMock()
-        runner._session_db.get_session.return_value = {
-            "billing_provider": "openai-codex",
-            "billing_base_url": "https://chatgpt.com/backend-api/codex",
-        }
-        session_entry = MagicMock()
-        session_entry.session_id = "sess-1"
-        runner.session_store.get_or_create_session.return_value = session_entry
-        runner.session_store.load_transcript.return_value = [
-            {"role": "user", "content": "earlier"},
-        ]
-
-        calls = {}
-
-        async def _fake_to_thread(fn, *args, **kwargs):
-            calls["args"] = args
-            calls["kwargs"] = kwargs
-            return fn(*args, **kwargs)
-
-        monkeypatch.setattr("gateway.run.asyncio.to_thread", _fake_to_thread)
-        monkeypatch.setattr(
-            "gateway.run.fetch_account_usage",
-            lambda provider, base_url=None, api_key=None: object(),
-        )
-        monkeypatch.setattr(
-            "gateway.run.render_account_usage_lines",
-            lambda snapshot, markdown=False: [
-                "📈 **Account limits**",
-                "Provider: openai-codex (Pro)",
-            ],
-        )
-
-        event = MagicMock()
-        result = await runner._handle_usage_command(event)
-
-        assert calls["args"] == ("openai-codex",)
-        assert calls["kwargs"]["base_url"] == "https://chatgpt.com/backend-api/codex"
-        assert "📊 **Session Info**" in result
-        assert "📈 **Account limits**" in result
--- a/tests/test_account_usage.py
+++ b/tests/test_account_usage.py
@@ -1,203 +0,0 @@
-from datetime import datetime, timezone
-
-from agent.account_usage import (
-    AccountUsageSnapshot,
-    AccountUsageWindow,
-    fetch_account_usage,
-    render_account_usage_lines,
-)
-
-
-class _Response:
-    def __init__(self, payload, status_code=200):
-        self._payload = payload
-        self.status_code = status_code
-
-    def raise_for_status(self):
-        if self.status_code >= 400:
-            raise RuntimeError(f"HTTP {self.status_code}")
-
-    def json(self):
-        return self._payload
-
-
-class _Client:
-    def __init__(self, payload):
-        self._payload = payload
-
-    def __enter__(self):
-        return self
-
-    def __exit__(self, exc_type, exc, tb):
-        return False
-
-    def get(self, url, headers=None):
-        return _Response(self._payload)
-
-
-class _RoutingClient:
-    def __init__(self, payloads):
-        self._payloads = payloads
-
-    def __enter__(self):
-        return self
-
-    def __exit__(self, exc_type, exc, tb):
-        return False
-
-    def get(self, url, headers=None):
-        return _Response(self._payloads[url])
-
-
-def test_fetch_account_usage_codex(monkeypatch):
-    monkeypatch.setattr(
-        "agent.account_usage.resolve_codex_runtime_credentials",
-        lambda refresh_if_expiring=True: {
-            "provider": "openai-codex",
-            "base_url": "https://chatgpt.com/backend-api/codex",
-            "api_key": "***",
-        },
-    )
-    monkeypatch.setattr(
-        "agent.account_usage._read_codex_tokens",
-        lambda: {"tokens": {"account_id": "acct_123"}},
-    )
-    monkeypatch.setattr(
-        "agent.account_usage.httpx.Client",
-        lambda timeout=15.0: _Client(
-            {
-                "plan_type": "pro",
-                "rate_limit": {
-                    "primary_window": {
-                        "used_percent": 15,
-                        "reset_at": 1_900_000_000,
-                        "limit_window_seconds": 18000,
-                    },
-                    "secondary_window": {
-                        "used_percent": 40,
-                        "reset_at": 1_900_500_000,
-                        "limit_window_seconds": 604800,
-                    },
-                },
-                "credits": {"has_credits": True, "balance": 12.5},
-            }
-        ),
-    )
-
-    snapshot = fetch_account_usage("openai-codex")
-
-    assert snapshot is not None
-    assert snapshot.plan == "Pro"
-    assert len(snapshot.windows) == 2
-    assert snapshot.windows[0].label == "Session"
-    assert snapshot.windows[0].used_percent == 15.0
-    assert snapshot.windows[0].reset_at == datetime.fromtimestamp(1_900_000_000, tz=timezone.utc)
-    assert "Credits balance: $12.50" in snapshot.details
-
-
-def test_render_account_usage_lines_includes_reset_and_provider():
-    snapshot = AccountUsageSnapshot(
-        provider="openai-codex",
-        source="usage_api",
-        fetched_at=datetime.now(timezone.utc),
-        plan="Pro",
-        windows=(
-            AccountUsageWindow(
-                label="Session",
-                used_percent=25,
-                reset_at=datetime.now(timezone.utc),
-            ),
-        ),
-        details=("Credits balance: $9.99",),
-    )
-    lines = render_account_usage_lines(snapshot)
-
-    assert lines[0] == "📈 Account limits"
-    assert "openai-codex (Pro)" in lines[1]
-    assert "Session: 75% remaining (25% used)" in lines[2]
-    assert "Credits balance: $9.99" in lines[3]
-
-
-def test_fetch_account_usage_openrouter_uses_limit_remaining_and_ignores_deprecated_rate_limit(monkeypatch):
-    monkeypatch.setattr(
-        "agent.account_usage.resolve_runtime_provider",
-        lambda requested, explicit_base_url=None, explicit_api_key=None: {
-            "provider": "openrouter",
-            "base_url": "https://openrouter.ai/api/v1",
-            "api_key": "***",
-        },
-    )
-    monkeypatch.setattr(
-        "agent.account_usage.httpx.Client",
-        lambda timeout=10.0: _RoutingClient(
-            {
-                "https://openrouter.ai/api/v1/credits": {
-                    "data": {"total_credits": 300.0, "total_usage": 10.92}
-                },
-                "https://openrouter.ai/api/v1/key": {
-                    "data": {
-                        "limit": 100.0,
-                        "limit_remaining": 70.0,
-                        "limit_reset": "monthly",
-                        "usage": 12.5,
-                        "usage_daily": 0.5,
-                        "usage_weekly": 2.0,
-                        "usage_monthly": 8.0,
-                        "rate_limit": {"requests": -1, "interval": "10s"},
-                    }
-                },
-            }
-        ),
-    )
-
-    snapshot = fetch_account_usage("openrouter")
-
-    assert snapshot is not None
-    assert snapshot.windows == (
-        AccountUsageWindow(
-            label="API key quota",
-            used_percent=30.0,
-            detail="$70.00 of $100.00 remaining • resets monthly",
-        ),
-    )
-    assert "Credits balance: $289.08" in snapshot.details
-    assert "API key usage: $12.50 total • $0.50 today • $2.00 this week • $8.00 this month" in snapshot.details
-    assert all("-1 requests / 10s" not in line for line in render_account_usage_lines(snapshot))
-
-
-def test_fetch_account_usage_openrouter_omits_quota_window_when_key_has_no_limit(monkeypatch):
-    monkeypatch.setattr(
-        "agent.account_usage.resolve_runtime_provider",
-        lambda requested, explicit_base_url=None, explicit_api_key=None: {
-            "provider": "openrouter",
-            "base_url": "https://openrouter.ai/api/v1",
-            "api_key": "***",
-        },
-    )
-    monkeypatch.setattr(
-        "agent.account_usage.httpx.Client",
-        lambda timeout=10.0: _RoutingClient(
-            {
-                "https://openrouter.ai/api/v1/credits": {
-                    "data": {"total_credits": 100.0, "total_usage": 25.5}
-                },
-                "https://openrouter.ai/api/v1/key": {
-                    "data": {
-                        "limit": None,
-                        "limit_remaining": None,
-                        "usage": 25.5,
-                        "usage_daily": 1.25,
-                        "usage_weekly": 4.5,
-                        "usage_monthly": 18.0,
-                    }
-                },
-            }
-        ),
-    )
-
-    snapshot = fetch_account_usage("openrouter")
-
-    assert snapshot is not None
-    assert snapshot.windows == ()
-    assert "Credits balance: $74.50" in snapshot.details
-    assert "API key usage: $25.50 total • $1.25 today • $4.50 this week • $18.00 this month" in snapshot.details
Author	SHA1	Message	Date
Alexander Whitestone	a23f54d15f	fix: add numbered approval and clarify shortcuts (#956 ) All checks were successful Lint / lint (pull_request) Successful in 21s Details	2026-04-22 10:44:09 -04:00
Alexander Whitestone	ffb3257cb5	fix: repair auxiliary client syntax corruption	2026-04-22 10:44:00 -04:00