feat(xai-oauth): add xAI Grok OAuth (SuperGrok Subscription) provider

Adds a new authentication provider that lets SuperGrok subscribers sign
in to Hermes with their xAI account via the standard OAuth 2.0 PKCE
loopback flow, instead of pasting a raw API key from console.x.ai.

Highlights
----------
* OAuth 2.0 PKCE loopback login against accounts.x.ai with discovery,
  state/nonce, and a strict CORS-origin allowlist on the callback.
* Authorize URL carries `plan=generic` (required for non-allowlisted
  loopback clients) and `referrer=hermes-agent` for best-effort
  attribution in xAI's OAuth server logs.
* Token storage in `auth.json` with file-locked atomic writes; JWT
  `exp`-based expiry detection with skew; refresh-token rotation
  synced both ways between the singleton store and the credential
  pool so multi-process / multi-profile setups don't tear each other's
  refresh tokens.
* Reactive 401 retry: on a 401 from the xAI Responses API, the agent
  refreshes the token, swaps it back into `self.api_key`, and retries
  the call once. Guarded against silent account swaps when the active
  key was sourced from a different (manual) pool entry.
* Auxiliary tasks (curator, vision, embeddings, etc.) route through a
  dedicated xAI Responses-mode auxiliary client instead of falling back
  to OpenRouter billing.
* Direct HTTP tools (`tools/xai_http.py`, transcription, TTS, image-gen
  plugin) resolve credentials through a unified runtime → singleton →
  env-var fallback chain so xai-oauth users get them for free.
* `hermes auth add xai-oauth` and `hermes auth remove xai-oauth N` are
  wired through the standard auth-commands surface; remove cleans up
  the singleton loopback_pkce entry so it doesn't silently reinstate.
* `hermes model` provider picker shows
  "xAI Grok OAuth (SuperGrok Subscription)" and the model-flow falls
  back to pool credentials when the singleton is missing.

Hardening
---------
* Discovery and refresh responses validate the returned
  `token_endpoint` host against the same `*.x.ai` allowlist as the
  authorization endpoint, blocking MITM persistence of a hostile
  endpoint.
* Discovery / refresh / token-exchange `response.json()` calls are
  wrapped to raise typed `AuthError` on malformed bodies (captive
  portals, proxy error pages) instead of leaking JSONDecodeError
  tracebacks.
* `prompt_cache_key` is routed through `extra_body` on the codex
  transport (sending it as a top-level kwarg trips xAI's SDK with a
  TypeError).
* Credential-pool sync-back preserves `active_provider` so refreshing
  an OAuth entry doesn't silently flip the active provider out from
  under the running agent.

Testing
-------
* New `tests/hermes_cli/test_auth_xai_oauth_provider.py` (~63 tests)
  covers JWT expiry, OAuth URL params (plan + referrer), CORS origins,
  redirect URI validation, singleton↔pool sync, concurrency races,
  refresh error paths, runtime resolution, and malformed-JSON guards.
* Extended `test_credential_pool.py`, `test_codex_transport.py`, and
  `test_run_agent_codex_responses.py` cover the pool sync-back,
  `extra_body` routing, and 401 reactive refresh paths.
* 165 tests passing on this branch via `scripts/run_tests.sh`.
This commit is contained in:
Jaaneek 2026-05-15 16:10:38 +01:00 committed by Teknium
parent 9fb40e6a3d
commit b62c997973
27 changed files with 3843 additions and 131 deletions

View file

@ -266,10 +266,12 @@ def _get_provider(stt_config: dict) -> str:
return "none"
if provider == "xai":
if get_env_value("XAI_API_KEY"):
from tools.xai_http import resolve_xai_http_credentials
if resolve_xai_http_credentials().get("api_key"):
return "xai"
logger.warning(
"STT provider 'xai' configured but XAI_API_KEY not set"
"STT provider 'xai' configured but no xAI credentials are available"
)
return "none"
@ -289,9 +291,14 @@ def _get_provider(stt_config: dict) -> str:
if _HAS_OPENAI and _has_openai_audio_backend():
logger.info("No local STT available, using OpenAI Whisper API")
return "openai"
if get_env_value("XAI_API_KEY"):
logger.info("No local STT available, using xAI Grok STT API")
return "xai"
try:
from tools.xai_http import resolve_xai_http_credentials
if resolve_xai_http_credentials().get("api_key"):
logger.info("No local STT available, using xAI Grok STT API")
return "xai"
except Exception:
pass
return "none"
# ---------------------------------------------------------------------------
@ -704,14 +711,22 @@ def _transcribe_xai(file_path: str, model_name: str) -> Dict[str, Any]:
Supports Inverse Text Normalization, diarization, and word-level timestamps.
Requires ``XAI_API_KEY`` environment variable.
"""
api_key = get_env_value("XAI_API_KEY")
from tools.xai_http import resolve_xai_http_credentials
creds = resolve_xai_http_credentials()
api_key = str(creds.get("api_key") or "").strip()
if not api_key:
return {"success": False, "transcript": "", "error": "XAI_API_KEY not set"}
return {
"success": False,
"transcript": "",
"error": "No xAI credentials found. Configure xAI OAuth in `hermes model` or set XAI_API_KEY",
}
stt_config = _load_stt_config()
xai_config = stt_config.get("xai", {})
base_url = str(
xai_config.get("base_url")
or creds.get("base_url")
or get_env_value("XAI_STT_BASE_URL")
or XAI_STT_BASE_URL
).strip().rstrip("/")
@ -872,7 +887,7 @@ def transcribe_audio(file_path: str, model: Optional[str] = None) -> Dict[str, A
"No STT provider available. Install faster-whisper for free local "
f"transcription, configure {LOCAL_STT_COMMAND_ENV} or install a local whisper CLI, "
"set GROQ_API_KEY for free Groq Whisper, set MISTRAL_API_KEY for Mistral "
"Voxtral Transcribe, set XAI_API_KEY for xAI Grok STT, or set VOICE_TOOLS_OPENAI_KEY "
"Voxtral Transcribe, configure xAI OAuth or set XAI_API_KEY for xAI Grok STT, or set VOICE_TOOLS_OPENAI_KEY "
"or OPENAI_API_KEY for the OpenAI Whisper API."
),
}

View file

@ -9,7 +9,7 @@ Built-in TTS providers:
- MiniMax TTS: High-quality with voice cloning, needs MINIMAX_API_KEY
- Mistral (Voxtral TTS): Multilingual, native Opus, needs MISTRAL_API_KEY
- Google Gemini TTS: Controllable, 30 prebuilt voices, needs GEMINI_API_KEY
- xAI TTS: Grok voices, needs XAI_API_KEY
- xAI TTS: Grok voices, uses xAI Grok OAuth credentials or XAI_API_KEY
- NeuTTS (local, free, no API key): On-device TTS via neutts
- KittenTTS (local, free, no API key): On-device 25MB model
- Piper (local, free, no API key): OHF-Voice/piper1-gpl neural VITS, 44 languages
@ -902,9 +902,12 @@ def _generate_xai_tts(text: str, output_path: str, tts_config: Dict[str, Any]) -
"""
import requests
api_key = (get_env_value("XAI_API_KEY") or "").strip()
from tools.xai_http import resolve_xai_http_credentials
creds = resolve_xai_http_credentials()
api_key = str(creds.get("api_key") or "").strip()
if not api_key:
raise ValueError("XAI_API_KEY not set. Get one at https://console.x.ai/")
raise ValueError("No xAI credentials found. Configure xAI OAuth in `hermes model` or set XAI_API_KEY.")
xai_config = tts_config.get("xai", {})
voice_id = str(xai_config.get("voice_id", DEFAULT_XAI_VOICE_ID)).strip() or DEFAULT_XAI_VOICE_ID
@ -913,6 +916,7 @@ def _generate_xai_tts(text: str, output_path: str, tts_config: Dict[str, Any]) -
bit_rate = int(xai_config.get("bit_rate", DEFAULT_XAI_BIT_RATE))
base_url = str(
xai_config.get("base_url")
or creds.get("base_url")
or get_env_value("XAI_BASE_URL")
or DEFAULT_XAI_BASE_URL
).strip().rstrip("/")
@ -1917,8 +1921,13 @@ def check_tts_requirements() -> bool:
pass
if get_env_value("MINIMAX_API_KEY"):
return True
if get_env_value("XAI_API_KEY"):
return True
try:
from tools.xai_http import resolve_xai_http_credentials
if resolve_xai_http_credentials().get("api_key"):
return True
except Exception:
pass
if get_env_value("GEMINI_API_KEY") or get_env_value("GOOGLE_API_KEY"):
return True
try:

View file

@ -2,6 +2,9 @@
from __future__ import annotations
import os
from typing import Dict
def hermes_xai_user_agent() -> str:
"""Return a stable Hermes-specific User-Agent for xAI HTTP calls."""
@ -10,3 +13,49 @@ def hermes_xai_user_agent() -> str:
except Exception:
__version__ = "unknown"
return f"Hermes-Agent/{__version__}"
def resolve_xai_http_credentials() -> Dict[str, str]:
"""Resolve bearer credentials for direct xAI HTTP endpoints.
Prefers Hermes-managed xAI OAuth credentials when available, then falls back
to ``XAI_API_KEY`` from the environment. This keeps direct xAI endpoints
(images, TTS, STT, etc.) aligned with the main runtime auth model.
"""
try:
from hermes_cli.runtime_provider import resolve_runtime_provider
runtime = resolve_runtime_provider(requested="xai-oauth")
access_token = str(runtime.get("api_key") or "").strip()
base_url = str(runtime.get("base_url") or "").strip().rstrip("/")
if access_token:
return {
"provider": "xai-oauth",
"api_key": access_token,
"base_url": base_url or "https://api.x.ai/v1",
}
except Exception:
pass
try:
from hermes_cli.auth import resolve_xai_oauth_runtime_credentials
creds = resolve_xai_oauth_runtime_credentials()
access_token = str(creds.get("api_key") or "").strip()
base_url = str(creds.get("base_url") or "").strip().rstrip("/")
if access_token:
return {
"provider": "xai-oauth",
"api_key": access_token,
"base_url": base_url or "https://api.x.ai/v1",
}
except Exception:
pass
api_key = os.getenv("XAI_API_KEY", "").strip()
base_url = (os.getenv("XAI_BASE_URL") or "https://api.x.ai/v1").strip().rstrip("/")
return {
"provider": "xai",
"api_key": api_key,
"base_url": base_url,
}