refactor(agent/models): table-driven credits header parsing, unified account-usage fetchers

- credits_tracker: _MICROS_FIELDS/_USD_FIELDS/_BOOL_FIELDS drive parse_credits_headers;
  _sticky_notice() replaces repeated AgentNotice literals; collapsed defensive layers
- account_usage: _USAGE_FETCHERS dispatch table replaces provider if/elif; shared
  fetch/parse helpers across the codex/anthropic/openrouter fetchers; drop dead
  _resolve_codex_usage_url
- billing_links, fast_mode: compacted docstrings, collapsed if/return chains
- model_metadata: fold effective_provider inference into one expression
Notice keys, header names, URLs, log strings and dataclass defaults unchanged.

agent/billing_links.py

@@ -1,11 +1,10 @@
 """Provider-agnostic billing/credit recovery links.
 
-Maps a billing-classified failure onto a recovery link + label. *Detection*
-is not done here — that is :mod:`agent.error_classifier`
-(``FailoverReason.billing``), the single source of truth for "credit wall vs.
-rate limit / auth / transport". The resulting :class:`BillingBlock` rides the
-turn result and the gateway ``message.complete`` event so every surface (CLI,
-TUI, desktop) renders one structured signal instead of re-parsing error text.
+Maps a billing-classified failure onto a recovery link + label. *Detection* is
+:mod:`agent.error_classifier` (``FailoverReason.billing``), the single source of
+truth for "credit wall vs. rate limit / auth / transport". The :class:`BillingBlock`
+rides the turn result and the gateway ``message.complete`` event so every surface
+renders one structured signal instead of re-parsing error text.
 """
 
 from __future__ import annotations
@@ -20,10 +19,9 @@ from utils import base_url_host_matches
 class BillingBlock:
     """Structured billing-wall descriptor shared across every surface.
 
-    ``is_nous`` is the routing bit: Nous has a first-class in-app billing surface
-    (desktop Settings → Billing, TUI/CLI ``/topup``), so surfaces prefer that over
-    ``billing_url``; third-party providers have no in-app flow, so ``billing_url``
-    is the deep link the user actually needs.
+    ``is_nous`` is the routing bit: Nous has an in-app billing surface (desktop
+    Settings → Billing, TUI/CLI ``/topup``) preferred over ``billing_url``;
+    third-party providers have none, so ``billing_url`` is the deep link needed.
     """
 
     provider: str
@@ -45,11 +43,10 @@ class _Provider:
     hosts: tuple[str, ...] = ()
 
 
-# Single source of truth: internal slug(s) + base_url host(s) → billing page.
-# Curated "add credits / manage billing" landing pages, not marketing homes.
-# Hosts back the OpenAI-compatible fallback where the slug is a generic bucket
-# (e.g. "openai_compatible") but base_url reveals the real upstream. An unknown
-# provider degrades to a readable label with no invented URL.
+# Single source of truth: slug(s) + base_url host(s) → curated "add credits"
+# page (not marketing homes). Hosts back the OpenAI-compatible fallback where the
+# slug is a generic bucket but base_url reveals the real upstream. Unknown
+# providers degrade to a readable label with no invented URL.
 _PROVIDERS: tuple[_Provider, ...] = (
     _Provider("OpenAI", "https://platform.openai.com/settings/organization/billing", ("openai",), ("api.openai.com",)),
     _Provider("Anthropic", "https://console.anthropic.com/settings/billing", ("anthropic",), ("api.anthropic.com",)),
@@ -72,16 +69,15 @@ _BY_SLUG: dict[str, _Provider] = {slug: p for p in _PROVIDERS for slug in p.slug
 
 def is_nous_inference_route(provider: str, base_url: str) -> bool:
     """True when the failing route is the Nous-managed inference gateway."""
-    if (provider or "").strip().lower() == "nous":
-        return True
-    return base_url_host_matches(str(base_url or ""), "inference-api.nousresearch.com")
+    return (provider or "").strip().lower() == "nous" or base_url_host_matches(
+        str(base_url or ""), "inference-api.nousresearch.com"
+    )
 
 
 def _nous_billing_url() -> Optional[str]:
     """Best-effort Nous portal billing URL (text-surface fallback; Nous prefers the in-app flow)."""
     try:
         from hermes_cli.nous_account import nous_portal_billing_url
-
         return nous_portal_billing_url(None)
     except Exception:
         return "https://portal.nousresearch.com/billing"
@@ -92,22 +88,14 @@ def _resolve_provider_link(slug: str, base_url: str) -> tuple[str, Optional[str]
     hit = _BY_SLUG.get(slug)
     if hit:
         return hit.label, hit.url
-
     base = str(base_url or "")
     for p in _PROVIDERS:
         if any(base_url_host_matches(base, host) for host in p.hosts):
             return p.label, p.url
-
     return slug.replace("_", " ").replace("-", " ").strip().title() or "your provider", None
 
 
-def build_billing_block(
-    *,
-    provider: str,
-    base_url: str,
-    model: str,
-    message: str = "",
-) -> BillingBlock:
+def build_billing_block(*, provider: str, base_url: str, model: str, message: str = "") -> BillingBlock:
     """Build the billing descriptor for a billing-classified failure.
 
     ``message`` is the guidance already assembled by the agent loop
@@ -116,9 +104,7 @@ def build_billing_block(
     """
     slug = (provider or "").strip().lower()
     model = (model or "").strip()
-
     if is_nous_inference_route(slug, base_url):
         return BillingBlock(slug or "nous", "Nous Portal", model, _nous_billing_url(), True, message or "")
-
     label, url = _resolve_provider_link(slug, base_url)
     return BillingBlock(slug, label, model, url, False, message or "")

agent/fast_mode.py

@@ -1,17 +1,12 @@
 """Bounded fast-mode windows (``/fast auto`` and ``/fast cold``).
 
-``agent.service_tier`` is ``None`` (normal), ``"priority"`` (static fast),
-``"auto"`` or ``"cold"``. The static value is pinned into
-``agent.request_overrides`` at agent build time; the two bounded modes
-instead open a wall-clock window at each user-turn boundary and layer the
-provider's fast override onto the request kwargs only while it is open:
-
-- ``auto`` — every user turn opens a window of ``agent.fast_auto_seconds``.
-- ``cold`` — only the first turn of a session (no prior history) opens it.
-
-Only per-request params (``service_tier`` / ``speed``) vary between requests;
-the system prompt, tools, and messages are untouched, so the prompt cache is
-preserved across the window boundary.
+``agent.service_tier`` is ``None`` (normal), ``"priority"`` (static fast, pinned
+into ``agent.request_overrides`` at build time), ``"auto"`` (every user turn opens
+a window of ``agent.fast_auto_seconds``) or ``"cold"`` (only a session's first
+turn, no prior history, opens it). The provider's fast override is layered onto
+the request kwargs only while the window is open. Only per-request params
+(``service_tier`` / ``speed``) vary — system prompt, tools, and messages are
+untouched, so the prompt cache survives the boundary.
 """
 
 from __future__ import annotations
@@ -44,19 +39,16 @@ def begin_turn(agent: Any, conversation_history: Any) -> None:
 def effective_request_overrides(agent: Any) -> dict[str, Any]:
     """``agent.request_overrides`` plus the fast override while the window is open."""
     overrides = dict(getattr(agent, "request_overrides", None) or {})
-    if getattr(agent, "service_tier", None) not in BOUNDED_MODES:
-        return overrides
-    if time.monotonic() >= getattr(agent, "_fast_until", 0.0):
+    if getattr(agent, "service_tier", None) not in BOUNDED_MODES or time.monotonic() >= getattr(
+        agent, "_fast_until", 0.0
+    ):
         return overrides
     from hermes_cli.models import resolve_fast_mode_overrides
-
     base_url = getattr(agent, "base_url", None)
     if getattr(agent, "api_mode", None) == "anthropic_messages":
         base_url = getattr(agent, "_anthropic_base_url", None) or base_url
     fast = resolve_fast_mode_overrides(
-        getattr(agent, "model", None),
-        provider=getattr(agent, "provider", None),
-        base_url=base_url,
+        getattr(agent, "model", None), provider=getattr(agent, "provider", None), base_url=base_url
     )
     if fast:
         overrides.update(fast)

agent/model_metadata.py

@@ -2714,11 +2714,8 @@ def get_model_context_length(
     # model has different limits per provider (claude-opus-4.6: 1M on
     # Anthropic, 128K on Copilot). Generic providers are inferred from the URL.
     effective_provider = provider
-    if not effective_provider or effective_provider in {"openrouter", "custom"}:
-        if base_url:
-            inferred = _infer_provider_from_url(base_url)
-            if inferred:
-                effective_provider = inferred
+    if base_url and (not effective_provider or effective_provider in {"openrouter", "custom"}):
+        effective_provider = _infer_provider_from_url(base_url) or effective_provider
 
     # 5a. Copilot live /models — account-specific models (claude-opus-4.6-1m)
     # absent from models.dev, and the provider-enforced limit for the rest.