feat: verify AI Gateway provider UX and attribution headers (#950)

Closes #950

- promote Vercel AI Gateway near the top of the provider picker
- add dedicated AI Gateway model flow with Vercel API-key deep link and live pricing
- use curated AI Gateway catalog refresh with free Moonshot auto-promotion
- apply AI Gateway attribution headers on runtime clients
- add targeted QA tests for provider UX and attribution headers

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

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
@@ -38,6 +38,7 @@ import json
 import logging
 import os
 import threading
+from agent.telemetry_logger import log_token_usage
 import time
 from pathlib import Path  # noqa: F401 — used by test mocks
 from types import SimpleNamespace
@@ -122,6 +123,16 @@ _OR_HEADERS = {
     "X-OpenRouter-Categories": "productivity,cli-agent",
 }
 
+# Vercel AI Gateway app attribution headers. HTTP-Referer maps to
+# referrerUrl and X-Title maps to appName in the gateway analytics.
+from hermes_cli import __version__ as _HERMES_VERSION
+
+_AI_GATEWAY_HEADERS = {
+    "HTTP-Referer": "https://hermes-agent.nousresearch.com",
+    "X-Title": "Hermes Agent",
+    "User-Agent": f"HermesAgent/{_HERMES_VERSION}",
+}
+
 # Nous Portal extra_body for product attribution.
 # Callers should pass this as extra_body in chat.completions.create()
 # when the auxiliary client is backed by Nous Portal.
@@ -396,7 +407,8 @@ 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 +541,8 @@ 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,

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'):

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:

hermes_cli/main.py

@@ -168,7 +168,7 @@ import time as _time
 from datetime import datetime
 
 from hermes_cli import __version__, __release_date__
-from hermes_constants import OPENROUTER_BASE_URL
+from hermes_constants import AI_GATEWAY_BASE_URL, OPENROUTER_BASE_URL
 
 logger = logging.getLogger(__name__)
 
@@ -1112,6 +1112,8 @@ def select_provider_and_model(args=None):
     # Step 2: Provider-specific setup + model selection
     if selected_provider == "openrouter":
         _model_flow_openrouter(config, current_model)
+    elif selected_provider == "ai-gateway":
+        _model_flow_ai_gateway(config, current_model)
     elif selected_provider == "nous":
         _model_flow_nous(config, current_model, args=args)
     elif selected_provider == "openai-codex":
@@ -1267,6 +1269,55 @@ def _model_flow_openrouter(config, current_model=""):
         print("No change.")
 
 
+def _model_flow_ai_gateway(config, current_model=""):
+    """Vercel AI Gateway provider: ensure API key, then pick model with pricing."""
+    from hermes_cli.auth import _prompt_model_selection, _save_model_choice, deactivate_provider
+    from hermes_cli.config import get_env_value, save_env_value
+    from hermes_cli.models import ai_gateway_model_ids, get_pricing_for_provider
+
+    api_key = get_env_value("AI_GATEWAY_API_KEY")
+    if not api_key:
+        print("No Vercel AI Gateway API key configured.")
+        print("Create API key here: https://vercel.com/d?to=%2F%5Bteam%5D%2F%7E%2Fai-gateway&title=AI+Gateway")
+        print("Add a payment method to get $5 in free credits.")
+        print()
+        try:
+            import getpass
+            key = getpass.getpass("AI Gateway API key (or Enter to cancel): ").strip()
+        except (KeyboardInterrupt, EOFError):
+            print()
+            return
+        if not key:
+            print("Cancelled.")
+            return
+        save_env_value("AI_GATEWAY_API_KEY", key)
+        print("API key saved.")
+        print()
+
+    models_list = ai_gateway_model_ids(force_refresh=True)
+    pricing = get_pricing_for_provider("ai-gateway", force_refresh=True)
+
+    selected = _prompt_model_selection(models_list, current_model=current_model, pricing=pricing)
+    if selected:
+        _save_model_choice(selected)
+
+        from hermes_cli.config import load_config, save_config
+
+        cfg = load_config()
+        model = cfg.get("model")
+        if not isinstance(model, dict):
+            model = {"default": model} if model else {}
+            cfg["model"] = model
+        model["provider"] = "ai-gateway"
+        model["base_url"] = AI_GATEWAY_BASE_URL
+        model["api_mode"] = "chat_completions"
+        save_config(cfg)
+        deactivate_provider()
+        print(f"Default model set to: {selected} (via Vercel AI Gateway)")
+    else:
+        print("No change.")
+
+
 def _model_flow_nous(config, current_model="", args=None):
     """Nous Portal provider: ensure logged in, then pick model."""
     from hermes_cli.auth import (

hermes_cli/models.py

@@ -58,6 +58,28 @@ OPENROUTER_MODELS: list[tuple[str, str]] = [
 
 _openrouter_catalog_cache: list[tuple[str, str]] | None = None
 
+# Fallback Vercel AI Gateway snapshot used when the live catalog is unavailable.
+# OSS / open-weight models prioritized first, then closed-source by family.
+VERCEL_AI_GATEWAY_MODELS: list[tuple[str, str]] = [
+    ("moonshotai/kimi-k2.6", "recommended"),
+    ("alibaba/qwen3.6-plus", ""),
+    ("zai/glm-5.1", ""),
+    ("minimax/minimax-m2.7", ""),
+    ("anthropic/claude-sonnet-4.6", ""),
+    ("anthropic/claude-opus-4.7", ""),
+    ("anthropic/claude-opus-4.6", ""),
+    ("anthropic/claude-haiku-4.5", ""),
+    ("openai/gpt-5.4", ""),
+    ("openai/gpt-5.4-mini", ""),
+    ("openai/gpt-5.3-codex", ""),
+    ("google/gemini-3.1-pro-preview", ""),
+    ("google/gemini-3-flash", ""),
+    ("google/gemini-3.1-flash-lite-preview", ""),
+    ("xai/grok-4.20-reasoning", ""),
+]
+
+_ai_gateway_catalog_cache: list[tuple[str, str]] | None = None
+
 
 def _codex_curated_models() -> list[str]:
     """Derive the openai-codex curated list from codex_models.py.
@@ -258,18 +280,21 @@ _PROVIDER_MODELS: dict[str, list[str]] = {
         "minimax-m2.5",
     ],
     "ai-gateway": [
-        "anthropic/claude-opus-4.6",
+        "moonshotai/kimi-k2.6",
+        "alibaba/qwen3.6-plus",
+        "zai/glm-5.1",
+        "minimax/minimax-m2.7",
         "anthropic/claude-sonnet-4.6",
-        "anthropic/claude-sonnet-4.5",
+        "anthropic/claude-opus-4.7",
+        "anthropic/claude-opus-4.6",
         "anthropic/claude-haiku-4.5",
-        "openai/gpt-5",
-        "openai/gpt-4.1",
-        "openai/gpt-4.1-mini",
-        "google/gemini-3-pro-preview",
+        "openai/gpt-5.4",
+        "openai/gpt-5.4-mini",
+        "openai/gpt-5.3-codex",
+        "google/gemini-3.1-pro-preview",
         "google/gemini-3-flash",
-        "google/gemini-2.5-pro",
-        "google/gemini-2.5-flash",
-        "deepseek/deepseek-v3.2",
+        "google/gemini-3.1-flash-lite-preview",
+        "xai/grok-4.20-reasoning",
     ],
     "kilocode": [
         "anthropic/claude-opus-4.6",
@@ -516,6 +541,7 @@ class ProviderEntry(NamedTuple):
 CANONICAL_PROVIDERS: list[ProviderEntry] = [
     ProviderEntry("nous",           "Nous Portal",              "Nous Portal (Nous Research subscription)"),
     ProviderEntry("openrouter",     "OpenRouter",               "OpenRouter (100+ models, pay-per-use)"),
+    ProviderEntry("ai-gateway",     "Vercel AI Gateway",        "Vercel AI Gateway (200+ models, $5 free credit, no markup)"),
     ProviderEntry("anthropic",      "Anthropic",                "Anthropic (Claude models — API key or Claude Code)"),
     ProviderEntry("openai-codex",   "OpenAI Codex",             "OpenAI Codex"),
     ProviderEntry("xiaomi",         "Xiaomi MiMo",              "Xiaomi MiMo (MiMo-V2 models — pro, omni, flash)"),
@@ -536,7 +562,6 @@ CANONICAL_PROVIDERS: list[ProviderEntry] = [
     ProviderEntry("kilocode",       "Kilo Code",                "Kilo Code (Kilo Gateway API)"),
     ProviderEntry("opencode-zen",   "OpenCode Zen",             "OpenCode Zen (35+ curated models, pay-as-you-go)"),
     ProviderEntry("opencode-go",    "OpenCode Go",              "OpenCode Go (open models, $10/month subscription)"),
-    ProviderEntry("ai-gateway",     "Vercel AI Gateway",        "Vercel AI Gateway (200+ models, pay-per-use)"),
 ]
 
 # Derived dicts — used throughout the codebase
@@ -679,6 +704,90 @@ def model_ids(*, force_refresh: bool = False) -> list[str]:
 
 
 
+def _ai_gateway_model_is_free(pricing: Any) -> bool:
+    """Return True if an AI Gateway model has $0 input AND output pricing."""
+    if not isinstance(pricing, dict):
+        return False
+    try:
+        return float(pricing.get("input", "0")) == 0 and float(pricing.get("output", "0")) == 0
+    except (TypeError, ValueError):
+        return False
+
+
+def fetch_ai_gateway_models(
+    timeout: float = 8.0,
+    *,
+    force_refresh: bool = False,
+) -> list[tuple[str, str]]:
+    """Return the curated AI Gateway picker list, refreshed from the live catalog when possible."""
+    global _ai_gateway_catalog_cache
+
+    if _ai_gateway_catalog_cache is not None and not force_refresh:
+        return list(_ai_gateway_catalog_cache)
+
+    from hermes_constants import AI_GATEWAY_BASE_URL
+
+    fallback = list(VERCEL_AI_GATEWAY_MODELS)
+    preferred_ids = [mid for mid, _ in fallback]
+
+    try:
+        req = urllib.request.Request(
+            f"{AI_GATEWAY_BASE_URL.rstrip('/')}/models",
+            headers={"Accept": "application/json"},
+        )
+        with urllib.request.urlopen(req, timeout=timeout) as resp:
+            payload = json.loads(resp.read().decode())
+    except Exception:
+        return list(_ai_gateway_catalog_cache or fallback)
+
+    live_items = payload.get("data", [])
+    if not isinstance(live_items, list):
+        return list(_ai_gateway_catalog_cache or fallback)
+
+    live_by_id: dict[str, dict[str, Any]] = {}
+    for item in live_items:
+        if not isinstance(item, dict):
+            continue
+        mid = str(item.get("id") or "").strip()
+        if not mid:
+            continue
+        live_by_id[mid] = item
+
+    curated: list[tuple[str, str]] = []
+    for preferred_id in preferred_ids:
+        live_item = live_by_id.get(preferred_id)
+        if live_item is None:
+            continue
+        desc = "free" if _ai_gateway_model_is_free(live_item.get("pricing")) else ""
+        curated.append((preferred_id, desc))
+
+    if not curated:
+        return list(_ai_gateway_catalog_cache or fallback)
+
+    free_moonshot = next(
+        (
+            mid
+            for mid, item in live_by_id.items()
+            if mid.startswith("moonshotai/") and _ai_gateway_model_is_free(item.get("pricing"))
+        ),
+        None,
+    )
+    if free_moonshot:
+        curated = [(mid, desc) for mid, desc in curated if mid != free_moonshot]
+        curated.insert(0, (free_moonshot, "recommended"))
+    else:
+        first_id, _ = curated[0]
+        curated[0] = (first_id, "recommended")
+
+    _ai_gateway_catalog_cache = curated
+    return list(curated)
+
+
+def ai_gateway_model_ids(*, force_refresh: bool = False) -> list[str]:
+    """Return just the AI Gateway model-id strings."""
+    return [mid for mid, _ in fetch_ai_gateway_models(force_refresh=force_refresh)]
+
+
 # ---------------------------------------------------------------------------
 # Pricing helpers — fetch live pricing from OpenRouter-compatible /v1/models
 # ---------------------------------------------------------------------------
@@ -821,6 +930,51 @@ def fetch_models_with_pricing(
     return result
 
 
+def fetch_ai_gateway_pricing(
+    timeout: float = 8.0,
+    *,
+    force_refresh: bool = False,
+) -> dict[str, dict[str, str]]:
+    """Fetch Vercel AI Gateway /v1/models and return Hermes-shaped pricing."""
+    from hermes_constants import AI_GATEWAY_BASE_URL
+
+    cache_key = AI_GATEWAY_BASE_URL.rstrip("/")
+    if not force_refresh and cache_key in _pricing_cache:
+        return _pricing_cache[cache_key]
+
+    try:
+        req = urllib.request.Request(
+            f"{cache_key}/models",
+            headers={"Accept": "application/json"},
+        )
+        with urllib.request.urlopen(req, timeout=timeout) as resp:
+            payload = json.loads(resp.read().decode())
+    except Exception:
+        _pricing_cache[cache_key] = {}
+        return {}
+
+    result: dict[str, dict[str, str]] = {}
+    for item in payload.get("data", []):
+        if not isinstance(item, dict):
+            continue
+        mid = item.get("id")
+        pricing = item.get("pricing")
+        if not (mid and isinstance(pricing, dict)):
+            continue
+        entry: dict[str, str] = {
+            "prompt": str(pricing.get("input", "")),
+            "completion": str(pricing.get("output", "")),
+        }
+        if pricing.get("input_cache_read"):
+            entry["input_cache_read"] = str(pricing["input_cache_read"])
+        if pricing.get("input_cache_write"):
+            entry["input_cache_write"] = str(pricing["input_cache_write"])
+        result[mid] = entry
+
+    _pricing_cache[cache_key] = result
+    return result
+
+
 def _resolve_openrouter_api_key() -> str:
     """Best-effort OpenRouter API key for pricing fetch."""
     return os.getenv("OPENROUTER_API_KEY", "").strip()
@@ -839,7 +993,7 @@ def _resolve_nous_pricing_credentials() -> tuple[str, str]:
 
 
 def get_pricing_for_provider(provider: str, *, force_refresh: bool = False) -> dict[str, dict[str, str]]:
-    """Return live pricing for providers that support it (openrouter, nous)."""
+    """Return live pricing for providers that support it (openrouter, ai-gateway, nous)."""
     normalized = normalize_provider(provider)
     if normalized == "openrouter":
         return fetch_models_with_pricing(
@@ -847,11 +1001,11 @@ def get_pricing_for_provider(provider: str, *, force_refresh: bool = False) -> d
             base_url="https://openrouter.ai/api",
             force_refresh=force_refresh,
         )
+    if normalized == "ai-gateway":
+        return fetch_ai_gateway_pricing(force_refresh=force_refresh)
     if normalized == "nous":
         api_key, base_url = _resolve_nous_pricing_credentials()
         if base_url:
-            # Nous base_url typically looks like https://inference-api.nousresearch.com/v1
-            # We need the part before /v1 for our fetch function
             stripped = base_url.rstrip("/")
             if stripped.endswith("/v1"):
                 stripped = stripped[:-3]
@@ -1253,9 +1407,7 @@ def provider_model_ids(provider: Optional[str], *, force_refresh: bool = False)
         if live:
             return live
     if normalized == "ai-gateway":
-        live = _fetch_ai_gateway_models()
-        if live:
-            return live
+        return ai_gateway_model_ids()
     if normalized == "custom":
         base_url = _get_custom_base_url()
         if base_url:

run_agent.py

@@ -908,6 +908,10 @@ class AIAgent:
                         "X-OpenRouter-Title": "Hermes Agent",
                         "X-OpenRouter-Categories": "productivity,cli-agent",
                     }
+                elif "ai-gateway.vercel.sh" in effective_base.lower():
+                    from agent.auxiliary_client import _AI_GATEWAY_HEADERS
+
+                    client_kwargs["default_headers"] = dict(_AI_GATEWAY_HEADERS)
                 elif "api.githubcopilot.com" in effective_base.lower():
                     from hermes_cli.models import copilot_default_headers
 
@@ -4667,11 +4671,13 @@ class AIAgent:
         return True
 
     def _apply_client_headers_for_base_url(self, base_url: str) -> None:
-        from agent.auxiliary_client import _OR_HEADERS
+        from agent.auxiliary_client import _AI_GATEWAY_HEADERS, _OR_HEADERS
 
         normalized = (base_url or "").lower()
         if "openrouter" in normalized:
             self._client_kwargs["default_headers"] = dict(_OR_HEADERS)
+        elif "ai-gateway.vercel.sh" in normalized:
+            self._client_kwargs["default_headers"] = dict(_AI_GATEWAY_HEADERS)
         elif "api.githubcopilot.com" in normalized:
             from hermes_cli.models import copilot_default_headers
 

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

tests/hermes_cli/test_ai_gateway_models.py

@@ -0,0 +1,222 @@
+"""AI Gateway provider UX, live pricing, and model promotion tests."""
+
+from __future__ import annotations
+
+import json
+from unittest.mock import MagicMock, patch
+
+import pytest
+
+from hermes_cli import models as models_module
+from hermes_cli.models import (
+    CANONICAL_PROVIDERS,
+    VERCEL_AI_GATEWAY_MODELS,
+    _ai_gateway_model_is_free,
+    ai_gateway_model_ids,
+    fetch_ai_gateway_models,
+    fetch_ai_gateway_pricing,
+    get_pricing_for_provider,
+)
+
+
+def _mock_urlopen(payload):
+    resp = MagicMock()
+    resp.read.return_value = json.dumps(payload).encode()
+    ctx = MagicMock()
+    ctx.__enter__.return_value = resp
+    ctx.__exit__.return_value = False
+    return ctx
+
+
+def _reset_caches():
+    models_module._ai_gateway_catalog_cache = None
+    models_module._pricing_cache.clear()
+
+
+@pytest.fixture
+def config_home(tmp_path, monkeypatch):
+    home = tmp_path / "hermes"
+    home.mkdir()
+    (home / "config.yaml").write_text("model: some-old-model\n")
+    (home / ".env").write_text("")
+    monkeypatch.setenv("HERMES_HOME", str(home))
+    monkeypatch.delenv("AI_GATEWAY_API_KEY", raising=False)
+    monkeypatch.delenv("AI_GATEWAY_BASE_URL", raising=False)
+    return home
+
+
+def test_ai_gateway_provider_is_promoted_near_top_of_picker():
+    slugs = [entry.slug for entry in CANONICAL_PROVIDERS]
+    assert "ai-gateway" in slugs[:3]
+
+
+def test_ai_gateway_pricing_translates_input_output_to_prompt_completion():
+    _reset_caches()
+    payload = {
+        "data": [
+            {
+                "id": "moonshotai/kimi-k2.5",
+                "type": "language",
+                "pricing": {
+                    "input": "0.0000006",
+                    "output": "0.0000025",
+                    "input_cache_read": "0.00000015",
+                    "input_cache_write": "0.0000006",
+                },
+            }
+        ]
+    }
+    with patch("urllib.request.urlopen", return_value=_mock_urlopen(payload)):
+        result = fetch_ai_gateway_pricing(force_refresh=True)
+
+    entry = result["moonshotai/kimi-k2.5"]
+    assert entry["prompt"] == "0.0000006"
+    assert entry["completion"] == "0.0000025"
+    assert entry["input_cache_read"] == "0.00000015"
+    assert entry["input_cache_write"] == "0.0000006"
+
+
+def test_get_pricing_for_provider_supports_ai_gateway():
+    _reset_caches()
+    payload = {
+        "data": [
+            {
+                "id": "moonshotai/kimi-k2.5",
+                "type": "language",
+                "pricing": {"input": "0.0001", "output": "0.0002"},
+            }
+        ]
+    }
+    with patch("urllib.request.urlopen", return_value=_mock_urlopen(payload)):
+        result = get_pricing_for_provider("ai-gateway", force_refresh=True)
+    assert result["moonshotai/kimi-k2.5"] == {"prompt": "0.0001", "completion": "0.0002"}
+
+
+def test_ai_gateway_pricing_returns_empty_on_fetch_failure():
+    _reset_caches()
+    with patch("urllib.request.urlopen", side_effect=OSError("network down")):
+        result = fetch_ai_gateway_pricing(force_refresh=True)
+    assert result == {}
+
+
+def test_ai_gateway_pricing_skips_entries_without_pricing_dict():
+    _reset_caches()
+    payload = {
+        "data": [
+            {"id": "x/y", "pricing": None},
+            {"id": "a/b", "pricing": {"input": "0", "output": "0"}},
+        ]
+    }
+    with patch("urllib.request.urlopen", return_value=_mock_urlopen(payload)):
+        result = fetch_ai_gateway_pricing(force_refresh=True)
+    assert "x/y" not in result
+    assert result["a/b"] == {"prompt": "0", "completion": "0"}
+
+
+def test_ai_gateway_free_detector():
+    assert _ai_gateway_model_is_free({"input": "0", "output": "0"}) is True
+    assert _ai_gateway_model_is_free({"input": "0", "output": "0.01"}) is False
+    assert _ai_gateway_model_is_free({"input": "0.01", "output": "0"}) is False
+    assert _ai_gateway_model_is_free(None) is False
+    assert _ai_gateway_model_is_free({"input": "not a number"}) is False
+
+
+def test_fetch_ai_gateway_models_filters_against_live_catalog():
+    _reset_caches()
+    preferred = [mid for mid, _ in VERCEL_AI_GATEWAY_MODELS]
+    live_ids = preferred[:3]
+    payload = {
+        "data": [
+            {"id": mid, "pricing": {"input": "0.001", "output": "0.002"}}
+            for mid in live_ids
+        ]
+    }
+    with patch("urllib.request.urlopen", return_value=_mock_urlopen(payload)):
+        result = fetch_ai_gateway_models(force_refresh=True)
+
+    assert [mid for mid, _ in result] == live_ids
+    assert result[0][1] == "recommended"
+    assert ai_gateway_model_ids(force_refresh=False) == live_ids
+
+
+def test_fetch_ai_gateway_models_tags_free_models():
+    _reset_caches()
+    first_id = VERCEL_AI_GATEWAY_MODELS[0][0]
+    second_id = VERCEL_AI_GATEWAY_MODELS[1][0]
+    payload = {
+        "data": [
+            {"id": first_id, "pricing": {"input": "0.001", "output": "0.002"}},
+            {"id": second_id, "pricing": {"input": "0", "output": "0"}},
+        ]
+    }
+    with patch("urllib.request.urlopen", return_value=_mock_urlopen(payload)):
+        result = fetch_ai_gateway_models(force_refresh=True)
+
+    by_id = dict(result)
+    assert by_id[first_id] == "recommended"
+    assert by_id[second_id] == "free"
+
+
+def test_free_moonshot_model_auto_promoted_to_top_even_if_not_curated():
+    _reset_caches()
+    first_curated = VERCEL_AI_GATEWAY_MODELS[0][0]
+    unlisted_free_moonshot = "moonshotai/kimi-coder-free-preview"
+    payload = {
+        "data": [
+            {"id": first_curated, "pricing": {"input": "0.001", "output": "0.002"}},
+            {"id": unlisted_free_moonshot, "pricing": {"input": "0", "output": "0"}},
+        ]
+    }
+    with patch("urllib.request.urlopen", return_value=_mock_urlopen(payload)):
+        result = fetch_ai_gateway_models(force_refresh=True)
+
+    assert result[0] == (unlisted_free_moonshot, "recommended")
+    assert any(mid == first_curated for mid, _ in result)
+
+
+def test_paid_moonshot_does_not_get_auto_promoted():
+    _reset_caches()
+    first_curated = VERCEL_AI_GATEWAY_MODELS[0][0]
+    payload = {
+        "data": [
+            {"id": first_curated, "pricing": {"input": "0.001", "output": "0.002"}},
+            {"id": "moonshotai/some-paid-variant", "pricing": {"input": "0.001", "output": "0.002"}},
+        ]
+    }
+    with patch("urllib.request.urlopen", return_value=_mock_urlopen(payload)):
+        result = fetch_ai_gateway_models(force_refresh=True)
+
+    assert result[0][0] == first_curated
+
+
+def test_fetch_ai_gateway_models_falls_back_on_error():
+    _reset_caches()
+    with patch("urllib.request.urlopen", side_effect=OSError("network")):
+        result = fetch_ai_gateway_models(force_refresh=True)
+    assert result == list(VERCEL_AI_GATEWAY_MODELS)
+
+
+def test_ai_gateway_setup_flow_shows_deeplink_and_passes_pricing(config_home, monkeypatch, capsys):
+    from hermes_cli.main import _model_flow_ai_gateway
+    from hermes_cli.config import load_config
+
+    pricing = {"moonshotai/kimi-k2.6": {"prompt": "0", "completion": "0"}}
+    monkeypatch.setenv("HERMES_HOME", str(config_home))
+
+    with patch("getpass.getpass", return_value="vercel-key"), \
+         patch("hermes_cli.models.ai_gateway_model_ids", return_value=["moonshotai/kimi-k2.6"]), \
+         patch("hermes_cli.models.get_pricing_for_provider", return_value=pricing), \
+         patch("hermes_cli.auth._prompt_model_selection", return_value="moonshotai/kimi-k2.6") as prompt_selection, \
+         patch("hermes_cli.auth.deactivate_provider"):
+        _model_flow_ai_gateway(load_config(), "")
+
+    out = capsys.readouterr().out
+    assert "vercel.com/d?to=%2F%5Bteam%5D%2F%7E%2Fai-gateway&title=AI+Gateway" in out
+    assert "free credits" in out.lower()
+    assert prompt_selection.call_args.kwargs["pricing"] == pricing
+
+    import yaml
+    config = yaml.safe_load((config_home / "config.yaml").read_text()) or {}
+    model = config["model"]
+    assert model["provider"] == "ai-gateway"
+    assert model["api_mode"] == "chat_completions"

tests/run_agent/test_provider_attribution_headers.py

@@ -0,0 +1,62 @@
+"""Attribution default_headers applied per provider via base-URL detection."""
+
+from unittest.mock import MagicMock, patch
+
+from run_agent import AIAgent
+
+
+@patch("run_agent.OpenAI")
+def test_openrouter_base_url_applies_or_headers(mock_openai):
+    mock_openai.return_value = MagicMock()
+    agent = AIAgent(
+        api_key="test-key",
+        base_url="https://openrouter.ai/api/v1",
+        model="test/model",
+        quiet_mode=True,
+        skip_context_files=True,
+        skip_memory=True,
+    )
+
+    agent._apply_client_headers_for_base_url("https://openrouter.ai/api/v1")
+
+    headers = agent._client_kwargs["default_headers"]
+    assert headers["HTTP-Referer"] == "https://hermes-agent.nousresearch.com"
+    assert headers["X-OpenRouter-Title"] == "Hermes Agent"
+
+
+@patch("run_agent.OpenAI")
+def test_ai_gateway_base_url_applies_attribution_headers(mock_openai):
+    mock_openai.return_value = MagicMock()
+    agent = AIAgent(
+        api_key="test-key",
+        base_url="https://openrouter.ai/api/v1",
+        model="test/model",
+        quiet_mode=True,
+        skip_context_files=True,
+        skip_memory=True,
+    )
+
+    agent._apply_client_headers_for_base_url("https://ai-gateway.vercel.sh/v1")
+
+    headers = agent._client_kwargs["default_headers"]
+    assert headers["HTTP-Referer"] == "https://hermes-agent.nousresearch.com"
+    assert headers["X-Title"] == "Hermes Agent"
+    assert headers["User-Agent"].startswith("HermesAgent/")
+
+
+@patch("run_agent.OpenAI")
+def test_unknown_base_url_clears_default_headers(mock_openai):
+    mock_openai.return_value = MagicMock()
+    agent = AIAgent(
+        api_key="test-key",
+        base_url="https://openrouter.ai/api/v1",
+        model="test/model",
+        quiet_mode=True,
+        skip_context_files=True,
+        skip_memory=True,
+    )
+    agent._client_kwargs["default_headers"] = {"X-Stale": "yes"}
+
+    agent._apply_client_headers_for_base_url("https://api.example.com/v1")
+
+    assert "default_headers" not in agent._client_kwargs

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