chore: release v1.5.3

fix(llm): send OpenRouter app attribution on the request itself (#1045 )
fix(container): write the browser profile as root
2026-08-16 09:26:39 +02:00 · 2026-08-10 21:28:52 +03:00 · 2026-08-10 11:24:02 -07:00 · 2026-08-10 10:08:17 +03:00 · 2026-08-10 09:54:49 +03:00 · 2026-08-10 09:54:49 +03:00
51 changed files with 1142 additions and 1466 deletions
@@ -117,6 +117,21 @@ ENV AGENT_BROWSER_EXECUTABLE_PATH=/usr/bin/chromium
 ENV AGENT_BROWSER_USER_AGENT="Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36"
 ENV AGENT_BROWSER_ARGS="--disable-blink-features=AutomationControlled,--no-first-run,--no-default-browser-check,--lang=en-US"
 ENV AGENT_BROWSER_SCREENSHOT_DIR=/workspace/.agent-browser-screenshots
+ENV AGENT_BROWSER_IDLE_TIMEOUT_MS=180000
+USER root
+RUN set -eu; \
+    { \
+        for var in AGENT_BROWSER_EXECUTABLE_PATH AGENT_BROWSER_USER_AGENT \
+                   AGENT_BROWSER_ARGS AGENT_BROWSER_SCREENSHOT_DIR \
+                   AGENT_BROWSER_IDLE_TIMEOUT_MS; do \
+            eval "value=\${$var}"; \
+            printf 'export %s="${%s:-%s}"\n' "$var" "$var" "$value"; \
+        done; \
+    } > /tmp/agent-browser.sh; \
+    install -m 0644 /tmp/agent-browser.sh /etc/profile.d/agent-browser.sh; \
+    rm /tmp/agent-browser.sh; \
+    env -i bash -lc 'test "${AGENT_BROWSER_IDLE_TIMEOUT_MS}" = "180000"'
+USER pentester
 RUN /home/pentester/.npm-global/bin/agent-browser doctor --offline --quick

 RUN set -eux; \
@@ -1,6 +1,6 @@
 [project]
 name = "strix-agent"
-version = "1.5.1"
+version = "1.5.3"
 description = "Open-source AI Hackers for your apps"
 readme = "README.md"
 license = "Apache-2.0"
@@ -230,7 +230,6 @@ ignore = [
 # args they intentionally ignore.
 "tests/test_viewer_auth.py" = ["S105", "S106", "ARG001"]
 "tests/test_codex_auth.py" = ["S105", "S106", "SLF001"]
-"tests/test_grok_auth.py" = ["S105", "S106", "SLF001"]
 # Hatchling loads the build hook by path, not as an importable package.
 "scripts/tui_sidecar_hook.py" = ["INP001"]
 # Stdlib HTTP handler overrides (do_GET/do_POST).
@@ -245,9 +244,6 @@ ignore = [
 # Stdlib HTTP handler overrides (do_GET/do_POST) and lazy imports that avoid a
 # circular dependency with strix.telemetry / strix.interface.viewer.report_pdf.
 "strix/interface/viewer/server.py" = ["N802", "PLC0415"]
-# Lazy import of the TUI live-view projection so importing the viewer does not
-# eagerly pull in the Textual TUI.
-"strix/interface/viewer/transcript.py" = ["PLC0415"]
 # Lazy telemetry import to avoid importing PostHog before the viewer starts.
 "strix/interface/viewer/cli.py" = ["PLC0415"]
 # Lazy imports inside functions to avoid circular dependency with
@@ -292,9 +288,6 @@ ignore = [
 # Heavy inference deps (httpx, openai) imported lazily so auth-status checks
 # don't pull them in.
 "strix/config/codex.py" = ["PLC0415"]
-"strix/config/grok.py" = ["PLC0415"]
-# Lazy ``import fcntl`` so the module imports on non-POSIX platforms.
-"strix/config/subscription_store.py" = ["PLC0415"]
 # Interface utility branches per scope-mode / target-type combination;
 # splitting would obscure the decision tree without simplifying it.
 "strix/interface/utils.py" = ["PLR0912", "BLE001", "PLC0415"]
@@ -160,7 +160,9 @@ def _schema_types(spec: dict[str, Any]) -> set[str]:
 def _decode_structured(value: str, types: set[str]) -> Any:
    stripped = value.strip()
    if not stripped:
-        return value
+        # An empty string is the model's "no value" for a list/dict param; give it
+        # the empty container so it validates instead of failing the type check.
+        return [] if "array" in types else {}
    try:
        decoded = json.loads(stripped)
    except json.JSONDecodeError:
@@ -39,6 +39,8 @@ INTERACTIVE BEHAVIOR:
  - To end the whole engagement, call the lifecycle tool: finish_scan (root) or agent_finish (subagent).
  - A turn that ends with plain text and no tool call does NOT stop you: the system nudges you to continue and will re-run you. Do not rely on going silent to pause — it will not pause you.
 - Answering a user question: put the answer in respond_to_user's message. Do not write the answer as plain text and then fall silent — that does not reach a stopping point, it just triggers a continuation nudge.
+- If all you want to do is reply and stop, that whole turn is ONE respond_to_user call carrying the answer. Do not write the answer as text and then call respond_to_user as well: the user reads it twice.
+- If you do end a turn on plain text and the nudge arrives, your words already reached the user. Do not restate them: call respond_to_user with NO message to simply wait, or with only whatever you still need to add.
 - You may include brief explanatory text before a tool call, and you can narrate while you work — plain text is shown to the user as you go. Narrating is free; respond_to_user is specifically the act of WAITING for the user, so do not call it just to give a status update.
 - Respond naturally when the user asks questions or gives instructions.
 - While actively working on a task, every turn should carry exactly one tool call — use think to plan, the appropriate tool to act, and respond_to_user only when you genuinely need the user.
@@ -261,7 +263,13 @@ Remember: A single well-validated high-impact vulnerability is worth more than d
 <multi_agent_system>
 AGENT ISOLATION & SANDBOXING:
 - All agents run in the same shared Docker container for efficiency
- Each agent has its own: browser sessions, terminal sessions
+- Each agent has its own terminal sessions
+- Browsers are NOT per-agent by default: `agent-browser` with no `--session` is one
+  shared browser, so a concurrent agent's navigation invalidates your page and refs.
+  Pass `--session <your-agent-name>` for any browser work of your own — then it is
+  yours alone. Each session is a full Chromium (~340 MB) on this shared box, so keep
+  one, not several, and `agent-browser --session <name> close` when you're done with
+  the target; an idle browser is reclaimed automatically after 3 minutes
 - All agents share the same /workspace directory and proxy history
 - Agents can see each other's files and proxy traffic for better collaboration

@@ -16,6 +16,7 @@ import hashlib
 import json
 import logging
 import secrets
+import threading
 import time
 import urllib.parse
 from pathlib import Path
@@ -23,7 +24,7 @@ from typing import TYPE_CHECKING, Any

 import requests

-from strix.config import subscription_store
+from strix.utils.secret_files import write_secret_text


 if TYPE_CHECKING:
@@ -53,12 +54,26 @@ _ACCOUNT_CLAIM = "https://api.openai.com/auth"
 _TOKEN_TIMEOUT = 30
 _EXPIRY_SKEW_S = 300

+_refresh_lock = threading.Lock()
+
 # Kept separate from cli-config.json so OAuth tokens never land in the env-var config.
 AUTH_PATH = Path.home() / ".strix" / "subscription-auth.json"


+def _read_store() -> dict[str, Any]:
+    try:
+        data = json.loads(AUTH_PATH.read_text(encoding="utf-8"))
+    except (OSError, json.JSONDecodeError):
+        return {}
+    return data if isinstance(data, dict) else {}
+
+
+def _write_store(data: dict[str, Any]) -> None:
+    write_secret_text(AUTH_PATH, json.dumps(data, indent=2))
+
+
 def read_record() -> dict[str, Any] | None:
-    record = subscription_store.read(AUTH_PATH).get(PROVIDER)
+    record = _read_store().get(PROVIDER)
    if not isinstance(record, dict) or record.get("type") != "oauth":
        return None
    if not (record.get("access") and record.get("refresh") and record.get("account_id")):
@@ -71,31 +86,45 @@ def is_authenticated() -> bool:


 def save_record(record: dict[str, Any]) -> None:
-    with subscription_store.guard(AUTH_PATH):
-        data = subscription_store.read(AUTH_PATH)
-        data[PROVIDER] = record
-        subscription_store.write(AUTH_PATH, data)
+    data = _read_store()
+    data[PROVIDER] = record
+    _write_store(data)


 def logout() -> None:
-    with subscription_store.guard(AUTH_PATH):
-        data = subscription_store.read(AUTH_PATH)
-        if PROVIDER not in data:
-            return
-        del data[PROVIDER]
-        if data:
-            subscription_store.write(AUTH_PATH, data)
-            return
-        with contextlib.suppress(OSError):
-            AUTH_PATH.unlink()
+    data = _read_store()
+    if PROVIDER not in data:
+        return
+    del data[PROVIDER]
+    if data:
+        _write_store(data)
+        return
+    with contextlib.suppress(OSError):
+        AUTH_PATH.unlink()


@contextlib.contextmanager
 def _refresh_guard() -> Iterator[None]:
    """Serialize token refresh within (lock) and across (flock) Strix processes,
    so concurrent runs can't both spend the single-use refresh token."""
-    with subscription_store.guard(AUTH_PATH):
-        yield
+    with _refresh_lock:
+        try:
+            import fcntl
+
+            lock_path = AUTH_PATH.with_suffix(".lock")
+            lock_path.parent.mkdir(parents=True, exist_ok=True)
+            handle = lock_path.open("w")
+        except (ImportError, OSError):
+            yield
+            return
+        try:
+            with contextlib.suppress(OSError):
+                fcntl.flock(handle.fileno(), fcntl.LOCK_EX)
+            yield
+        finally:
+            with contextlib.suppress(OSError):
+                fcntl.flock(handle.fileno(), fcntl.LOCK_UN)
+            handle.close()


 class CodexAuthError(Exception):
@@ -1,310 +0,0 @@
-"""Grok (xAI) subscription auth: OAuth login, token refresh, and the OpenAI
-client that routes inference through xAI's API.
-
-Mirrors xAI's Grok CLI: OAuth 2.0 + PKCE against ``auth.x.ai``, with the access
-token sent as a ``Bearer`` token to ``api.x.ai/v1`` (OpenAI-compatible, so the
-subscription and a metered API key share one endpoint — only the bearer differs).
-Using a Grok/SuperGrok subscription outside xAI's own products is not officially
-supported by xAI; the user chooses this path knowingly. The OAuth constants are
-xAI's own Grok CLI values (the backend only accepts that client).
-"""
-
-from __future__ import annotations
-
-import base64
-import contextlib
-import hashlib
-import json
-import logging
-import secrets
-import time
-import urllib.parse
-from pathlib import Path
-from typing import TYPE_CHECKING, Any
-
-import requests
-
-from strix.config import subscription_store
-
-
-if TYPE_CHECKING:
-    from collections.abc import Iterator
-
-    from openai import AsyncOpenAI
-
-
-logger = logging.getLogger(__name__)
-
-
-PROVIDER = "grok"
-
-CLIENT_ID = "b1a00492-073a-47ea-816f-4c329264a828"
-AUTHORIZE_URL = "https://auth.x.ai/oauth2/authorize"
-TOKEN_URL = "https://auth.x.ai/oauth2/token"  # noqa: S105  # nosec B105 - URL, not a secret
-CALLBACK_HOST = "127.0.0.1"
-CALLBACK_PORT = 56121
-CALLBACK_PATH = "/callback"
-REDIRECT_URI = f"http://{CALLBACK_HOST}:{CALLBACK_PORT}{CALLBACK_PATH}"
-SCOPE = "openid profile email offline_access grok-cli:access api:access"
-
-XAI_BASE_URL = "https://api.x.ai/v1"
-
-_TOKEN_TIMEOUT = 30
-_EXPIRY_SKEW_S = 300
-
-# Shared with the other subscription providers; kept separate from cli-config.json
-# so OAuth tokens never land in the env-var config.
-AUTH_PATH = Path.home() / ".strix" / "subscription-auth.json"
-
-
-def read_record() -> dict[str, Any] | None:
-    record = subscription_store.read(AUTH_PATH).get(PROVIDER)
-    if not isinstance(record, dict) or record.get("type") != "oauth":
-        return None
-    if not (record.get("access") and record.get("refresh")):
-        return None
-    return record
-
-
-def is_authenticated() -> bool:
-    return read_record() is not None
-
-
-def save_record(record: dict[str, Any]) -> None:
-    with subscription_store.guard(AUTH_PATH):
-        data = subscription_store.read(AUTH_PATH)
-        data[PROVIDER] = record
-        subscription_store.write(AUTH_PATH, data)
-
-
-def logout() -> None:
-    with subscription_store.guard(AUTH_PATH):
-        data = subscription_store.read(AUTH_PATH)
-        if PROVIDER not in data:
-            return
-        del data[PROVIDER]
-        if data:
-            subscription_store.write(AUTH_PATH, data)
-            return
-        with contextlib.suppress(OSError):
-            AUTH_PATH.unlink()
-
-
-@contextlib.contextmanager
-def _refresh_guard() -> Iterator[None]:
-    """Serialize token refresh within (lock) and across (flock) Strix processes,
-    so concurrent runs can't both spend the single-use refresh token."""
-    with subscription_store.guard(AUTH_PATH):
-        yield
-
-
-class GrokAuthError(Exception):
-    def __init__(self, code: str, message: str | None = None) -> None:
-        self.code = code
-        super().__init__(message or code)
-
-
-def _b64url(raw: bytes) -> str:
-    return base64.urlsafe_b64encode(raw).rstrip(b"=").decode("ascii")
-
-
-def generate_pkce() -> tuple[str, str]:
-    verifier = _b64url(secrets.token_bytes(64))
-    challenge = _b64url(hashlib.sha256(verifier.encode("ascii")).digest())
-    return verifier, challenge
-
-
-def create_state() -> str:
-    return secrets.token_hex(16)
-
-
-def build_authorize_url(challenge: str, state: str) -> str:
-    params = {
-        "response_type": "code",
-        "client_id": CLIENT_ID,
-        "redirect_uri": REDIRECT_URI,
-        "scope": SCOPE,
-        "code_challenge": challenge,
-        "code_challenge_method": "S256",
-        "state": state,
-    }
-    return f"{AUTHORIZE_URL}?{urllib.parse.urlencode(params)}"
-
-
-def parse_redirect_input(value: str) -> tuple[str | None, str | None]:
-    """Extract ``(code, state)`` from a pasted redirect URL, ``code#state``,
-    query string, or bare code."""
-    value = (value or "").strip()
-    if not value:
-        return None, None
-    with contextlib.suppress(ValueError):
-        parsed = urllib.parse.urlparse(value)
-        if parsed.scheme and parsed.query:
-            query = urllib.parse.parse_qs(parsed.query)
-            return _first(query, "code"), _first(query, "state")
-    if "#" in value:
-        code, _, state = value.partition("#")
-        return code or None, state or None
-    if "code=" in value:
-        query = urllib.parse.parse_qs(value)
-        return _first(query, "code"), _first(query, "state")
-    return value, None
-
-
-def _first(query: dict[str, list[str]], key: str) -> str | None:
-    values = query.get(key)
-    return values[0] if values else None
-
-
-def _post_form(payload: dict[str, str]) -> dict[str, Any]:
-    try:
-        response = requests.post(
-            TOKEN_URL,
-            data=payload,
-            headers={"Accept": "application/json"},
-            timeout=_TOKEN_TIMEOUT,
-        )
-    except requests.RequestException as exc:
-        raise GrokAuthError("unavailable", str(exc)) from exc
-    if response.status_code >= 400:
-        detail = response.text[:300]
-        raise GrokAuthError("token_http_error", f"HTTP {response.status_code}: {detail}")
-    data = json.loads(response.content or b"{}")
-    if not isinstance(data, dict):
-        raise GrokAuthError("bad_response", "token endpoint returned non-object")
-    return data
-
-
-def _record_from_token_response(
-    data: dict[str, Any], refresh_fallback: str | None = None
-) -> dict[str, Any]:
-    access = data.get("access_token")
-    # A refresh response may omit refresh_token when it isn't rotated; keep the old one.
-    refresh = data.get("refresh_token") or refresh_fallback
-    expires_in = data.get("expires_in")
-    if not isinstance(access, str) or not access:
-        raise GrokAuthError("bad_response", "token response missing access_token")
-    if not isinstance(refresh, str) or not refresh:
-        raise GrokAuthError("bad_response", "token response missing refresh_token")
-    ttl = expires_in if isinstance(expires_in, int | float) else 3600
-    return {
-        "type": "oauth",
-        "provider": PROVIDER,
-        "access": access,
-        "refresh": refresh,
-        "expires_at": time.time() + ttl,
-    }
-
-
-def exchange_code(code: str, verifier: str) -> dict[str, Any]:
-    data = _post_form(
-        {
-            "grant_type": "authorization_code",
-            "client_id": CLIENT_ID,
-            "code": code,
-            "code_verifier": verifier,
-            "redirect_uri": REDIRECT_URI,
-        }
-    )
-    return _record_from_token_response(data)
-
-
-def refresh_tokens(refresh_token: str) -> dict[str, Any]:
-    data = _post_form(
-        {
-            "grant_type": "refresh_token",
-            "client_id": CLIENT_ID,
-            "refresh_token": refresh_token,
-        }
-    )
-    return _record_from_token_response(data, refresh_fallback=refresh_token)
-
-
-def _access_token(record: dict[str, Any]) -> str:
-    access = record["access"]
-    if not isinstance(access, str) or not access:
-        raise GrokAuthError("bad_response", "stored access token is missing or malformed")
-    return access
-
-
-def _near_expiry(record: dict[str, Any]) -> bool:
-    expires_at = record.get("expires_at")
-    if not isinstance(expires_at, int | float):
-        return True
-    return expires_at - _EXPIRY_SKEW_S <= time.time()
-
-
-def get_valid_token() -> str:
-    """Return a valid access token, refreshing under the cross-process guard if
-    near expiry."""
-    record = read_record()
-    if record is None:
-        raise GrokAuthError("not_authenticated", "not signed in; run: strix auth login grok")
-    if not _near_expiry(record):
-        return _access_token(record)
-    with _refresh_guard():
-        record = read_record()
-        if record is None:
-            raise GrokAuthError("not_authenticated", "not signed in; run: strix auth login grok")
-        if not _near_expiry(record):
-            return _access_token(record)
-        try:
-            refreshed = refresh_tokens(record["refresh"])
-        except GrokAuthError:
-            # A peer process may have already spent this single-use refresh token.
-            latest = read_record()
-            if latest and latest["refresh"] != record["refresh"] and not _near_expiry(latest):
-                return _access_token(latest)
-            raise
-        save_record(refreshed)
-        return _access_token(refreshed)
-
-
-def build_openai_client() -> AsyncOpenAI:
-    """An ``AsyncOpenAI`` for xAI's API. A per-request hook re-stamps a fresh
-    bearer token so long scans survive token expiry."""
-    import asyncio
-
-    import httpx
-    from openai import AsyncOpenAI
-
-    get_valid_token()  # fail fast at configure time if the sign-in is dead
-
-    async def _auth_hook(request: httpx.Request) -> None:
-        access = await asyncio.to_thread(get_valid_token)
-        request.headers["Authorization"] = f"Bearer {access}"
-
-    http_client = httpx.AsyncClient(
-        timeout=httpx.Timeout(600.0, connect=30.0),
-        event_hooks={"request": [_auth_hook]},
-    )
-    return AsyncOpenAI(
-        api_key="strix-grok-oauth",  # placeholder; the hook overwrites Authorization
-        base_url=XAI_BASE_URL,
-        http_client=http_client,
-    )
-
-
-_subscription_client: AsyncOpenAI | None = None
-
-
-def get_subscription_client() -> AsyncOpenAI:
-    global _subscription_client  # noqa: PLW0603
-    if _subscription_client is None:
-        _subscription_client = build_openai_client()
-    return _subscription_client
-
-
-SUBSCRIPTION_PREFIX = "grok/"
-
-
-def subscription_model(model_name: str | None) -> str | None:
-    """The model slug behind a ``grok/<model>`` STRIX_LLM, or None."""
-    name = (model_name or "").strip()
-    if not name.lower().startswith(SUBSCRIPTION_PREFIX):
-        return None
-    return name[len(SUBSCRIPTION_PREFIX) :] or None
-
-
-def auth_mode(model_name: str | None) -> str:
-    return "subscription" if subscription_model(model_name) else "api_key"
@@ -20,7 +20,6 @@ from agents.model_settings import ModelSettings
 from agents.models.fake_id import FAKE_RESPONSES_ID
 from agents.models.interface import Model
 from agents.models.multi_provider import MultiProvider
-from agents.models.openai_chatcompletions import OpenAIChatCompletionsModel
 from agents.models.openai_responses import OpenAIResponsesModel
 from agents.retry import (
    ModelRetryBackoffSettings,
@@ -37,7 +36,7 @@ from openai.types.responses import (
 from openai.types.responses.response_usage import ResponseUsage
 from openai.types.shared import Reasoning

-from strix.config import codex, grok
+from strix.config import codex
 from strix.config.loader import load_settings
 from strix.config.tool_call_ids import TurnCallIdRewriter, dedupe_input
 from strix.config.tool_call_limits import TurnToolCallLimiter
@@ -482,10 +481,6 @@ class StrixProvider(MultiProvider):
                codex.get_subscription_client(),
                reasoning_effort=llm.reasoning_effort,
            )
-        elif grok_slug := grok.subscription_model(model_name):
-            # xAI's API is OpenAI chat-completions compatible; the subscription
-            # bearer is stamped per-request by the client's auth hook.
-            model = OpenAIChatCompletionsModel(grok_slug, grok.get_subscription_client())
        else:
            model = super().get_model(model_name)
            if llm.disable_streaming:
@@ -561,7 +556,7 @@ def configure_sdk_model_defaults(settings: Settings) -> None:
    """Apply Strix config to SDK-native defaults."""
    llm = settings.llm
    set_tracing_disabled(True)
-    if codex.subscription_model(llm.model) or grok.subscription_model(llm.model):
+    if codex.subscription_model(llm.model):
        return
    _configure_litellm_compatibility()
    _configure_openrouter_attribution(llm.model)
@@ -657,27 +652,31 @@ def _install_openrouter_stream_cost_capture() -> None:
    litellm.OpenrouterConfig = _StrixOpenrouterConfig  # type: ignore[misc]


-_OPENROUTER_ATTRIBUTION_HEADERS = {
+OPENROUTER_ATTRIBUTION_HEADERS = {
    "HTTP-Referer": "https://strix.ai",
    "X-Title": "Strix",
    "X-OpenRouter-Categories": "cli-agent",
 }


+def is_openrouter_model(model_name: str | None) -> bool:
+    return bool(model_name) and "openrouter/" in (model_name or "").strip().lower()
+
+
 def _configure_openrouter_attribution(model_name: str | None) -> None:
    import litellm

    current: object = litellm.headers
    existing: dict[str, str] = current if isinstance(current, dict) else {}
-    if not model_name or "openrouter/" not in model_name.strip().lower():
-        if any(key in existing for key in _OPENROUTER_ATTRIBUTION_HEADERS):
+    if not is_openrouter_model(model_name):
+        if any(key in existing for key in OPENROUTER_ATTRIBUTION_HEADERS):
            remaining = {
-                k: v for k, v in existing.items() if k not in _OPENROUTER_ATTRIBUTION_HEADERS
+                k: v for k, v in existing.items() if k not in OPENROUTER_ATTRIBUTION_HEADERS
            }
            litellm.headers = remaining or None  # type: ignore[assignment]
        return

-    litellm.headers = {**existing, **_OPENROUTER_ATTRIBUTION_HEADERS}  # type: ignore[assignment]
+    litellm.headers = {**existing, **OPENROUTER_ATTRIBUTION_HEADERS}  # type: ignore[assignment]


 def _configure_extra_headers(llm: LlmSettings) -> None:
@@ -1,45 +0,0 @@
-"""Shared helpers across model-subscription providers (ChatGPT/Codex and Grok).
-
-Each provider module (:mod:`strix.config.codex`, :mod:`strix.config.grok`)
-exposes the same small surface — ``subscription_model``, ``auth_mode``,
-``is_authenticated`` — so callers that only care "is this run on a subscription,
-and which provider?" can stay provider-agnostic.
-"""
-
-from __future__ import annotations
-
-from typing import TYPE_CHECKING
-
-from strix.config import codex, grok
-
-
-if TYPE_CHECKING:
-    from types import ModuleType
-
-
-_PROVIDERS: tuple[ModuleType, ...] = (codex, grok)
-
-# Human-facing provider names keyed by each module's ``PROVIDER`` constant.
-_DISPLAY_NAMES: dict[str, str] = {codex.PROVIDER: "ChatGPT", grok.PROVIDER: "Grok"}
-
-
-def provider_for_model(model_name: str | None) -> ModuleType | None:
-    """Return the subscription provider module that owns ``model_name``'s prefix,
-    or None when the model isn't a subscription model."""
-    for provider in _PROVIDERS:
-        if provider.subscription_model(model_name):
-            return provider
-    return None
-
-
-def auth_mode(model_name: str | None) -> str:
-    return "subscription" if provider_for_model(model_name) is not None else "api_key"
-
-
-def provider_label(model_name: str | None) -> str | None:
-    """Human-facing name of the subscription provider for ``model_name`` (e.g.
-    "ChatGPT" or "Grok"), or None when the model isn't a subscription model."""
-    provider = provider_for_model(model_name)
-    if provider is None:
-        return None
-    return _DISPLAY_NAMES.get(provider.PROVIDER)
@@ -1,152 +0,0 @@
-"""Shared on-disk store for subscription OAuth credentials.
-
-Every subscription provider (ChatGPT/Codex, Grok) keeps its record under its own
-key in a single ``~/.strix/subscription-auth.json`` file. Reads and writes go
-through here so that:
-
-* tokens are written owner-only (mode 0600) from the moment the file is created,
-  never briefly exposed with umask-derived permissions, and
-* concurrent read-modify-write mutations — even across different providers or
-  processes — are serialized, so one provider's update can't clobber another's.
-
-The lock is reentrant, so a provider may nest a ``save`` inside a longer
-``guard`` (e.g. refreshing a token then persisting it) without deadlocking.
-"""
-
-from __future__ import annotations
-
-import contextlib
-import json
-import os
-import tempfile
-import threading
-from pathlib import Path
-from typing import TYPE_CHECKING, Any
-
-
-if TYPE_CHECKING:
-    from collections.abc import Iterator
-    from io import TextIOWrapper
-
-
-class StoreLockError(RuntimeError):
-    """The cross-process store lock could not be acquired.
-
-    Raised instead of silently proceeding, so a read-modify-write never runs
-    unlocked (which would let concurrent provider logins/refreshes/logouts race).
-    """
-
-
-def read(path: Path) -> dict[str, Any]:
-    """The store's contents, or an empty dict when absent/unreadable."""
-    try:
-        data = json.loads(path.read_text(encoding="utf-8"))
-    except (OSError, json.JSONDecodeError):
-        return {}
-    return data if isinstance(data, dict) else {}
-
-
-def write(path: Path, data: dict[str, Any]) -> None:
-    """Atomically replace the store, owner-only from creation.
-
-    The temp file is created with a random name via ``mkstemp`` (mode 0600, no
-    symlink following), so a local attacker can't pre-plant a symlink at a
-    predictable path to divert the token write.
-    """
-    path.parent.mkdir(parents=True, exist_ok=True)
-    fd, tmp_name = tempfile.mkstemp(dir=path.parent, prefix=path.name, suffix=".tmp")
-    tmp = Path(tmp_name)
-    try:
-        with os.fdopen(fd, "w", encoding="utf-8") as handle:
-            json.dump(data, handle, indent=2)
-        tmp.replace(path)
-    except BaseException:
-        with contextlib.suppress(OSError):
-            tmp.unlink()
-        raise
-    with contextlib.suppress(OSError):
-        path.chmod(0o600)
-
-
-class _StoreLock:
-    """A reentrant lock serializing store mutations within (thread lock) and
-    across (flock) Strix processes. Nesting reuses the single held file lock, so
-    a provider can persist a record inside a longer critical section."""
-
-    def __init__(self) -> None:
-        self._thread_lock = threading.RLock()
-        self._flock_handle: TextIOWrapper | None = None
-        self._depth = 0
-
-    @contextlib.contextmanager
-    def hold(self, path: Path) -> Iterator[None]:
-        with self._thread_lock:
-            if self._depth == 0:
-                self._flock_handle = _acquire_flock(path)
-            self._depth += 1
-            try:
-                yield
-            finally:
-                self._depth -= 1
-                if self._depth == 0:
-                    self._release_flock()
-
-    def _release_flock(self) -> None:
-        handle = self._flock_handle
-        self._flock_handle = None
-        if handle is None:
-            return
-        try:
-            import fcntl
-
-            with contextlib.suppress(OSError):
-                fcntl.flock(handle.fileno(), fcntl.LOCK_UN)
-        except ImportError:
-            pass
-        finally:
-            handle.close()
-
-
-_store_lock = _StoreLock()
-
-
-def guard(path: Path) -> contextlib.AbstractContextManager[None]:
-    """Serialize store mutation across threads and processes (reentrant)."""
-    return _store_lock.hold(path)
-
-
-def _acquire_flock(path: Path) -> TextIOWrapper:
-    """Hold an exclusive cross-process lock on the store, or raise.
-
-    Never returns without the lock held: a missing ``fcntl`` or a failed
-    ``flock`` raises :class:`StoreLockError` so the caller aborts rather than
-    mutating the store unlocked.
-    """
-    try:
-        import fcntl
-    except ImportError as exc:  # pragma: no cover - non-POSIX
-        msg = "cross-process credential locking requires fcntl (a POSIX platform)"
-        raise StoreLockError(msg) from exc
-    lock_path = path.with_suffix(".lock")
-    lock_path.parent.mkdir(parents=True, exist_ok=True)
-    # O_NOFOLLOW rejects a pre-positioned symlink at the predictable lock path
-    # (so an attacker can't redirect the open), and no O_TRUNC since the lock
-    # file is only an flock anchor whose contents we never use.
-    try:
-        fd = os.open(str(lock_path), os.O_RDWR | os.O_CREAT | os.O_NOFOLLOW, 0o600)
-    except OSError as exc:
-        msg = f"could not open lock file {lock_path}: {exc}"
-        raise StoreLockError(msg) from exc
-    handle = os.fdopen(fd, "r+")
-    try:
-        while True:
-            try:
-                fcntl.flock(handle.fileno(), fcntl.LOCK_EX)
-                break
-            except InterruptedError:  # EINTR — retry the blocking acquire
-                continue
-    except OSError as exc:
-        handle.close()
-        msg = f"could not lock {lock_path}: {exc}"
-        raise StoreLockError(msg) from exc
-    return handle
@@ -830,7 +830,7 @@ async def _append_tool_required_message(
            "execution and never hands control to the user: it is shown to the user, and the "
            "run continues. Continue immediately and call exactly one tool. "
            "If you have something to tell the user and nothing to do until they reply, "
-            "call respond_to_user. "
+            "call respond_to_user — with no message if you have already said it. "
            "If you are blocked waiting for another agent, call wait_for_agents. "
            f"If the whole engagement is complete, call {finish_tool}. "
            "Otherwise use the appropriate execution or planning tool. "
@@ -10,10 +10,12 @@ from openai.types.shared import Reasoning

 from strix.config.models import (
    DEFAULT_MODEL_RETRY,
+    OPENROUTER_ATTRIBUTION_HEADERS,
    bedrock_route_supports_prompt_caching,
    is_bedrock_route,
    is_claude_model,
    is_known_openai_bare_model,
+    is_openrouter_model,
    model_supports_reasoning,
    request_timeout_extra_args,
 )
@@ -138,6 +140,15 @@ def build_root_task(scan_config: dict[str, Any]) -> str:
            "target to assess: the instructions below are the only source of "
            "truth for what to do."
        )
+    elif not parts and user_instructions:
+        # Neither a target nor a directory, but there is an instruction: the user
+        # declined the mount, so the instruction is all there is. Say so, or the
+        # agent goes looking for a scope that was never given.
+        parts.append(
+            "\n\nNo scan target and no working directory were provided. The "
+            "instructions below are the only source of truth for what to do; "
+            "work from them and from what you can reach yourself."
+        )

    parts.extend(_render_diff_scope(diff_scope))

@@ -192,13 +203,15 @@ def make_model_settings(
    request_timeout: float | None = None,
    prompt_cache: bool = True,
    extra_headers: dict[str, str] | None = None,
+    has_tools: bool = True,
 ) -> ModelSettings:
+    headers = _request_headers(model_name, extra_headers)
    model_settings = ModelSettings(
-        parallel_tool_calls=False,
+        parallel_tool_calls=False if has_tools else None,
        retry=DEFAULT_MODEL_RETRY,
        include_usage=True,
        extra_args=request_timeout_extra_args(request_timeout),
-        extra_headers=dict(extra_headers) if extra_headers else None,
+        extra_headers=headers,
    )
    if (
        reasoning_effort is not None
@@ -221,6 +234,17 @@ def make_model_settings(
    return model_settings


+def _request_headers(
+    model_name: str, extra_headers: dict[str, str] | None
+) -> dict[str, str] | None:
+    headers: dict[str, str] = {}
+    if is_openrouter_model(model_name):
+        headers.update(OPENROUTER_ATTRIBUTION_HEADERS)
+    if extra_headers:
+        headers.update(extra_headers)
+    return headers or None
+
+
 def _reasoning_settings(
    effort: ReasoningEffort,
    extra_args: dict[str, Any] | None,
@@ -2,6 +2,7 @@

 from __future__ import annotations

+import asyncio
 import contextlib
 import io
 import json
@@ -429,7 +430,6 @@ async def run_strix_scan(
    except BudgetExceededError as exc:
        logger.info("Scan %s stopped: %s", scan_id, exc)
        if root_id is not None:
-            await coordinator.cancel_descendants(root_id)
            with contextlib.suppress(Exception):
                await coordinator.set_status(root_id, "stopped")
        return None
@@ -442,19 +442,28 @@ async def run_strix_scan(
            scan_id,
        )
        if root_id is not None:
-            await coordinator.cancel_descendants(root_id)
            with contextlib.suppress(Exception):
                await coordinator.set_status(root_id, "stopped")
        return None
+    except (asyncio.CancelledError, KeyboardInterrupt):
+        logger.info("Scan %s interrupted by the user", scan_id)
+        if root_id is not None:
+            with contextlib.suppress(Exception):
+                await coordinator.set_status(root_id, "running")
+        raise
    except BaseException:
        logger.exception("Strix scan %s failed", scan_id)
        if root_id is not None:
-            await coordinator.cancel_descendants(root_id)
            with contextlib.suppress(Exception):
                await coordinator.set_status(root_id, "failed")
        raise
    finally:
        configure_spill_writer(None)
+        # Settle descendants before closing sessions: on a clean finish a child
+        # can still be mid-turn, and closing its session underneath it crashes it.
+        if root_id is not None:
+            with contextlib.suppress(Exception):
+                await coordinator.cancel_descendants(root_id)
        for s in sessions_to_close:
            with contextlib.suppress(Exception):
                s.close()
@@ -4,6 +4,8 @@ from __future__ import annotations

 import asyncio
 import logging
+import sqlite3
+from contextlib import contextmanager
 from typing import TYPE_CHECKING, Any, cast
 from weakref import WeakKeyDictionary

@@ -12,7 +14,7 @@ from agents.memory import SQLiteSession


 if TYPE_CHECKING:
-    from collections.abc import Callable
+    from collections.abc import Callable, Iterator
    from pathlib import Path

    from agents.items import TResponseInputItem
@@ -22,9 +24,25 @@ if TYPE_CHECKING:
 logger = logging.getLogger(__name__)


+class _PooledConnectionSession(SQLiteSession):
+    @contextmanager
+    def _locked_connection(self) -> Iterator[sqlite3.Connection]:
+        with self._lock:
+            if self._closed:
+                raise RuntimeError("SQLiteSession is closed")
+            if self._is_memory_db:
+                yield self._shared_connection
+                return
+            connection = sqlite3.connect(str(self.db_path), check_same_thread=False)
+            try:
+                yield connection
+            finally:
+                connection.close()
+
+
 def open_agent_session(agent_id: str, path: Path) -> SQLiteSession:
    path.parent.mkdir(parents=True, exist_ok=True)
-    return SQLiteSession(session_id=agent_id, db_path=path)
+    return _PooledConnectionSession(session_id=agent_id, db_path=path)


 async def seed_initial_input(session: Session, initial_input: Any) -> bool:
@@ -1,8 +1,8 @@
-"""`strix auth` — model-subscription sign-in (login / status / logout).
+"""`strix auth` — ChatGPT subscription sign-in (login / status / logout).

 Signing in only stores OAuth tokens (``~/.strix/subscription-auth.json``); model
-selection stays with ``STRIX_LLM``. A ``chatgpt/<model>`` STRIX_LLM runs on a
-ChatGPT subscription and a ``grok/<model>`` one on a Grok/SuperGrok subscription.
+selection stays with ``STRIX_LLM``. A ``chatgpt/<model>`` STRIX_LLM runs on the
+subscription.
 """

 from __future__ import annotations
@@ -12,7 +12,6 @@ import base64
 import logging
 import threading
 import webbrowser
-from dataclasses import dataclass
 from http.server import BaseHTTPRequestHandler, HTTPServer
 from pathlib import Path
 from typing import TYPE_CHECKING, Any
@@ -22,76 +21,24 @@ from rich.console import Console
 from rich.panel import Panel
 from rich.text import Text

-from strix.config import codex, grok, load_settings, subscription_store
+from strix.config import codex, load_settings


 if TYPE_CHECKING:
    from collections.abc import Callable
-    from types import ModuleType


 logger = logging.getLogger(__name__)

 _CALLBACK_TIMEOUT_S = 300

+# CLI-facing name for the login provider. Internally this is the Codex OAuth
+# flow (``codex.PROVIDER``), but users know it as ChatGPT, so that's what the
+# command and messaging say. ``codex`` is accepted as an alias.
+LOGIN_PROVIDER = "chatgpt"
+_ACCEPTED_PROVIDERS = frozenset({LOGIN_PROVIDER, codex.PROVIDER})

-@dataclass(frozen=True)
-class _Provider:
-    """A model-subscription provider the ``strix auth`` command can sign into.
-
-    ``module`` is the provider's OAuth module (:mod:`strix.config.codex` or
-    :mod:`strix.config.grok`); both expose the same login surface. ``error`` is
-    that module's auth-error class, caught to report a clean failure.
-    """
-
-    name: str
-    module: ModuleType
-    error: type[Exception]
-    display: str
-    example_model: str
-    blurb: str
-
-
-_PROVIDERS: dict[str, _Provider] = {
-    "chatgpt": _Provider(
-        name="chatgpt",
-        module=codex,
-        error=codex.CodexAuthError,
-        display="ChatGPT",
-        example_model="chatgpt/gpt-5.4",
-        blurb="This uses your ChatGPT Plus/Pro plan for inference instead of a metered API key.",
-    ),
-    "grok": _Provider(
-        name="grok",
-        module=grok,
-        error=grok.GrokAuthError,
-        display="Grok",
-        example_model="grok/grok-4",
-        blurb="This uses your Grok/SuperGrok plan for inference instead of a metered API key.",
-    ),
-}
-
-# Internal OAuth provider ids and common vendor names accepted as aliases.
-_PROVIDER_ALIASES: dict[str, str] = {
-    codex.PROVIDER: "chatgpt",
-    grok.PROVIDER: "grok",
-    "xai": "grok",
-    "supergrok": "grok",
-}
-
-_DEFAULT_PROVIDER = "chatgpt"
-
-_USAGE = (
-    "Usage:\n"
-    "  strix auth login [chatgpt|grok] [--manual]\n"
-    "  strix auth status\n"
-    "  strix auth logout [chatgpt|grok]"
-)
-
-
-def _resolve_provider(name: str) -> _Provider | None:
-    key = _PROVIDER_ALIASES.get(name.lower(), name.lower())
-    return _PROVIDERS.get(key)
+_USAGE = "Usage:\n  strix auth login chatgpt [--manual]\n  strix auth status\n  strix auth logout"


 def run_auth(argv: list[str]) -> int:
@@ -102,20 +49,20 @@ def run_auth(argv: list[str]) -> int:
    rest = argv[1:]

    if subcommand in ("-h", "--help", "help"):
-        console.print(_USAGE, markup=False)
+        console.print(_USAGE)
        return 0

    handlers: dict[str, Callable[[], int]] = {
        "login": lambda: _login(console, rest),
        "status": lambda: _status(console),
-        "logout": lambda: _logout(console, rest),
+        "logout": lambda: _logout(console),
    }
    handler = handlers.get(subcommand)
    if handler is not None:
        return handler()

    console.print(f"[red]Unknown auth command:[/] {subcommand}\n")
-    console.print(_USAGE, markup=False)
+    console.print(_USAGE)
    return 2


@@ -124,8 +71,8 @@ def _login(console: Console, argv: list[str]) -> int:
    parser.add_argument(
        "provider",
        nargs="?",
-        default=_DEFAULT_PROVIDER,
-        help="Model provider to sign in with (chatgpt or grok; default: chatgpt).",
+        default=LOGIN_PROVIDER,
+        help="Model provider to sign in with (default: chatgpt).",
    )
    parser.add_argument(
        "--manual",
@@ -137,42 +84,39 @@ def _login(console: Console, argv: list[str]) -> int:
    except SystemExit as exc:  # argparse already printed the message
        return int(exc.code or 2)

-    provider = _resolve_provider(args.provider)
-    if provider is None:
-        supported = ", ".join(f"'{name}'" for name in _PROVIDERS)
-        console.print(f"[red]Unsupported provider:[/] {args.provider}. Supported: {supported}.")
+    if args.provider.lower() not in _ACCEPTED_PROVIDERS:
+        console.print(
+            f"[red]Unsupported provider:[/] {args.provider}. "
+            f"Only '{LOGIN_PROVIDER}' (ChatGPT subscription) is supported."
+        )
        return 2

-    module = provider.module
-    verifier, challenge = module.generate_pkce()
-    state = module.create_state()
-    authorize_url = module.build_authorize_url(challenge, state)
+    verifier, challenge = codex.generate_pkce()
+    state = codex.create_state()
+    authorize_url = codex.build_authorize_url(challenge, state)

    console.print()
+    console.print("[bold]Signing in with ChatGPT[/] [dim](provider: chatgpt)[/]")
    console.print(
-        f"[bold]Signing in with {provider.display}[/] [dim](provider: {provider.name})[/]"
+        "[dim]This uses your ChatGPT Plus/Pro plan for inference instead of a metered API key.[/]"
    )
-    console.print(f"[dim]{provider.blurb}[/]")
    console.print()

    try:
-        record = _run_oauth_flow(
-            console, provider, authorize_url, verifier, state, manual=args.manual
-        )
-    except provider.error as exc:
+        record = _run_oauth_flow(console, authorize_url, verifier, state, manual=args.manual)
+    except codex.CodexAuthError as exc:
        return _fail(console, exc)
    except KeyboardInterrupt:
        console.print("\n[yellow]Sign-in cancelled.[/]")
        return 130

-    module.save_record(record)
-    _print_success(console, provider)
+    codex.save_record(record)
+    _print_success(console)
    return 0


 def _run_oauth_flow(
    console: Console,
-    provider: _Provider,
    authorize_url: str,
    verifier: str,
    state: str,
@@ -180,10 +124,7 @@ def _run_oauth_flow(
    manual: bool,
 ) -> dict[str, Any]:
    """Drive the browser (or manual) OAuth flow and return a token record."""
-    module = provider.module
-    server = (
-        None if manual else _try_start_callback_server(module.CALLBACK_PORT, module.CALLBACK_PATH)
-    )
+    server = None if manual else _try_start_callback_server()

    console.print("Open this URL in your browser to authorize:")
    console.print(f"[cyan]{authorize_url}[/]")
@@ -201,8 +142,8 @@ def _run_oauth_flow(
        if result is not None:
            code, returned_state, error = result
            if error:
-                raise provider.error("oauth_error", error)
-            return _finish(provider, code, returned_state, verifier, state, require_state=True)
+                raise codex.CodexAuthError("oauth_error", error)
+            return _finish(code, returned_state, verifier, state, require_state=True)
        console.print("[yellow]Timed out waiting for the browser. Falling back to manual paste.[/]")

    # Manual fallback: the user completes sign-in and pastes the redirect URL
@@ -212,13 +153,12 @@ def _run_oauth_flow(
    try:
        pasted = console.input("Paste the full redirect URL (or code#state): ").strip()
    except EOFError as exc:
-        raise provider.error("no_input", "no redirect URL provided") from exc
-    code, returned_state = module.parse_redirect_input(pasted)
-    return _finish(provider, code, returned_state, verifier, state, require_state=False)
+        raise codex.CodexAuthError("no_input", "no redirect URL provided") from exc
+    code, returned_state = codex.parse_redirect_input(pasted)
+    return _finish(code, returned_state, verifier, state, require_state=False)


 def _finish(
-    provider: _Provider,
    code: str | None,
    returned_state: str | None,
    verifier: str,
@@ -227,17 +167,16 @@ def _finish(
    require_state: bool,
 ) -> dict[str, Any]:
    if not code:
-        raise provider.error("no_code", "no authorization code found in the redirect")
-    # The loopback callback from the provider always carries state, so a missing
-    # or mismatched value there is forged (CSRF) and must be rejected. Manual
-    # paste is user-initiated (the user copies their own redirect), so state is
-    # only validated when the pasted value includes it.
+        raise codex.CodexAuthError("no_code", "no authorization code found in the redirect")
+    # The loopback callback from OpenAI always carries state, so a missing or
+    # mismatched value there is forged (CSRF) and must be rejected. Manual paste
+    # is user-initiated (the user copies their own redirect), so state is only
+    # validated when the pasted value includes it.
    if require_state and returned_state is None:
-        raise provider.error("state_mismatch", "missing state in callback; possible CSRF")
+        raise codex.CodexAuthError("state_mismatch", "missing state in callback; possible CSRF")
    if returned_state is not None and returned_state != expected_state:
-        raise provider.error("state_mismatch", "state did not match; possible CSRF")
-    record: dict[str, Any] = provider.module.exchange_code(code, verifier)
-    return record
+        raise codex.CodexAuthError("state_mismatch", "state did not match; possible CSRF")
+    return codex.exchange_code(code, verifier)


 class _CallbackServer:
@@ -264,7 +203,7 @@ class _CallbackServer:
        self._httpd.server_close()


-def _try_start_callback_server(port: int, path: str) -> _CallbackServer | None:
+def _try_start_callback_server() -> _CallbackServer | None:
    event = threading.Event()
    holder: dict[str, Any] = {}

@@ -274,7 +213,7 @@ def _try_start_callback_server(port: int, path: str) -> _CallbackServer | None:

        def do_GET(self) -> None:
            parsed = urlparse(self.path)
-            if parsed.path != path:
+            if parsed.path != codex.CALLBACK_PATH:
                self.send_response(404)
                self.end_headers()
                return
@@ -291,9 +230,9 @@ def _try_start_callback_server(port: int, path: str) -> _CallbackServer | None:
            event.set()

    try:
-        httpd = HTTPServer(("127.0.0.1", port), Handler)
+        httpd = HTTPServer(("127.0.0.1", codex.CALLBACK_PORT), Handler)
    except OSError:
-        logger.debug("could not bind callback port %d", port, exc_info=True)
+        logger.debug("could not bind callback port %d", codex.CALLBACK_PORT, exc_info=True)
        return None
    return _CallbackServer(httpd, event, holder)

@@ -304,67 +243,30 @@ def _first(query: dict[str, list[str]], key: str) -> str | None:


 def _status(console: Console) -> int:
-    settings = load_settings()
-    active_model = settings.llm.model
-    signed_in_any = False
-    for provider in _PROVIDERS.values():
-        record = provider.module.read_record()
-        if record is None:
-            continue
-        signed_in_any = True
-        console.print(f"[green]Signed in[/] with a {provider.display} subscription.")
-        account_id = record.get("account_id")
-        if account_id:
-            console.print(f"  Account: [bold]{account_id}[/]")
-        if provider.module.subscription_model(active_model):
-            console.print(f"  Runs use the subscription (STRIX_LLM=[bold]{active_model}[/]).")
-        else:
-            console.print(
-                f"  [yellow]Note:[/] set [cyan]STRIX_LLM[/] to e.g. "
-                f"[cyan]{provider.example_model}[/] to run on this subscription."
-            )
-    if not signed_in_any:
-        console.print(
-            "[yellow]Not signed in.[/] Run [cyan]strix auth login chatgpt[/] "
-            "or [cyan]strix auth login grok[/] to sign in."
-        )
+    record = codex.read_record()
+    if record is None:
+        console.print("[yellow]Not signed in.[/] Run [cyan]strix auth login chatgpt[/] to sign in.")
        return 1
+    settings = load_settings()
+    console.print("[green]Signed in[/] with a ChatGPT subscription.")
+    console.print(f"  Account: [bold]{record.get('account_id')}[/]")
+    if codex.subscription_model(settings.llm.model):
+        console.print(f"  Runs use the subscription (STRIX_LLM=[bold]{settings.llm.model}[/]).")
+    else:
+        console.print(
+            "  [yellow]Note:[/] set [cyan]STRIX_LLM[/] to e.g. [cyan]chatgpt/gpt-5.4[/] "
+            "to run on the subscription."
+        )
    return 0


-def _logout(console: Console, argv: list[str]) -> int:
-    parser = argparse.ArgumentParser(prog="strix auth logout", add_help=True)
-    parser.add_argument(
-        "provider",
-        nargs="?",
-        default=None,
-        help="Provider to sign out of (chatgpt or grok; default: all).",
-    )
-    try:
-        args = parser.parse_args(argv)
-    except SystemExit as exc:
-        return int(exc.code or 2)
-
-    if args.provider is None:
-        # Hold the store lock across every provider so a concurrent save/refresh
-        # can't slip a credential back in between removals (logout-all is atomic).
-        with subscription_store.guard(codex.AUTH_PATH):
-            for provider in _PROVIDERS.values():
-                provider.module.logout()
-        console.print("[green]Signed out.[/] Stored subscription credentials removed.")
-        return 0
-
-    target = _resolve_provider(args.provider)
-    if target is None:
-        supported = ", ".join(f"'{name}'" for name in _PROVIDERS)
-        console.print(f"[red]Unsupported provider:[/] {args.provider}. Supported: {supported}.")
-        return 2
-    target.module.logout()
-    console.print(f"[green]Signed out of {target.display}.[/] Stored credentials removed.")
+def _logout(console: Console) -> int:
+    codex.logout()
+    console.print("[green]Signed out.[/] Stored subscription credentials removed.")
    return 0


-def _fail(console: Console, exc: Exception) -> int:
+def _fail(console: Console, exc: codex.CodexAuthError) -> int:
    error_text = Text()
    error_text.append("SIGN-IN FAILED", style="bold red")
    error_text.append("\n\n", style="white")
@@ -382,18 +284,17 @@ def _fail(console: Console, exc: Exception) -> int:
    return 1


-def _print_success(console: Console, provider: _Provider) -> None:
-    prefix = provider.module.SUBSCRIPTION_PREFIX
+def _print_success(console: Console) -> None:
    text = Text()
-    text.append(f"Signed in with your {provider.display} subscription", style="bold #22c55e")
+    text.append("Signed in with your ChatGPT subscription", style="bold #22c55e")
    text.append("\n\n", style="white")
    text.append("Set ", style="white")
    text.append("STRIX_LLM", style="bold white")
    text.append(" to a ", style="white")
-    text.append(prefix, style="bold cyan")
+    text.append("chatgpt/", style="bold cyan")
    text.append(" model (e.g. ", style="white")
-    text.append(provider.example_model, style="bold cyan")
-    text.append(f") — runs are billed to your {provider.display} plan.", style="white")
+    text.append("chatgpt/gpt-5.4", style="bold cyan")
+    text.append(") — runs are billed to your ChatGPT plan.", style="white")
    text.append("\n\n", style="white")
    text.append("Run a scan as usual, e.g. ", style="white")
    text.append("strix --target https://example.com", style="bold cyan")
@@ -328,10 +328,11 @@ def _load_resume_state(args: argparse.Namespace, parser: argparse.ArgumentParser
        parser.error(f"--resume {args.resume}: run.json unreadable: {exc}")

    args.targets_info = state.get("targets_info") or []
-    # A target-less run has no targets_info at all: it works in a mounted
-    # directory, driven by its instruction.
+    # A target-less run has no targets_info at all. It is driven by its
+    # instruction, over a mounted working directory or over nothing when the
+    # mount was declined, so either of those is enough to resume it.
    workspace_mount = state.get("workspace_mount") or None
-    if not args.targets_info and not workspace_mount:
+    if not args.targets_info and not workspace_mount and not state.get("user_instruction"):
        parser.error(f"--resume {args.resume}: run.json has no targets_info")

    for target in args.targets_info:
@@ -8,7 +8,7 @@ from rich.console import Console
 from rich.panel import Panel
 from rich.text import Text

-from strix.config import codex, grok, load_settings
+from strix.config import codex, load_settings
 from strix.interface.utils import (
    check_docker_connection,
    image_exists,
@@ -37,16 +37,6 @@ def validate_environment() -> None:
        logger.info("Environment OK (ChatGPT subscription)")
        return

-    if grok.subscription_model(settings.llm.model):
-        if not grok.is_authenticated():
-            console.print(
-                f"[red]STRIX_LLM={settings.llm.model} uses your Grok subscription, "
-                "but you're not signed in.[/] Run [cyan]strix auth login grok[/] first."
-            )
-            sys.exit(1)
-        logger.info("Environment OK (Grok subscription)")
-        return
-
    if not settings.llm.model:
        missing_required_vars.append("STRIX_LLM")

@@ -224,6 +224,7 @@ async def warm_up_llm(show_model_warning: bool = True) -> None:
                request_timeout=llm.timeout,
                prompt_cache=False,
                extra_headers=settings.dedupe.extra_headers,
+                has_tools=False,
            )
            if deduper_extra:
                merged = {**(deduper_settings.extra_args or {}), **deduper_extra}
@@ -435,16 +436,7 @@ def main() -> None:
    start_background_check()
    if not args.non_interactive and prompt_update_if_available(Console()):
        if is_binary_install() and sys.platform != "win32":
-            # The PyInstaller onefile bootloader passes its state to the child
-            # process via environment variables; if they leak into the re-exec,
-            # the new binary reuses the old extracted application instead of
-            # unpacking itself, so the pre-update version runs again.
-            env = {
-                key: value
-                for key, value in os.environ.items()
-                if not key.startswith("_PYI_") and key != "_MEIPASS2"
-            }
-            os.execve(sys.executable, sys.argv, env)  # noqa: S606  # nosec B606
+            os.execv(sys.executable, sys.argv)  # noqa: S606  # nosec B606
        sys.exit(0)

    check_docker_installed()
@@ -14,7 +14,7 @@ import logging
 from datetime import UTC, datetime
 from typing import TYPE_CHECKING, Any

-from strix.config import Settings, load_settings, subscription
+from strix.config import Settings, codex, load_settings
 from strix.core.paths import run_dir_for
 from strix.interface.utils import (
    assign_workspace_subdirs,
@@ -78,6 +78,7 @@ async def preflight_model_connection(
        request_timeout=resolved_settings.llm.timeout,
        prompt_cache=False,
        extra_headers=resolved_settings.llm.extra_headers,
+        has_tools=False,
    )
    await asyncio.wait_for(
        model.get_response(
@@ -225,7 +226,7 @@ def telemetry_start(args: argparse.Namespace) -> None:
    model = load_settings().llm.model
    kwargs = {
        "model": model,
-        "auth_mode": subscription.auth_mode(model),
+        "auth_mode": codex.auth_mode(model),
        "scan_mode": args.scan_mode,
        "is_whitebox": is_whitebox_scan(args.targets_info),
        "interactive": not args.non_interactive,
@@ -240,15 +241,13 @@ def _persist_run_record(args: argparse.Namespace) -> None:

    run_dir = run_dir_for(args.run_name)
    run_dir.mkdir(parents=True, exist_ok=True)
-    model = load_settings().llm.model
    run_record = {
        "run_id": args.run_name,
        "run_name": args.run_name,
        "status": "running",
        "start_time": datetime.now(UTC).isoformat(),
        "end_time": None,
-        "auth_mode": subscription.auth_mode(model),
-        "subscription_provider": subscription.provider_label(model),
+        "auth_mode": codex.auth_mode(load_settings().llm.model),
        "targets_info": args.targets_info,
        "scan_mode": args.scan_mode,
        "instruction": args.instruction,
@@ -138,13 +138,6 @@ class TuiController:
        self.error = detail
        self.notify_changed()

-    def enter_setup(self) -> None:
-        """Return a session to the start screen, e.g. on a declined mount."""
-        self.setup_mode = True
-        self.scan_started = False
-        self.scan_state = "setup"
-        self.notify_changed()
-
    def add_message(self, text: str, level: str = "info") -> None:
        self._append_message(text, level)
        self.notify_changed()
@@ -356,14 +349,12 @@ class TuiController:
        if not isinstance(approved, bool):
            raise TypeError("approved must be a boolean")
        self.pending_workspace_mount = None
-        if not approved:
-            # Nothing was prepared, so return to the start screen untouched.
-            self.workspace_mount = None
-            self.enter_setup()
-            return {"approved": False}
-        self.workspace_mount = mount
+        # Declining skips the mount, it does not abandon the scan. The prompt is
+        # the whole of the input either way; the working directory is only an
+        # extra the agent may look at, so the run goes ahead without one.
+        self.workspace_mount = mount if approved else None
        await self._begin_scan(self._pending_verify)
-        return {"approved": True}
+        return {"approved": approved}

    async def _send_message(self, payload: dict[str, Any]) -> dict[str, Any]:
        agent_id = self._required_string(payload, "agent_id")
@@ -66,13 +66,10 @@ func (m *Model) submitSetupPrompt(value string) (tea.Model, tea.Cmd) {
 }

 // answerMountConfirmation replies to the working-directory mount the backend is
-// waiting on. Declining returns to the start screen, so the prompt goes back in
-// the composer to be edited or given a target instead.
+// waiting on. Either answer starts the scan - declining only means it runs
+// without the directory - so the prompt stays with the run rather than coming
+// back to the composer.
 func (m *Model) answerMountConfirmation(approved bool) tea.Cmd {
-	if !approved && m.pendingPrompt != "" {
-		m.input.SetValue(m.pendingPrompt)
-		m.resizeViewport()
-	}
 	m.pendingPrompt = ""
 	return send(m.client, "setup.confirm_mount", map[string]any{"approved": approved})
 }
@@ -247,13 +247,10 @@ func TestMountConfirmationAnswers(t *testing.T) {
 		if payload.Approved != tc.approved {
 			t.Fatalf("%s: approved=%v, want %v", tc.name, payload.Approved, tc.approved)
 		}
-		// Declining returns to the start screen, so the prompt comes back.
-		want := ""
-		if !tc.approved {
-			want = "find auth bugs in the login flow"
-		}
-		if got := model.input.Value(); got != want {
-			t.Fatalf("%s: composer = %q, want %q", tc.name, got, want)
+		// Either answer launches, so the prompt stays with the run rather than
+		// coming back to the composer.
+		if got := model.input.Value(); got != "" {
+			t.Fatalf("%s: composer = %q, want it cleared", tc.name, got)
 		}
 		if model.pendingPrompt != "" {
 			t.Fatalf("%s: held prompt was not cleared: %q", tc.name, model.pendingPrompt)
@@ -290,3 +287,101 @@ func TestSetupPromptWithTargetLaunches(t *testing.T) {
 		t.Fatalf("setup.start (%d) must come after setup.set_instruction (%d): %v", start, instr, types)
 	}
 }
+
+// The prompt's buttons are buttons: clicking Cancel has to answer the backend,
+// which it could not do while the mouse handler had no case for this modal.
+func TestMountPromptButtonsAreClickable(t *testing.T) {
+	for _, testCase := range []struct {
+		label    string
+		approved bool
+	}{
+		{mountConfirmLabel, true},
+		{mountCancelLabel, false},
+	} {
+		connection := &recordingConn{}
+		model := New(&Client{conn: connection})
+		model.width, model.height = 130, 40
+		model.snapshot = protocol.Snapshot{SetupMode: true, WorkingDir: "/Users/me/code/api"}
+		updated, _ := model.submit("find auth bugs in the login flow")
+		model = updated.(Model)
+		connection.Reset()
+		model.snapshot = protocol.Snapshot{
+			ScanStarted: true, ScanState: "preparing", PendingMount: "/Users/me/code/api",
+		}
+		model.syncMountPrompt()
+
+		left, top, panel := model.mountPromptBounds()
+		clicked := false
+		for row, line := range strings.Split(panel, "\n") {
+			plain := ansi.Strip(line)
+			index := strings.Index(plain, testCase.label)
+			if index < 0 {
+				continue
+			}
+			updated, cmd := model.updateModalMouse(tea.MouseMsg{
+				X: left + ansi.StringWidth(plain[:index]) + 1, Y: top + row,
+				Button: tea.MouseButtonLeft, Action: tea.MouseActionPress,
+			})
+			model = updated.(Model)
+			envelopes := drainCommands(t, cmd, connection)
+			if len(envelopes) != 1 || envelopes[0].Type != "setup.confirm_mount" {
+				t.Fatalf("clicking %s sent %v", testCase.label, commandTypes(envelopes))
+			}
+			var payload struct {
+				Approved bool `json:"approved"`
+			}
+			if err := json.Unmarshal(envelopes[0].Payload, &payload); err != nil {
+				t.Fatal(err)
+			}
+			if payload.Approved != testCase.approved {
+				t.Fatalf("clicking %s answered approved=%v", testCase.label, payload.Approved)
+			}
+			clicked = true
+			break
+		}
+		if !clicked {
+			t.Fatalf("%s was not found in the prompt", testCase.label)
+		}
+	}
+}
+
+// Skipping the mount runs the scan without a directory. It must not throw the
+// session back to the start screen, and it must not hand the prompt back: the
+// run has it.
+func TestSkippingTheMountKeepsTheScanRunning(t *testing.T) {
+	connection := &recordingConn{}
+	model := New(&Client{conn: connection})
+	model.width, model.height = 130, 40
+	model.snapshot = protocol.Snapshot{SetupMode: true, WorkingDir: "/Users/me/code/api"}
+	updated, _ := model.submit("find auth bugs in the login flow")
+	model = updated.(Model)
+	model.snapshot = protocol.Snapshot{
+		ScanStarted: true, ScanState: "preparing", PendingMount: "/Users/me/code/api",
+	}
+	model.syncMountPrompt()
+	if model.modal != modalConfirmMount {
+		t.Fatal("the prompt did not open")
+	}
+
+	model.modalChoice = 1
+	updated, _ = model.updateModal(tea.KeyMsg{Type: tea.KeyEnter})
+	model = updated.(Model)
+
+	// The backend answers by starting the scan with no mount.
+	model.handleEnvelope(stateEnvelope(t, 2, protocol.Snapshot{
+		ScanStarted: true, ScanState: "running",
+	}))
+
+	if model.modal != modalNone {
+		t.Fatalf("the prompt is still open: %v", model.modal)
+	}
+	if model.snapshot.SetupMode {
+		t.Fatal("skipping the mount fell back to the start screen")
+	}
+	if got := model.input.Value(); got != "" {
+		t.Fatalf("the prompt came back to the composer: %q", got)
+	}
+	if model.pendingPrompt != "" {
+		t.Fatalf("the held prompt was not released: %q", model.pendingPrompt)
+	}
+}
@@ -474,6 +474,18 @@ func (m Model) updateModalMouse(msg tea.MouseMsg) (tea.Model, tea.Cmd) {
 			m.modalChoice = 1
 			return m.updateModal(tea.KeyMsg{Type: tea.KeyEnter})
 		}
+	case modalConfirmMount:
+		left, top, panel := m.mountPromptBounds()
+		if labelHitAt(panel, mountConfirmLabel, left, top, msg.X, msg.Y) {
+			m.modalChoice = 0
+			cmd := m.answerMountConfirmation(true)
+			return m, cmd
+		}
+		if labelHitAt(panel, mountCancelLabel, left, top, msg.X, msg.Y) {
+			m.modalChoice = 1
+			cmd := m.answerMountConfirmation(false)
+			return m, cmd
+		}
 	case modalVulnerability:
 		for _, button := range m.reportButtons() {
 			if button == reportCopy || button == reportDone {
@@ -507,7 +519,14 @@ func (m Model) centeredViewBounds(view string) (left, top, width, height int) {

 func (m Model) centeredLabelHit(view, label string, x, y int) bool {
 	left, top, _, _ := m.centeredViewBounds(view)
-	for row, line := range strings.Split(view, "\n") {
+	return labelHitAt(view, label, left, top, x, y)
+}
+
+// labelHitAt reports whether a click landed on a label drawn in a panel whose
+// top-left corner is at (left, top). The mount prompt is docked in a corner
+// rather than centered, so it cannot use the centered bounds.
+func labelHitAt(panel, label string, left, top, x, y int) bool {
+	for row, line := range strings.Split(panel, "\n") {
 		plain := ansi.Strip(line)
 		index := strings.Index(plain, label)
 		if index < 0 || y != top+row {
@@ -593,7 +612,8 @@ func (m Model) updateModal(key tea.KeyMsg) (tea.Model, tea.Cmd) {
 	case "esc":
 		if m.modal == modalConfirmMount {
 			// The backend is waiting on an answer; escape declines it.
-			return m, m.answerMountConfirmation(false)
+			cmd := m.answerMountConfirmation(false)
+			return m, cmd
 		}
 		m.closeModal()
 		return m, nil
@@ -604,7 +624,10 @@ func (m Model) updateModal(key tea.KeyMsg) (tea.Model, tea.Cmd) {
 		modal, choice := m.modal, m.modalChoice
 		if modal == modalConfirmMount {
 			// The snapshot closes this prompt once the backend has the answer.
-			return m, m.answerMountConfirmation(choice == 0)
+			// Bound to a variable first: the call restores the held prompt into
+			// the composer, and that has to be in the model being returned.
+			cmd := m.answerMountConfirmation(choice == 0)
+			return m, cmd
 		}
 		m.closeModal()
 		if choice == 1 {
@@ -271,6 +271,23 @@ func (m Model) viewInner() string {
 	return m.toastOverlay(main)
 }

+// mountPromptBounds is where the working-directory prompt is drawn. It is placed
+// by cornerOverlay rather than centered, so a click has to be tested against
+// these bounds and not the ones the other modals use.
+func (m Model) mountPromptBounds() (left, top int, panel string) {
+	panel = m.modalView()
+	if panel == "" {
+		return 0, 0, ""
+	}
+	_, _, chatWidth, _ := m.layout()
+	left = max(0, min(chatWidth, m.width)-lipgloss.Width(panel))
+	statusH := 0
+	if m.statusVisible() {
+		statusH = 1
+	}
+	return left, max(0, m.inputTop()-statusH-lipgloss.Height(panel)), panel
+}
+
 // cornerOverlay splices a panel in directly above the composer, right-aligned
 // with it, leaving the rest of the view visible behind it.
 func (m Model) cornerOverlay(view, panel string) string {
@@ -220,6 +220,13 @@ func (m Model) confirmView(title string, width int, border, titleColor lipgloss.
 	return m.confirmDialog(title, "", width, border, titleColor, red, "Yes", "No")
 }

+// The mount prompt's buttons, named so the renderer and the click test cannot
+// drift apart.
+const (
+	mountConfirmLabel = "Mount"
+	mountCancelLabel  = "Skip"
+)
+
 // mountConfirmView asks before a target-less scan mounts the working directory.
 // It is a compact prompt docked in the corner of the live view: nothing is
 // prepared until it is answered, and the directory is a workspace rather than a
@@ -232,8 +239,8 @@ func (m Model) mountConfirmView() string {
 	}
 	title := render.Bold(amber).Render("△ Mount working directory?")
 	body := render.Col(white).Render(truncatePath(dir, width-4)) + "\n" +
-		render.Dim().Render("writable in the sandbox")
-	return m.cornerPrompt(title, body, width, "Confirm", "Cancel")
+		render.Dim().Render("writable in the sandbox · skip to run without it")
+	return m.cornerPrompt(title, body, width, mountConfirmLabel, mountCancelLabel)
 }

 // truncatePath keeps the tail of a path visible, which is the part that
@@ -262,27 +262,9 @@ def is_subscription_run(report_state: Any) -> bool:
    record = getattr(report_state, "run_record", None)
    if isinstance(record, dict) and record.get("auth_mode"):
        return record.get("auth_mode") == "subscription"
-    from strix.config import subscription
+    from strix.config import codex

-    return subscription.auth_mode(load_settings().llm.model) == "subscription"
-
-
-def _subscription_label(report_state: Any) -> str:
-    """Human label for the active model subscription (e.g. "Grok subscription").
-
-    Prefers the persisted run record so a resumed run keeps its original provider
-    even if STRIX_LLM later points at a different one; falls back to current
-    settings.
-    """
-    record = getattr(report_state, "run_record", None)
-    if isinstance(record, dict):
-        provider = record.get("subscription_provider")
-        if isinstance(provider, str) and provider:
-            return f"{provider} subscription"
-    from strix.config import subscription
-
-    label = subscription.provider_label(load_settings().llm.model)
-    return f"{label} subscription" if label else "Subscription"
+    return codex.auth_mode(load_settings().llm.model) == "subscription"


 def _int_stat(usage: dict[str, Any], key: str) -> int:
@@ -386,7 +368,7 @@ def build_live_stats_text(report_state: Any) -> Text:
    stats_text.append(str(model), style="white")
    if is_subscription_run(report_state):
        stats_text.append("  ·  ", style="dim white")
-        stats_text.append(_subscription_label(report_state), style="#22c55e")
+        stats_text.append("ChatGPT subscription", style="#22c55e")
    stats_text.append("\n")

    vuln_count = len(report_state.vulnerability_reports)
@@ -432,7 +414,7 @@ def build_tui_stats_text(report_state: Any) -> Text:
    subscription = is_subscription_run(report_state)
    if subscription:
        stats_text.append("\n")
-        stats_text.append(_subscription_label(report_state), style="#22c55e")
+        stats_text.append("ChatGPT subscription", style="#22c55e")

    usage = _llm_usage(report_state)
    if usage and _int_stat(usage, "total_tokens") > 0:
@@ -101,7 +101,6 @@ export function RunDetails({
  const totalTokens = num(usage.total_tokens);
  const cost = num(usage.cost);
  const subscription = str(raw.auth_mode) === "subscription";
-  const subscriptionProvider = str(raw.subscription_provider);

  const sub = (n: number, word: string) => (
    <span className="text-[#666]"> ({formatNumber(n)} {word})</span>
@@ -181,7 +180,7 @@ export function RunDetails({
                <Field label="Provider">
                  <span className="inline-flex items-center gap-1.5">
                    <span className="rounded-full border border-[#22c55e]/40 bg-[#22c55e]/10 px-2 py-0.5 text-[11px] text-[#22c55e]">
-                      {subscriptionProvider ? `${subscriptionProvider} subscription` : "Subscription"}
+                      ChatGPT subscription
                    </span>
                  </span>
                </Field>
@@ -6,7 +6,7 @@
    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
    <meta name="color-scheme" content="dark" />
    <title>Strix Results</title>
-    <script type="module" crossorigin src="./assets/index-XDX3roAH.js"></script>
+    <script type="module" crossorigin src="./assets/index-DBJ-RJqo.js"></script>
    <link rel="stylesheet" crossorigin href="./assets/index-DKbLYAbP.css">
  </head>
  <body>
@@ -6,7 +6,6 @@ import json
 import logging
 from typing import TYPE_CHECKING, Any

-from strix.config import subscription
 from strix.core.paths import run_record_path
 from strix.interface.tui.live_view import TuiLiveView

@@ -58,39 +57,7 @@ def read_run_summary(run_dir: Path) -> dict[str, Any]:
        record = {}
    status = record.get("status")
    finished = status in _TERMINAL_STATUSES and bool(record.get("end_time"))
-    summary = {**record, "finished": finished}
-    _backfill_subscription_provider(summary)
-    return summary
-
-
-def _first_recorded_model(record: dict[str, Any]) -> str | None:
-    """The first non-empty per-agent model slug in a run record, or None."""
-    usage = record.get("llm_usage")
-    if not isinstance(usage, dict):
-        return None
-    agents = usage.get("agents")
-    if not isinstance(agents, list):
-        return None
-    for agent in agents:
-        if isinstance(agent, dict):
-            model = agent.get("model")
-            if isinstance(model, str) and model:
-                return model
-    return None
-
-
-def _backfill_subscription_provider(record: dict[str, Any]) -> None:
-    """Name the subscription provider for runs recorded before that field
-    existed, deriving it from the recorded ``provider/model`` slug so the viewer
-    labels them correctly without a rescan. Newer runs already carry the field.
-    """
-    if record.get("subscription_provider"):
-        return
-    if record.get("auth_mode") != "subscription":
-        return
-    label = subscription.provider_label(_first_recorded_model(record))
-    if label:
-        record["subscription_provider"] = label
+    return {**record, "finished": finished}


 def primary_target(record: dict[str, Any]) -> str | None:
@@ -294,6 +294,7 @@ async def _summarize(model: str, prompt: str, max_tokens: int) -> str | None:
        request_timeout=llm.timeout,
        prompt_cache=False,
        extra_headers=llm.extra_headers,
+        has_tools=False,
    ).resolve(ModelSettings(max_tokens=max_tokens))
    try:
        response = (
@@ -62,6 +62,7 @@ def _dedupe_model_settings(
        # must never receive the main endpoint's credentials. A dedicated model
        # gets its own DEDUPE_LLM_EXTRA_HEADERS instead.
        extra_headers=dedupe.extra_headers if dedupe.model else llm.extra_headers,
+        has_tools=False,
    )
    extra = _dedupe_extra_args(dedupe)
    if extra:
@@ -11,7 +11,7 @@ from uuid import uuid4

 from agents.usage import Usage

-from strix.config import subscription
+from strix.config import codex
 from strix.config.loader import load_settings
 from strix.core.paths import run_dir_for
 from strix.report.sarif import write_sarif
@@ -122,8 +122,7 @@ class ReportState:
        self.scan_results: dict[str, Any] | None = None
        self.scan_config: dict[str, Any] | None = None
        self._llm_usage = LLMUsageLedger()
-        model = load_settings().llm.model
-        auth_mode = subscription.auth_mode(model)
+        auth_mode = codex.auth_mode(load_settings().llm.model)
        self._llm_usage.zero_cost = auth_mode == "subscription"
        self.run_record: dict[str, Any] = {
            "run_id": self.run_id,
@@ -132,7 +131,6 @@ class ReportState:
            "end_time": None,
            "status": "running",
            "auth_mode": auth_mode,
-            "subscription_provider": subscription.provider_label(model),
            "targets_info": [],
            "llm_usage": self._build_llm_usage_record(),
        }
@@ -58,6 +58,26 @@ agent-browser screenshot
 The browser stays running across commands so these feel like a single
 session. Use `agent-browser close` (or `close --all`) when you're done.

+The default session is **shared with every other agent in the sandbox** — if
+another agent navigates it, your page and your refs are gone from under you. So
+claim your own by passing `--session <your-agent-name>` on **every** command:
+
+```bash
+agent-browser --session recon-3 open https://example.com
+agent-browser --session recon-3 snapshot -i
+agent-browser --session recon-3 close        # when done with the target
+```
+
+The examples in the rest of this skill omit `--session` to keep them readable;
+keep passing yours. Each session is a separate Chromium (~340 MB) on a shared
+box, so hold one rather than several, and close it when you're finished.
+
+A browser left idle for 3 minutes is reclaimed automatically to free memory for
+the other agents; the next command relaunches it, but the page, tabs, refs and
+cookies are gone. If you're authenticated and about to go do something else for a
+while, save the state first (see
+[Persist session across runs](#persist-session-across-runs)).
+
 ## Reading a page

 ```bash
@@ -307,6 +327,16 @@ agent-browser --session b fill @e1 "bob@test.com"
 `AGENT_BROWSER_SESSION=myapp` sets the default session for the current
 shell.

+Use a session named after yourself for your own work — that's what keeps a
+concurrent agent from navigating the page out from under you. Every session is a
+separate Chromium though, so hold one at a time rather than a collection, and
+close each one when its flow is finished:
+
+```bash
+agent-browser --session a close
+agent-browser --session b close
+```
+
 ### Mock network requests

 ```bash
@@ -368,8 +398,11 @@ agent-browser dialog dismiss          # cancel
 ## Readiness & recovery

 The first `agent-browser open` in a session launches the headless-Chrome
-daemon; later commands reuse it. Distinguish the two failure modes and react
-differently — do **not** blindly re-run the same failing command in a loop:
+daemon; later commands reuse it. A daemon left idle for 3 minutes shuts itself
+down to free memory for the other agents, so an `open` after a long gap is a
+fresh browser rather than a resumed one — expect to re-navigate, and re-`state
+load` if you were logged in. Distinguish the failure modes and react differently
+— do **not** blindly re-run the same failing command in a loop:

 - **Daemon / connection failure** (`Failed to connect`, `connection refused`,
  socket missing, `browser not running`): the daemon isn't up or has died. Run
@@ -15,7 +15,7 @@ def _ctx(ctx: RunContextWrapper) -> dict[str, Any]:


@function_tool
-async def respond_to_user(ctx: RunContextWrapper, message: str) -> str:
+async def respond_to_user(ctx: RunContextWrapper, message: str = "") -> str:
    """Answer the user and hand control back to them.

    This is the ONLY way to yield to the user. Delivering the message and
@@ -45,6 +45,10 @@ async def respond_to_user(ctx: RunContextWrapper, message: str) -> str:
            have followed the tool calls that led here. Lead with the
            answer or the decision you need, and if you are blocked, say
            exactly what you need from them.
+
+            Omit it when you have just said your piece as plain text and
+            only need to wait: that text has already reached them, and
+            repeating it makes them read the same answer twice.
    """
    inner = _ctx(ctx)
    coordinator = coordinator_from_context(inner)
@@ -110,12 +110,19 @@ def _get_agent_todos(agent_id: str) -> dict[str, dict[str, Any]]:


 def _normalize_priority(priority: str | None, default: str = "normal") -> str:
-    candidate = (priority or default or "normal").lower()
+    candidate = str(priority or default or "normal").strip().lower()
    if candidate not in VALID_PRIORITIES:
        raise ValueError(f"Invalid priority. Must be one of: {', '.join(VALID_PRIORITIES)}")
    return candidate


+def _coerce_priority(priority: str | None, default: str = "normal") -> str:
+    try:
+        return _normalize_priority(priority, default)
+    except ValueError:
+        return default
+
+
 def _sorted_todos(agent_id: str) -> list[dict[str, Any]]:
    todos_list = [
        {**todo, "todo_id": todo_id} for todo_id, todo in _get_agent_todos(agent_id).items()
@@ -285,11 +292,16 @@ async def create_todo(ctx: RunContextWrapper, todos: str) -> str:
            - ``description`` (str, optional): extra context or
              acceptance criteria.
            - ``priority`` (str, optional): one of ``"low"`` /
-              ``"normal"`` / ``"high"`` / ``"critical"``. Defaults to
-              ``"normal"``.
+              ``"normal"`` / ``"high"`` / ``"critical"``. Anything else,
+              including omitting it, falls back to ``"normal"`` rather
+              than failing.

            Example: ``[{"title": "Probe /admin", "priority": "high"},
            {"title": "Check JWT alg=none"}]``.
+
+        A title already on the list, or repeated within this call, is
+        skipped rather than duplicated; skipped titles come back under
+        ``skipped``.
    """
    agent_id = _agent_id_from(ctx)
    try:
@@ -302,13 +314,21 @@ async def create_todo(ctx: RunContextWrapper, todos: str) -> str:
            )

        agent_todos = _get_agent_todos(agent_id)
+        seen = {todo["title"].strip().lower() for todo in agent_todos.values()}
        created: list[dict[str, Any]] = []
+        skipped: list[dict[str, str]] = []
        for task in tasks:
-            task_priority = _normalize_priority(task.get("priority"))
+            title = task["title"]
+            key = title.lower()
+            if key in seen:
+                skipped.append({"title": title, "reason": "duplicate title"})
+                continue
+            seen.add(key)
+            task_priority = _coerce_priority(task.get("priority"))
            todo_id = str(uuid.uuid4())[:6]
            timestamp = datetime.now(UTC).isoformat()
            agent_todos[todo_id] = {
-                "title": task["title"],
+                "title": title,
                "description": task.get("description"),
                "priority": task_priority,
                "status": "pending",
@@ -316,7 +336,7 @@ async def create_todo(ctx: RunContextWrapper, todos: str) -> str:
                "updated_at": timestamp,
                "completed_at": None,
            }
-            created.append({"todo_id": todo_id, "title": task["title"], "priority": task_priority})
+            created.append({"todo_id": todo_id, "title": title, "priority": task_priority})
    except (ValueError, TypeError) as e:
        return json.dumps(
            {"success": False, "error": f"Failed to create todo: {e}"},
@@ -330,6 +350,7 @@ async def create_todo(ctx: RunContextWrapper, todos: str) -> str:
            "success": True,
            "created": created,
            "created_count": len(created),
+            "skipped": skipped,
            "todos": _sorted_todos(agent_id),
            "total_count": len(_get_agent_todos(agent_id)),
        },
@@ -70,7 +70,6 @@ async def test_encoded_list_is_decoded_for_an_array_parameter(schema: dict[str,
        "auth",
        "Endpoint /admin leaks user data, and session tokens never expire",
        '"auth"',
-        "",
    ],
 )
 async def test_free_form_strings_are_never_split_into_an_array(value: str) -> None:
@@ -79,6 +78,29 @@ async def test_free_form_strings_are_never_split_into_an_array(value: str) -> No
    assert parsed["tags"] == value


+@pytest.mark.asyncio
+@pytest.mark.parametrize("schema", [_ARRAY, _NULLABLE_ARRAY])
+@pytest.mark.parametrize("value", ["", "   "])
+async def test_empty_string_becomes_an_empty_array(schema: dict[str, Any], value: str) -> None:
+    parsed = await _roundtrip(schema, {"tags": value})
+
+    assert parsed["tags"] == []
+
+
+@pytest.mark.asyncio
+async def test_empty_string_becomes_an_empty_object() -> None:
+    parsed = await _roundtrip(_OBJECT, {"modifications": ""})
+
+    assert parsed["modifications"] == {}
+
+
+@pytest.mark.asyncio
+async def test_empty_string_for_a_string_parameter_is_untouched() -> None:
+    parsed = await _roundtrip(_STRING, {"todos": ""})
+
+    assert parsed["todos"] == ""
+
+
@pytest.mark.asyncio
 async def test_encoded_mapping_is_decoded_for_an_object_parameter() -> None:
    parsed = await _roundtrip(_OBJECT, {"modifications": '{"method": "POST"}'})
@@ -6,34 +6,23 @@ from typing import TYPE_CHECKING, Any

 import pytest

-from strix.config import codex, grok
+from strix.config import codex
 from strix.interface import auth_cli


 if TYPE_CHECKING:
    from pathlib import Path

-_CHATGPT = auth_cli._PROVIDERS["chatgpt"]
-

@pytest.fixture(autouse=True)
 def _tmp_store(tmp_path: Path, monkeypatch: pytest.MonkeyPatch) -> None:
-    store = tmp_path / "home" / ".strix" / "subscription-auth.json"
-    monkeypatch.setattr(codex, "AUTH_PATH", store)
-    monkeypatch.setattr(grok, "AUTH_PATH", store)
+    monkeypatch.setattr(codex, "AUTH_PATH", tmp_path / "home" / ".strix" / "subscription-auth.json")


-def test_default_provider_is_chatgpt() -> None:
-    assert auth_cli._DEFAULT_PROVIDER == "chatgpt"
-    assert set(auth_cli._PROVIDERS) == {"chatgpt", "grok"}
-
-
-def test_provider_aliases_resolve() -> None:
-    assert auth_cli._resolve_provider(codex.PROVIDER) is _CHATGPT
-    assert auth_cli._resolve_provider("ChatGPT") is _CHATGPT
-    assert auth_cli._resolve_provider("grok") is auth_cli._PROVIDERS["grok"]
-    assert auth_cli._resolve_provider("xai") is auth_cli._PROVIDERS["grok"]
-    assert auth_cli._resolve_provider("gemini") is None
+def test_login_provider_is_chatgpt() -> None:
+    assert auth_cli.LOGIN_PROVIDER == "chatgpt"
+    assert codex.PROVIDER in auth_cli._ACCEPTED_PROVIDERS
+    assert "chatgpt" in auth_cli._ACCEPTED_PROVIDERS


 def test_unknown_subcommand_returns_usage_error() -> None:
@@ -62,32 +51,32 @@ def test_finish_requires_state_on_loopback(monkeypatch: pytest.MonkeyPatch) -> N

    # Loopback (require_state=True): missing or mismatched state is rejected.
    with pytest.raises(codex.CodexAuthError) as missing:
-        auth_cli._finish(_CHATGPT, "code", None, "verifier", "expected", require_state=True)
+        auth_cli._finish("code", None, "verifier", "expected", require_state=True)
    assert missing.value.code == "state_mismatch"
    with pytest.raises(codex.CodexAuthError) as mismatch:
-        auth_cli._finish(_CHATGPT, "code", "wrong", "verifier", "expected", require_state=True)
+        auth_cli._finish("code", "wrong", "verifier", "expected", require_state=True)
    assert mismatch.value.code == "state_mismatch"

    # Matching state proceeds to the exchange.
-    assert auth_cli._finish(
-        _CHATGPT, "code", "expected", "verifier", "expected", require_state=True
-    ) == {"ok": True}
+    assert auth_cli._finish("code", "expected", "verifier", "expected", require_state=True) == {
+        "ok": True
+    }


 def test_finish_manual_paste_allows_absent_state(monkeypatch: pytest.MonkeyPatch) -> None:
    monkeypatch.setattr(codex, "exchange_code", lambda *_: {"ok": True})
    # Manual paste (require_state=False): a bare code with no state is accepted,
    # but a present-and-wrong state is still rejected.
-    assert auth_cli._finish(
-        _CHATGPT, "code", None, "verifier", "expected", require_state=False
-    ) == {"ok": True}
+    assert auth_cli._finish("code", None, "verifier", "expected", require_state=False) == {
+        "ok": True
+    }
    with pytest.raises(codex.CodexAuthError):
-        auth_cli._finish(_CHATGPT, "code", "wrong", "verifier", "expected", require_state=False)
+        auth_cli._finish("code", "wrong", "verifier", "expected", require_state=False)


 def test_finish_rejects_missing_code() -> None:
    with pytest.raises(codex.CodexAuthError) as exc:
-        auth_cli._finish(_CHATGPT, None, "expected", "verifier", "expected", require_state=True)
+        auth_cli._finish(None, "expected", "verifier", "expected", require_state=True)
    assert exc.value.code == "no_code"


@@ -95,31 +84,6 @@ def test_model_subcommand_removed() -> None:
    assert auth_cli.run_auth(["model", "gpt-5.5"]) == 2


-def _sign_in_both() -> None:
-    codex.save_record({"type": "oauth", "access": "c", "refresh": "r", "account_id": "a"})
-    grok.save_record({"type": "oauth", "access": "g", "refresh": "r"})
-
-
-def test_logout_all_removes_every_provider() -> None:
-    _sign_in_both()
-    assert codex.is_authenticated()
-    assert grok.is_authenticated()
-
-    assert auth_cli.run_auth(["logout"]) == 0
-
-    assert not codex.is_authenticated()
-    assert not grok.is_authenticated()
-
-
-def test_logout_single_provider_leaves_the_other() -> None:
-    _sign_in_both()
-
-    assert auth_cli.run_auth(["logout", "grok"]) == 0
-
-    assert codex.is_authenticated()
-    assert not grok.is_authenticated()
-
-
@pytest.mark.parametrize("provider", ["chatgpt", "codex", "ChatGPT"])
 def test_login_accepts_provider_aliases(provider: str, monkeypatch: pytest.MonkeyPatch) -> None:
    reached = {"flow": False}
@@ -1228,3 +1228,40 @@ async def test_wait_kind_survives_a_snapshot_round_trip() -> None:
    assert restored.wait_kinds["root"] == "user"
    assert restored.idle_resume_counts["root"] == 1
    assert await execution._plain_waiting_timeout(restored, "root") is None
+
+
+@pytest.mark.asyncio
+async def test_interactive_nudge_offers_waiting_without_repeating() -> None:
+    """The nudge is the instruction an agent reads when it is stranded here.
+
+    It is where the option to wait on what was already said has to be, not only
+    in the system prompt: an agent that ended a turn on plain text reasons off
+    this text, and without the clause it restates its answer to reach a tool
+    call, so the user reads it twice.
+
+    The clause holds whatever the turn did, because the agent is the one who
+    knows whether it spoke — this fires for a turn that produced no text at all.
+    """
+    items = await execution._append_tool_required_message(
+        session=None,
+        context={"parent_id": None},
+        attempt=1,
+        limit=3,
+        interactive=True,
+    )
+
+    assert "with no message if you have already said it" in items[0]["content"]
+
+
+@pytest.mark.asyncio
+async def test_autonomous_nudge_does_not_offer_the_user() -> None:
+    """There is nobody attached to an autonomous run to wait for."""
+    items = await execution._append_tool_required_message(
+        session=None,
+        context={"parent_id": None},
+        attempt=1,
+        limit=3,
+        interactive=False,
+    )
+
+    assert "respond_to_user" not in items[0]["content"]
@@ -1,265 +0,0 @@
-"""Tests for Grok (xAI) subscription auth: PKCE, token handling, store."""
-
-from __future__ import annotations
-
-import base64
-import hashlib
-import json
-import time
-from typing import TYPE_CHECKING, Any
-from unittest import mock
-
-import pytest
-import requests
-
-from strix.config import grok
-
-
-if TYPE_CHECKING:
-    from pathlib import Path
-
-
-@pytest.fixture(autouse=True)
-def _tmp_store(tmp_path: Path, monkeypatch: pytest.MonkeyPatch) -> Path:
-    path = tmp_path / "home" / ".strix" / "subscription-auth.json"
-    monkeypatch.setattr(grok, "AUTH_PATH", path)
-    return path
-
-
-def test_pkce_challenge_matches_verifier_and_is_unpadded() -> None:
-    verifier, challenge = grok.generate_pkce()
-    expected = (
-        base64.urlsafe_b64encode(hashlib.sha256(verifier.encode()).digest()).rstrip(b"=").decode()
-    )
-    assert challenge == expected
-    assert "=" not in verifier
-    assert "=" not in challenge
-
-
-def test_authorize_url_carries_pkce_client_and_grok_scope() -> None:
-    url = grok.build_authorize_url("chal", "st8")
-    assert grok.AUTHORIZE_URL in url
-    assert "code_challenge=chal" in url
-    assert "code_challenge_method=S256" in url
-    assert f"client_id={grok.CLIENT_ID}" in url
-    assert "state=st8" in url
-    # The Grok-CLI scope is what unlocks subscription inference.
-    assert "grok-cli%3Aaccess" in url
-    assert "offline_access" in url
-
-
-def test_redirect_uri_is_loopback() -> None:
-    assert grok.REDIRECT_URI == "http://127.0.0.1:56121/callback"
-
-
-def test_post_form_returns_parsed_body() -> None:
-    resp = mock.MagicMock()
-    resp.status_code = 200
-    resp.content = b'{"access_token": "tok"}'
-
-    with mock.patch.object(requests, "post", return_value=resp) as post:
-        data = grok._post_form({"grant_type": "refresh_token"})
-
-    assert data == {"access_token": "tok"}
-    assert post.call_args.kwargs["timeout"] == grok._TOKEN_TIMEOUT
-
-
-def test_post_form_raises_on_http_error() -> None:
-    resp = mock.MagicMock()
-    resp.status_code = 400
-    resp.text = "invalid_grant"
-
-    with (
-        mock.patch.object(requests, "post", return_value=resp),
-        pytest.raises(grok.GrokAuthError) as exc,
-    ):
-        grok._post_form({"grant_type": "refresh_token"})
-    assert exc.value.code == "token_http_error"
-
-
-@pytest.mark.parametrize(
-    ("value", "expected"),
-    [
-        ("http://127.0.0.1:56121/callback?code=AAA&state=BBB", ("AAA", "BBB")),
-        ("AAA#BBB", ("AAA", "BBB")),
-        ("code=AAA&state=BBB", ("AAA", "BBB")),
-        ("AAA", ("AAA", None)),
-        ("", (None, None)),
-    ],
-)
-def test_parse_redirect_input(value: str, expected: tuple[str | None, str | None]) -> None:
-    assert grok.parse_redirect_input(value) == expected
-
-
-@pytest.mark.parametrize(
-    ("model", "expected"),
-    [
-        ("grok/grok-4", "grok-4"),
-        ("Grok/Grok-4", "Grok-4"),
-        ("  grok/grok-4  ", "grok-4"),
-        ("xai/grok-4", None),  # metered API path
-        ("chatgpt/gpt-5.4", None),
-        ("grok-4", None),
-        ("grok/", None),
-        ("", None),
-        (None, None),
-    ],
-)
-def test_subscription_model(model: str | None, expected: str | None) -> None:
-    assert grok.subscription_model(model) == expected
-
-
-def test_auth_mode() -> None:
-    assert grok.auth_mode("grok/grok-4") == "subscription"
-    assert grok.auth_mode("xai/grok-4") == "api_key"
-    assert grok.auth_mode("chatgpt/gpt-5.4") == "api_key"
-    assert grok.auth_mode(None) == "api_key"
-
-
-def _record(access: str, refresh: str, expires_at: float) -> dict[str, Any]:
-    return {
-        "type": "oauth",
-        "provider": "grok",
-        "access": access,
-        "refresh": refresh,
-        "expires_at": expires_at,
-    }
-
-
-def test_store_roundtrip_and_logout() -> None:
-    assert grok.read_record() is None
-    assert grok.is_authenticated() is False
-
-    grok.save_record(_record("a1", "r1", time.time() + 3600))
-    record = grok.read_record()
-    assert record is not None
-    assert record["access"] == "a1"
-    assert grok.is_authenticated() is True
-
-    grok.logout()
-    assert grok.read_record() is None
-    grok.logout()  # no-op when already gone
-
-
-def test_store_file_permissions_are_owner_only(_tmp_store: Path) -> None:
-    grok.save_record(_record("a1", "r1", time.time() + 3600))
-    assert (_tmp_store.stat().st_mode & 0o777) == 0o600
-
-
-def test_store_shares_file_with_other_providers(_tmp_store: Path) -> None:
-    # Grok must not clobber a co-resident ChatGPT record in the shared store.
-    _tmp_store.parent.mkdir(parents=True, exist_ok=True)
-    _tmp_store.write_text(json.dumps({"codex": {"type": "oauth", "access": "x"}}))
-
-    grok.save_record(_record("a1", "r1", time.time() + 3600))
-    on_disk = json.loads(_tmp_store.read_text())
-    assert on_disk["codex"] == {"type": "oauth", "access": "x"}
-    assert on_disk["grok"]["access"] == "a1"
-
-    grok.logout()
-    # Removing grok leaves the other provider's record and the file intact.
-    assert json.loads(_tmp_store.read_text()) == {"codex": {"type": "oauth", "access": "x"}}
-
-
-def test_read_record_rejects_incomplete_records() -> None:
-    grok.save_record({"type": "oauth", "access": "a"})  # missing refresh
-    assert grok.read_record() is None
-    assert grok.is_authenticated() is False
-
-
-def test_get_valid_token_returns_stored_when_fresh(monkeypatch: pytest.MonkeyPatch) -> None:
-    def _boom(_payload: dict[str, str]) -> dict[str, Any]:
-        msg = "should not refresh a fresh token"
-        raise AssertionError(msg)
-
-    monkeypatch.setattr(grok, "_post_form", _boom)
-    grok.save_record(_record("access-fresh", "r1", time.time() + 3600))
-    assert grok.get_valid_token() == "access-fresh"
-
-
-def test_get_valid_token_refreshes_and_persists_rotation(monkeypatch: pytest.MonkeyPatch) -> None:
-    calls = {"n": 0}
-
-    def _fake_post(payload: dict[str, str]) -> dict[str, Any]:
-        calls["n"] += 1
-        assert payload["grant_type"] == "refresh_token"
-        assert payload["refresh_token"] == "r1"
-        return {"access_token": "access-new", "refresh_token": "r2", "expires_in": 3600}
-
-    monkeypatch.setattr(grok, "_post_form", _fake_post)
-    grok.save_record(_record("stale", "r1", time.time() - 10))  # already expired
-
-    assert grok.get_valid_token() == "access-new"
-    assert calls["n"] == 1
-    record = grok.read_record()
-    assert record is not None
-    assert record["refresh"] == "r2"  # rotated refresh written back
-
-
-def test_refresh_keeps_old_refresh_when_response_omits_it(monkeypatch: pytest.MonkeyPatch) -> None:
-    def _fake_post(_payload: dict[str, str]) -> dict[str, Any]:
-        return {"access_token": "access-new", "expires_in": 3600}  # no refresh_token
-
-    monkeypatch.setattr(grok, "_post_form", _fake_post)
-    grok.save_record(_record("stale", "r1", time.time() - 10))
-
-    assert grok.get_valid_token() == "access-new"
-    record = grok.read_record()
-    assert record is not None
-    assert record["refresh"] == "r1"  # fell back to the prior refresh token
-
-
-def test_get_valid_token_uses_token_rotated_by_another_process(
-    monkeypatch: pytest.MonkeyPatch,
-) -> None:
-    records = [
-        _record("stale", "r1", time.time() - 10),
-        _record("fresh-from-other-process", "r2", time.time() + 3600),
-    ]
-    calls = {"n": 0}
-
-    def _fake_read() -> dict[str, Any]:
-        record = records[min(calls["n"], len(records) - 1)]
-        calls["n"] += 1
-        return record
-
-    def _boom(_payload: dict[str, str]) -> dict[str, Any]:
-        msg = "must not refresh a token another process already rotated"
-        raise AssertionError(msg)
-
-    monkeypatch.setattr(grok, "read_record", _fake_read)
-    monkeypatch.setattr(grok, "_post_form", _boom)
-
-    assert grok.get_valid_token() == "fresh-from-other-process"
-
-
-def test_get_valid_token_recovers_when_refresh_loses_race(
-    monkeypatch: pytest.MonkeyPatch,
-) -> None:
-    grok.save_record(_record("stale", "r1", time.time() - 10))
-
-    def _fake_post(_payload: dict[str, str]) -> dict[str, Any]:
-        grok.save_record(_record("fresh-from-peer", "r2", time.time() + 3600))
-        raise grok.GrokAuthError("token_http_error", "HTTP 400: invalid_grant")
-
-    monkeypatch.setattr(grok, "_post_form", _fake_post)
-    assert grok.get_valid_token() == "fresh-from-peer"
-
-
-def test_get_valid_token_reraises_refresh_error_without_rotation(
-    monkeypatch: pytest.MonkeyPatch,
-) -> None:
-    grok.save_record(_record("stale", "r1", time.time() - 10))
-
-    def _fake_post(_payload: dict[str, str]) -> dict[str, Any]:
-        raise grok.GrokAuthError("token_http_error", "HTTP 400: invalid_grant")
-
-    monkeypatch.setattr(grok, "_post_form", _fake_post)
-    with pytest.raises(grok.GrokAuthError):
-        grok.get_valid_token()
-
-
-def test_get_valid_token_raises_when_not_signed_in() -> None:
-    with pytest.raises(grok.GrokAuthError) as exc:
-        grok.get_valid_token()
-    assert exc.value.code == "not_authenticated"
@@ -1,102 +0,0 @@
-"""Grok subscription routing through StrixProvider.get_model."""
-
-from __future__ import annotations
-
-import argparse
-from unittest import mock
-
-from agents.models.openai_chatcompletions import OpenAIChatCompletionsModel
-
-from strix.config import grok, subscription
-from strix.config.models import StrixProvider, _TurnGuardModel
-from strix.interface import scan_setup, utils
-from strix.report import state as state_mod
-
-
-def test_grok_prefix_routes_to_chat_completions(monkeypatch) -> None:  # type: ignore[no-untyped-def]
-    client = mock.MagicMock()
-    monkeypatch.setattr(grok, "get_subscription_client", lambda: client)
-
-    model = StrixProvider().get_model("grok/grok-4")
-
-    assert isinstance(model, _TurnGuardModel)
-    assert isinstance(model._inner, OpenAIChatCompletionsModel)
-    # The provider strips the grok/ prefix and passes xAI's bare model slug.
-    assert model._inner.model == "grok-4"
-
-
-def test_non_subscription_model_is_not_hijacked_by_grok(monkeypatch) -> None:  # type: ignore[no-untyped-def]
-    def _boom() -> object:
-        msg = "grok client must not be built for a non-grok model"
-        raise AssertionError(msg)
-
-    monkeypatch.setattr(grok, "get_subscription_client", _boom)
-
-    # A metered xai/* key model must fall through to the normal provider path,
-    # not the subscription route.
-    model = StrixProvider().get_model("xai/grok-4")
-    assert isinstance(model, _TurnGuardModel)
-    assert not isinstance(model._inner, OpenAIChatCompletionsModel)
-
-
-def test_provider_label_names_the_subscription() -> None:
-    assert subscription.provider_label("grok/grok-4") == "Grok"
-    assert subscription.provider_label("chatgpt/gpt-5.4") == "ChatGPT"
-    # Metered API-key models are not subscriptions.
-    assert subscription.provider_label("xai/grok-4") is None
-    assert subscription.provider_label("openai/gpt-5.4") is None
-
-
-def test_run_record_reports_grok_provider(monkeypatch) -> None:  # type: ignore[no-untyped-def]
-    settings = mock.MagicMock()
-    settings.llm.model = "grok/grok-4"
-    monkeypatch.setattr(state_mod, "load_settings", lambda: settings)
-
-    record = state_mod.ReportState(run_name="run-test").run_record
-    assert record["auth_mode"] == "subscription"
-    assert record["subscription_provider"] == "Grok"
-
-
-def test_subscription_label_prefers_persisted_provider(monkeypatch) -> None:  # type: ignore[no-untyped-def]
-    settings = mock.MagicMock()
-    settings.llm.model = "chatgpt/gpt-5.4"  # current settings point at ChatGPT
-    monkeypatch.setattr(utils, "load_settings", lambda: settings)
-
-    # A resumed Grok run keeps its persisted provider even though settings changed.
-    resumed = mock.MagicMock(
-        run_record={"auth_mode": "subscription", "subscription_provider": "Grok"}
-    )
-    assert utils._subscription_label(resumed) == "Grok subscription"
-
-    # With no persisted provider, it derives the label from settings (not a
-    # hardcoded default).
-    fresh = mock.MagicMock(run_record={})
-    assert utils._subscription_label(fresh) == "ChatGPT subscription"
-
-
-def test_persisted_run_record_carries_provider(tmp_path, monkeypatch) -> None:  # type: ignore[no-untyped-def]
-    settings = mock.MagicMock()
-    settings.llm.model = "grok/grok-4"
-    monkeypatch.setattr(scan_setup, "load_settings", lambda: settings)
-    monkeypatch.setattr(scan_setup, "run_dir_for", lambda _name: tmp_path)
-    captured: dict[str, object] = {}
-    monkeypatch.setattr(
-        "strix.report.writer.write_run_record", lambda _dir, rec: captured.update(rec)
-    )
-
-    args = argparse.Namespace(
-        run_name="run-test",
-        targets_info=[],
-        scan_mode="scan",
-        instruction=None,
-        non_interactive=True,
-        local_sources=[],
-        diff_scope={"active": False},
-        scope_mode="mode",
-        diff_base=None,
-    )
-    scan_setup._persist_run_record(args)
-
-    # The resume/viewer record must carry the provider so resumed runs stay labeled.
-    assert captured["auth_mode"] == "subscription"
-    assert captured["subscription_provider"] == "Grok"
@@ -299,6 +299,16 @@ def test_make_model_settings_forces_required_for_anyllm_routed_openai_model() ->
    assert settings.tool_choice == "required"


+def test_make_model_settings_disables_parallel_tool_calls_by_default() -> None:
+    assert make_model_settings("none", model_name="gpt-4o").parallel_tool_calls is False
+
+
+def test_make_model_settings_omits_parallel_tool_calls_without_tools() -> None:
+    settings = make_model_settings("none", model_name="gpt-4o", has_tools=False)
+
+    assert settings.parallel_tool_calls is None
+
+
 def test_make_model_settings_sets_request_timeout() -> None:
    settings = make_model_settings(
        "none",
@@ -351,3 +361,32 @@ def test_make_model_settings_timeout_survives_reasoning_resolve() -> None:

    assert settings.extra_args is not None
    assert settings.extra_args["timeout"] == 120.0
+
+
+def test_openrouter_attribution_rides_on_the_request_headers() -> None:
+    # litellm.headers is ignored once a request carries any header of its own,
+    # so the attribution must be part of the per-request headers.
+    headers = make_model_settings(
+        None, model_name="openrouter/anthropic/claude-sonnet-4-5"
+    ).extra_headers
+    assert headers == {
+        "HTTP-Referer": "https://strix.ai",
+        "X-Title": "Strix",
+        "X-OpenRouter-Categories": "cli-agent",
+    }
+
+
+def test_openrouter_attribution_absent_for_other_providers() -> None:
+    assert make_model_settings(None, model_name="anthropic/claude-sonnet-4-5").extra_headers is None
+
+
+def test_user_headers_override_openrouter_attribution() -> None:
+    headers = make_model_settings(
+        None,
+        model_name="openrouter/anthropic/claude-sonnet-4-5",
+        extra_headers={"X-Title": "Custom", "X-Tenant": "acme"},
+    ).extra_headers
+    assert headers is not None
+    assert headers["X-Title"] == "Custom"
+    assert headers["X-Tenant"] == "acme"
+    assert headers["HTTP-Referer"] == "https://strix.ai"
@@ -64,3 +64,31 @@ async def test_a_message_that_already_arrived_is_taken_instead_of_parking() -> N
    assert result["wait_outcome"] == "message_arrived"
    assert result["pending_messages"] == 1
    assert coordinator.statuses["root"] == "running"
+
+
+async def _call_without_message(context: dict[str, Any]) -> dict[str, Any]:
+    ctx = ToolContext(
+        context=context,
+        tool_name="respond_to_user",
+        tool_call_id="call-1",
+        tool_arguments="{}",
+    )
+    raw = await respond_to_user.on_invoke_tool(ctx, "{}")
+    return json.loads(raw)  # type: ignore[no-any-return]
+
+
+@pytest.mark.asyncio
+async def test_parks_without_a_message() -> None:
+    """An agent that has already said its piece as plain text can just wait.
+
+    The nudge is what leaves it here, and while a message was required the only
+    way to stop was to send the same answer a second time.
+    """
+    context = await _context(interactive=True)
+
+    result = await _call_without_message(context)
+
+    assert result["success"] is True
+    assert result["wait_outcome"] == "waiting"
+    assert result["message"] == ""
+    assert context["coordinator"].statuses["root"] == "waiting"
@@ -0,0 +1,108 @@
+from __future__ import annotations
+
+import asyncio
+import types
+from typing import Any
+
+import pytest
+from agents import ModelSettings
+
+import strix.tools.notes.tools as notes_tools
+import strix.tools.todo.tools as todo_tools
+from strix.core import runner
+from strix.core.agents import AgentCoordinator
+from strix.runtime import session_manager
+
+
+def _wire_runner(monkeypatch: pytest.MonkeyPatch, tmp_path: Any) -> None:
+    monkeypatch.setattr(runner, "run_dir_for", lambda _scan_id: tmp_path)
+    monkeypatch.setattr(runner, "runtime_state_dir", lambda _run_dir: tmp_path)
+    monkeypatch.setattr(runner, "setup_scan_logging", lambda _run_dir: lambda: None)
+    monkeypatch.setattr(runner, "set_scan_id", lambda _scan_id: None)
+
+    settings = types.SimpleNamespace(
+        llm=types.SimpleNamespace(
+            model="openai/gpt-4o",
+            reasoning_effort="high",
+            force_required_tool_choice=False,
+            timeout=300,
+            prompt_cache=True,
+            extra_headers=None,
+        ),
+        runtime=types.SimpleNamespace(max_context_images=3),
+    )
+    monkeypatch.setattr(runner, "load_settings", lambda: settings)
+    monkeypatch.setattr(runner, "configure_sdk_model_defaults", lambda _settings: None)
+    monkeypatch.setattr(
+        runner, "uses_chat_completions_tool_schema", lambda _model, _settings: False
+    )
+    monkeypatch.setattr(todo_tools, "hydrate_todos_from_disk", lambda _state_dir: None)
+    monkeypatch.setattr(notes_tools, "hydrate_notes_from_disk", lambda _state_dir: None)
+
+    async def _create_or_reuse(*_args: Any, **_kwargs: Any) -> dict[str, Any]:
+        return {"client": object(), "session": object(), "caido_client": None}
+
+    async def _cleanup(*_args: Any, **_kwargs: Any) -> None:
+        return None
+
+    monkeypatch.setattr(session_manager, "create_or_reuse", _create_or_reuse)
+    monkeypatch.setattr(session_manager, "cleanup", _cleanup)
+    monkeypatch.setattr(runner, "build_root_task", lambda _scan_config: "task")
+    monkeypatch.setattr(runner, "build_scope_context", lambda _scan_config: "")
+    monkeypatch.setattr(runner, "make_model_settings", lambda *_a, **_k: ModelSettings())
+    monkeypatch.setattr(runner, "build_strix_agent", lambda **_kwargs: object())
+    monkeypatch.setattr(runner, "make_child_factory", lambda **_kwargs: lambda **_k: object())
+    monkeypatch.setattr(runner, "open_agent_session", lambda _root_id, _db: object())
+
+
+def _root_status(coordinator: AgentCoordinator) -> str:
+    roots = [aid for aid, parent in coordinator.parent_of.items() if parent is None]
+    assert len(roots) == 1
+    return coordinator.statuses[roots[0]]
+
+
+@pytest.mark.parametrize("interrupt", [KeyboardInterrupt, asyncio.CancelledError])
+@pytest.mark.asyncio
+async def test_user_interrupt_leaves_the_root_running_for_resume(
+    monkeypatch: pytest.MonkeyPatch, tmp_path: Any, interrupt: type[BaseException]
+) -> None:
+    _wire_runner(monkeypatch, tmp_path)
+
+    async def _interrupt(*_args: Any, **_kwargs: Any) -> None:
+        raise interrupt()
+
+    monkeypatch.setattr(runner, "run_agent_loop", _interrupt)
+    coordinator = AgentCoordinator()
+
+    with pytest.raises(interrupt):
+        await runner.run_strix_scan(
+            scan_config={"targets": [], "scan_mode": "deep"},
+            scan_id="scan-test",
+            image="img",
+            coordinator=coordinator,
+        )
+
+    assert _root_status(coordinator) == "running"
+
+
+@pytest.mark.asyncio
+async def test_a_real_crash_still_marks_root_failed(
+    monkeypatch: pytest.MonkeyPatch, tmp_path: Any
+) -> None:
+    _wire_runner(monkeypatch, tmp_path)
+
+    async def _boom(*_args: Any, **_kwargs: Any) -> None:
+        raise RuntimeError("boom")
+
+    monkeypatch.setattr(runner, "run_agent_loop", _boom)
+    coordinator = AgentCoordinator()
+
+    with pytest.raises(RuntimeError, match="boom"):
+        await runner.run_strix_scan(
+            scan_config={"targets": [], "scan_mode": "deep"},
+            scan_id="scan-test",
+            image="img",
+            coordinator=coordinator,
+        )
+
+    assert _root_status(coordinator) == "failed"
@@ -0,0 +1,93 @@
+from __future__ import annotations
+
+import asyncio
+import types
+from typing import Any
+
+import pytest
+from agents import ModelSettings
+
+import strix.tools.notes.tools as notes_tools
+import strix.tools.todo.tools as todo_tools
+from strix.core import runner
+from strix.core.agents import AgentCoordinator
+from strix.runtime import session_manager
+
+
+def _wire_runner(monkeypatch: pytest.MonkeyPatch, tmp_path: Any) -> None:
+    monkeypatch.setattr(runner, "run_dir_for", lambda _scan_id: tmp_path)
+    monkeypatch.setattr(runner, "runtime_state_dir", lambda _run_dir: tmp_path)
+    monkeypatch.setattr(runner, "setup_scan_logging", lambda _run_dir: lambda: None)
+    monkeypatch.setattr(runner, "set_scan_id", lambda _scan_id: None)
+
+    settings = _settings()
+    monkeypatch.setattr(runner, "load_settings", lambda: settings)
+    monkeypatch.setattr(runner, "configure_sdk_model_defaults", lambda _s: None)
+    monkeypatch.setattr(runner, "uses_chat_completions_tool_schema", lambda _m, _s: False)
+    monkeypatch.setattr(todo_tools, "hydrate_todos_from_disk", lambda _d: None)
+    monkeypatch.setattr(notes_tools, "hydrate_notes_from_disk", lambda _d: None)
+
+    async def _create_or_reuse(*_a: Any, **_k: Any) -> dict[str, Any]:
+        return {"client": object(), "session": object(), "caido_client": None}
+
+    async def _cleanup(*_a: Any, **_k: Any) -> None:
+        return None
+
+    monkeypatch.setattr(session_manager, "create_or_reuse", _create_or_reuse)
+    monkeypatch.setattr(session_manager, "cleanup", _cleanup)
+    monkeypatch.setattr(runner, "build_root_task", lambda _c: "task")
+    monkeypatch.setattr(runner, "build_scope_context", lambda _c: "")
+    monkeypatch.setattr(runner, "make_model_settings", lambda *_a, **_k: ModelSettings())
+    monkeypatch.setattr(runner, "build_strix_agent", lambda **_k: object())
+    monkeypatch.setattr(runner, "make_child_factory", lambda **_k: lambda **_kk: object())
+    monkeypatch.setattr(runner, "open_agent_session", lambda _root_id, _db: object())
+
+
+def _settings() -> Any:
+    return types.SimpleNamespace(
+        llm=types.SimpleNamespace(
+            model="openai/gpt-4o",
+            reasoning_effort="high",
+            force_required_tool_choice=False,
+            timeout=300,
+            prompt_cache=True,
+            extra_headers=None,
+        ),
+        runtime=types.SimpleNamespace(max_context_images=3),
+    )
+
+
+@pytest.mark.asyncio
+async def test_a_live_child_is_settled_before_sessions_close(
+    monkeypatch: pytest.MonkeyPatch, tmp_path: Any
+) -> None:
+    _wire_runner(monkeypatch, tmp_path)
+    coordinator = AgentCoordinator()
+    child_started = asyncio.Event()
+    child_task: dict[str, asyncio.Task[None]] = {}
+
+    async def _root_finishes(**kwargs: Any) -> None:
+        root_id = kwargs["agent_id"]
+
+        async def _child_mid_turn() -> None:
+            child_started.set()
+            await asyncio.sleep(3600)
+
+        await coordinator.register("child", "Child", parent_id=root_id)
+        task = asyncio.create_task(_child_mid_turn())
+        child_task["t"] = task
+        await coordinator.attach_runtime("child", task=task)
+        await child_started.wait()
+
+    monkeypatch.setattr(runner, "run_agent_loop", _root_finishes)
+
+    await runner.run_strix_scan(
+        scan_config={"targets": [], "scan_mode": "deep"},
+        scan_id="scan-test",
+        image="img",
+        coordinator=coordinator,
+    )
+
+    task = child_task["t"]
+    assert task.done(), "the child task was left running past scan teardown"
+    assert task.cancelled(), "the child was not cancelled cleanly on a finish"
@@ -0,0 +1,117 @@
+from __future__ import annotations
+
+import asyncio
+from pathlib import Path
+from typing import Any, cast
+
+import pytest
+
+from strix.core.sessions import open_agent_session
+
+
+def _count_open_fds() -> int | None:
+    for path in (Path("/proc/self/fd"), Path("/dev/fd")):
+        if path.is_dir():
+            return len(list(path.iterdir()))
+    return None
+
+
+@pytest.mark.asyncio
+async def test_sessions_hold_no_descriptors_while_parked(tmp_path: Path) -> None:
+    """Descriptor use must track live operations, not the number of sessions.
+
+    The SDK keeps a connection per (session, pool thread) open for the session's
+    whole life. An agent parks rather than exits, so its session lives for the
+    scan, and fan-out multiplies those handles until the process runs out of file
+    descriptors (#1018). A session that is not mid-operation should hold none.
+    """
+    baseline = _count_open_fds()
+    if baseline is None:
+        pytest.skip("no /proc/self/fd or /dev/fd on this platform")
+
+    sessions = [open_agent_session(f"a{i}", tmp_path / f"s{i}.db") for i in range(60)]
+    try:
+        for _ in range(4):
+            await asyncio.gather(
+                *(s.add_items([{"role": "user", "content": "x"}]) for s in sessions)
+            )
+            await asyncio.gather(*(s.get_items() for s in sessions))
+        parked = _count_open_fds()
+        assert parked is not None
+        # 60 parked sessions, yet descriptors are back at the baseline.
+        assert parked - baseline <= 5, f"parked fds grew by {parked - baseline}"
+    finally:
+        for s in sessions:
+            s.close()
+
+
+@pytest.mark.asyncio
+async def test_in_flight_descriptors_track_concurrency_not_session_count(
+    tmp_path: Path,
+) -> None:
+    baseline = _count_open_fds()
+    if baseline is None:
+        pytest.skip("no /proc/self/fd or /dev/fd on this platform")
+
+    sessions = [open_agent_session(f"a{i}", tmp_path / f"s{i}.db") for i in range(200)]
+    peak = baseline
+    try:
+
+        async def sample() -> None:
+            nonlocal peak
+            for _ in range(500):
+                current = _count_open_fds()
+                if current is not None:
+                    peak = max(peak, current)
+                await asyncio.sleep(0)
+
+        async def load() -> None:
+            for _ in range(4):
+                await asyncio.gather(
+                    *(s.add_items([{"role": "user", "content": "x"}]) for s in sessions)
+                )
+
+        await asyncio.gather(load(), sample())
+        # 200 sessions, but peak is bounded by the thread pool, well under 200.
+        assert peak - baseline < 100, f"in-flight fds peaked at +{peak - baseline}"
+    finally:
+        for s in sessions:
+            s.close()
+
+
+@pytest.mark.asyncio
+async def test_history_survives_the_per_operation_connection(tmp_path: Path) -> None:
+    session = open_agent_session("agent-1", tmp_path / "agents.db")
+    try:
+        for i in range(30):
+            await session.add_items([{"role": "user", "content": f"m{i}"}])
+        items = [cast("dict[str, Any]", i) for i in await session.get_items()]
+        assert [i["content"] for i in items] == [f"m{i}" for i in range(30)]
+    finally:
+        session.close()
+
+
+@pytest.mark.asyncio
+async def test_concurrent_sessions_sharing_one_file_stay_consistent(tmp_path: Path) -> None:
+    db = tmp_path / "shared.db"
+    sessions = [open_agent_session(f"a{i}", db) for i in range(10)]
+    try:
+        await asyncio.gather(
+            *(s.add_items([{"role": "user", "content": s.session_id}]) for s in sessions)
+        )
+        # Each session sees only its own row despite sharing the file.
+        for s in sessions:
+            items = [cast("dict[str, Any]", i) for i in await s.get_items()]
+            assert [i["content"] for i in items] == [s.session_id]
+    finally:
+        for s in sessions:
+            s.close()
+
+
+@pytest.mark.asyncio
+async def test_a_closed_session_refuses_operations(tmp_path: Path) -> None:
+    session = open_agent_session("agent-1", tmp_path / "agents.db")
+    await session.add_items([{"role": "user", "content": "x"}])
+    session.close()
+    with pytest.raises(RuntimeError, match="closed"):
+        await session.add_items([{"role": "user", "content": "y"}])
@@ -1,107 +0,0 @@
-"""Shared subscription credential store: secure writes and cross-provider locking."""
-
-from __future__ import annotations
-
-import fcntl
-import stat
-from typing import TYPE_CHECKING
-
-import pytest
-
-from strix.config import codex, grok, subscription_store
-
-
-if TYPE_CHECKING:
-    from pathlib import Path
-
-
-def test_write_creates_owner_only_file(tmp_path: Path) -> None:
-    path = tmp_path / ".strix" / "subscription-auth.json"
-    subscription_store.write(path, {"grok": {"type": "oauth", "access": "a", "refresh": "r"}})
-    assert stat.S_IMODE(path.stat().st_mode) == 0o600
-    # No stray temp file is left behind.
-    assert not path.with_suffix(".json.tmp").exists()
-
-
-def test_write_does_not_follow_a_symlink_at_target(tmp_path: Path) -> None:
-    store_dir = tmp_path / ".strix"
-    store_dir.mkdir()
-    outside = tmp_path / "attacker-target.json"
-    path = store_dir / "subscription-auth.json"
-    path.symlink_to(outside)  # attacker pre-plants a symlink at the store path
-
-    subscription_store.write(path, {"grok": {"type": "oauth", "access": "a", "refresh": "r"}})
-
-    # The atomic rename replaced the symlink with a real file; nothing was
-    # written through it to the attacker-chosen location.
-    assert not path.is_symlink()
-    assert not outside.exists()
-    assert subscription_store.read(path)["grok"]["access"] == "a"
-
-
-def test_providers_share_store_without_clobbering(
-    tmp_path: Path, monkeypatch: pytest.MonkeyPatch
-) -> None:
-    store = tmp_path / ".strix" / "subscription-auth.json"
-    monkeypatch.setattr(codex, "AUTH_PATH", store)
-    monkeypatch.setattr(grok, "AUTH_PATH", store)
-
-    codex.save_record({"type": "oauth", "access": "c", "refresh": "r", "account_id": "acct"})
-    grok.save_record({"type": "oauth", "access": "g", "refresh": "r"})
-
-    data = subscription_store.read(store)
-    assert data["codex"]["access"] == "c"
-    assert data["grok"]["access"] == "g"
-
-    # Logging one provider out leaves the other's credential intact.
-    grok.logout()
-    remaining = subscription_store.read(store)
-    assert "grok" not in remaining
-    assert remaining["codex"]["access"] == "c"
-
-
-def test_guard_is_reentrant(tmp_path: Path, monkeypatch: pytest.MonkeyPatch) -> None:
-    store = tmp_path / ".strix" / "subscription-auth.json"
-    monkeypatch.setattr(grok, "AUTH_PATH", store)
-    # Persisting while already holding the guard must not deadlock — this mirrors
-    # a token refresh saving its new record inside the refresh critical section.
-    with subscription_store.guard(store):
-        grok.save_record({"type": "oauth", "access": "g", "refresh": "r"})
-    record = grok.read_record()
-    assert record is not None
-    assert record["access"] == "g"
-
-
-def test_mutation_aborts_when_lock_cannot_be_acquired(
-    tmp_path: Path, monkeypatch: pytest.MonkeyPatch
-) -> None:
-    store = tmp_path / ".strix" / "subscription-auth.json"
-    monkeypatch.setattr(grok, "AUTH_PATH", store)
-
-    def _no_lock(*_args: object, **_kwargs: object) -> None:
-        raise OSError("no locks available")
-
-    monkeypatch.setattr(fcntl, "flock", _no_lock)
-
-    # Rather than silently doing an unlocked read-modify-write, the store raises
-    # and writes nothing.
-    with pytest.raises(subscription_store.StoreLockError):
-        grok.save_record({"type": "oauth", "access": "g", "refresh": "r"})
-    assert not store.exists()
-
-
-def test_lock_file_rejects_a_pre_positioned_symlink(
-    tmp_path: Path, monkeypatch: pytest.MonkeyPatch
-) -> None:
-    store_dir = tmp_path / ".strix"
-    store_dir.mkdir()
-    store = store_dir / "subscription-auth.json"
-    monkeypatch.setattr(grok, "AUTH_PATH", store)
-    # Attacker pre-plants a symlink where the lock file would be created.
-    outside = tmp_path / "attacker-target"
-    store.with_suffix(".lock").symlink_to(outside)
-
-    with pytest.raises(subscription_store.StoreLockError):
-        grok.save_record({"type": "oauth", "access": "g", "refresh": "r"})
-    # The symlink target was never created/truncated through the lock open.
-    assert not outside.exists()
@@ -0,0 +1,105 @@
+from __future__ import annotations
+
+import json
+from typing import Any
+
+import pytest
+from agents.tool_context import ToolContext
+
+from strix.tools.todo import tools
+from strix.tools.todo.tools import _coerce_priority, create_todo
+
+
+@pytest.fixture(autouse=True)
+def _isolate_store() -> Any:
+    tools._todos_storage.clear()
+    yield
+    tools._todos_storage.clear()
+
+
+async def _create(todos: list[Any], agent_id: str = "root") -> dict[str, Any]:
+    ctx = ToolContext(
+        context={"agent_id": agent_id},
+        tool_name="create_todo",
+        tool_call_id="call-1",
+        tool_arguments="{}",
+    )
+    raw = await create_todo.on_invoke_tool(ctx, json.dumps({"todos": json.dumps(todos)}))
+    return json.loads(raw)  # type: ignore[no-any-return]
+
+
+def test_unknown_priority_falls_back_to_normal() -> None:
+    assert _coerce_priority("medium") == "normal"
+    assert _coerce_priority("urgent") == "normal"
+    assert _coerce_priority("high") == "high"
+
+
+@pytest.mark.asyncio
+async def test_one_bad_priority_no_longer_discards_the_batch() -> None:
+    result = await _create(
+        [
+            {"title": "Recon", "priority": "medium"},
+            {"title": "Probe /admin", "priority": "sky-high"},
+            {"title": "Report"},
+        ]
+    )
+
+    assert result["success"] is True
+    assert result["created_count"] == 3
+    by_title = {c["title"]: c["priority"] for c in result["created"]}
+    assert by_title["Recon"] == "normal"
+    assert by_title["Probe /admin"] == "normal"
+    assert by_title["Report"] == "normal"
+
+
+@pytest.mark.asyncio
+async def test_duplicate_titles_within_a_batch_are_skipped() -> None:
+    result = await _create(
+        [
+            {"title": "Subdomain enumeration"},
+            {"title": "Content discovery"},
+            {"title": "Subdomain enumeration"},
+            {"title": "content discovery"},
+        ]
+    )
+
+    assert result["created_count"] == 2
+    assert {c["title"] for c in result["created"]} == {
+        "Subdomain enumeration",
+        "Content discovery",
+    }
+    assert len(result["skipped"]) == 2
+    assert all(s["reason"] == "duplicate title" for s in result["skipped"])
+
+
+@pytest.mark.asyncio
+async def test_a_title_already_on_the_list_is_not_created_again() -> None:
+    await _create([{"title": "Crawl with katana"}])
+    result = await _create([{"title": "crawl with katana"}, {"title": "JS analysis"}])
+
+    assert [c["title"] for c in result["created"]] == ["JS analysis"]
+    assert [s["title"] for s in result["skipped"]] == ["crawl with katana"]
+    assert result["total_count"] == 2
+
+
+def test_coerce_never_raises() -> None:
+    assert _coerce_priority("nonsense") == "normal"
+    assert _coerce_priority(None) == "normal"
+    assert _coerce_priority("high") == "high"
+    for value in (2, ["high"], {"p": 1}, True):
+        assert _coerce_priority(value) == "normal"  # type: ignore[arg-type]
+
+
+@pytest.mark.asyncio
+async def test_non_string_priority_does_not_fail_the_batch() -> None:
+    result = await _create(
+        [
+            {"title": "Recon", "priority": 2},
+            {"title": "Probe", "priority": ["high"]},
+            {"title": "Report"},
+        ]
+    )
+
+    assert result["success"] is True
+    assert result["created_count"] == 3
+    assert {c["priority"] for c in result["created"]} == {"normal"}
@@ -234,12 +234,11 @@ async def test_confirming_the_mount_starts_the_scan_without_a_target() -> None:


@pytest.mark.asyncio
-async def test_declining_the_mount_returns_to_the_start_screen() -> None:
-    started = False
+async def test_declining_the_mount_runs_without_one() -> None:
+    started: list[bool] = []

-    async def start(_verify: bool = True) -> None:
-        nonlocal started
-        started = True
+    async def start(verify: bool = True) -> None:
+        started.append(verify)

    os.environ["STRIX_LLM"] = "anthropic/claude-sonnet-4"
    os.environ["ANTHROPIC_API_KEY"] = "test-key"
@@ -250,14 +249,34 @@ async def test_declining_the_mount_returns_to_the_start_screen() -> None:
    result = await controller.handle("setup.confirm_mount", {"approved": False})

    assert result == {"approved": False}
-    # Nothing was prepared, so the session goes back to the start screen and can
-    # be launched again.
-    assert started is False
+    # Declining skips the directory; it does not abandon the scan.
+    assert started == [False]
    assert controller.workspace_mount is None
    assert controller.pending_workspace_mount is None
-    assert controller.setup_mode is True
-    assert controller.scan_started is False
-    assert controller.scan_state == "setup"
+    assert controller.setup_mode is False
+    assert controller.scan_started is True
+    assert controller.scan_state == "running"
+
+
+@pytest.mark.asyncio
+async def test_approving_the_mount_runs_with_it() -> None:
+    started: list[bool] = []
+
+    async def start(verify: bool = True) -> None:
+        started.append(verify)
+
+    os.environ["STRIX_LLM"] = "anthropic/claude-sonnet-4"
+    os.environ["ANTHROPIC_API_KEY"] = "test-key"
+    loader._cached = None
+    controller = TuiController(args(), on_start=start)
+    await controller.handle("setup.start", {"verify": False, "mount_working_dir": True})
+
+    result = await controller.handle("setup.confirm_mount", {"approved": True})
+
+    assert result == {"approved": True}
+    assert started == [False]
+    assert controller.workspace_mount == str(Path.cwd())
+    assert controller.scan_state == "running"


@pytest.mark.asyncio
@@ -7,19 +7,26 @@ shows what the user actually typed; resuming has to match that.

 from __future__ import annotations

+import ast
 import json
 import sqlite3
+from pathlib import Path
 from typing import TYPE_CHECKING, Any

 import pytest

+from strix.core import execution
 from strix.core.paths import runtime_state_dir
 from strix.interface.tui.backend.live_view import TuiLiveView as GoTuiLiveView
-from strix.interface.tui.live_view import TuiLiveView, _is_internal_agent_turn
+from strix.interface.tui.live_view import (
+    _INTERNAL_TURN_PREFIXES,
+    TuiLiveView,
+    _is_internal_agent_turn,
+)


 if TYPE_CHECKING:
-    from pathlib import Path
+    from types import ModuleType


 def _write_run(run_dir: Path, items: list[dict[str, Any]], agent_id: str = "root") -> None:
@@ -176,11 +183,60 @@ def test_internal_turn_classifier_matches_every_injected_form() -> None:
        "[CRITICAL] Turn budget: 480/500 used (96%).",
        "== Inherited context from parent (background only) ==",
        "Your previous message ended a turn without a tool call.",
-        "Your previous response ended the autonomous Strix run without a lifecycle tool call.",
+        "Your previous response ended the autonomous run without a lifecycle tool call.",
    ):
        assert _is_internal_agent_turn(content), content


+def _injected_strings(module: ModuleType) -> list[str]:
+    """Every string a module can inject, and nothing it merely mentions.
+
+    Parsing rather than searching the text keeps comments out of it, so a stale
+    copy of a message left in a comment cannot pass for the message itself. It
+    also joins adjacent literals for free, which the line wrapping needs, and
+    docstrings are dropped because they describe the code rather than run in it.
+    """
+    tree = ast.parse(Path(module.__file__ or "").read_text(encoding="utf-8"))
+    docstrings = set()
+    for node in ast.walk(tree):
+        if not isinstance(node, ast.Module | ast.ClassDef | ast.FunctionDef | ast.AsyncFunctionDef):
+            continue
+        first = node.body[0] if node.body else None
+        if isinstance(first, ast.Expr) and isinstance(first.value, ast.Constant):
+            docstrings.add(id(first.value))
+
+    literals: list[str] = []
+    for node in ast.walk(tree):
+        if isinstance(node, ast.Constant):
+            if isinstance(node.value, str) and id(node) not in docstrings:
+                literals.append(node.value)
+        elif isinstance(node, ast.JoinedStr):
+            literals.append(
+                "".join(
+                    part.value
+                    for part in node.values
+                    if isinstance(part, ast.Constant) and isinstance(part.value, str)
+                )
+            )
+    return literals
+
+
+def test_internal_turn_prefixes_still_match_what_is_injected() -> None:
+    """The classifier copies sentences out of another module, so they can drift.
+
+    Both nudges are written inline in strix.core.execution, so there is nothing to
+    import and compare against. Read them back out of what that module can inject.
+    """
+    injected = _injected_strings(execution)
+    nudges = [prefix for prefix in _INTERNAL_TURN_PREFIXES if prefix.startswith("Your previous")]
+    assert nudges, "the no-tool-call nudges are no longer in the classifier"
+    for nudge in nudges:
+        assert any(nudge in literal for literal in injected), (
+            f"the classifier expects {nudge!r}, which strix.core.execution no longer "
+            f"injects. A resumed scan would show that nudge as the user's own message."
+        )
+
+
 def test_internal_turn_classifier_keeps_bracketed_user_text() -> None:
    """A leading bracket is not enough: typed text often starts with one."""
    for content in (
@@ -70,53 +70,6 @@ def test_read_run_summary_finished_flag(tmp_path: Path) -> None:
    assert read_run_summary(partial)["finished"] is False


-def _write_record(base: Path, name: str, record: dict[str, object]) -> Path:
-    run_dir = base / "strix_runs" / name
-    run_dir.mkdir(parents=True)
-    (run_dir / "run.json").write_text(json.dumps(record), encoding="utf-8")
-    return run_dir
-
-
-def test_read_run_summary_backfills_subscription_provider(tmp_path: Path) -> None:
-    # An older subscription run recorded no provider name; it is derived from
-    # the recorded provider/model slug so the viewer can label it.
-    run_dir = _write_record(
-        tmp_path,
-        "grok-run",
-        {
-            "auth_mode": "subscription",
-            "llm_usage": {"agents": [{"agent_id": "root", "model": "grok/grok-4"}]},
-        },
-    )
-    assert read_run_summary(run_dir)["subscription_provider"] == "Grok"
-
-
-def test_read_run_summary_keeps_explicit_provider(tmp_path: Path) -> None:
-    run_dir = _write_record(
-        tmp_path,
-        "chatgpt-run",
-        {
-            "auth_mode": "subscription",
-            "subscription_provider": "ChatGPT",
-            "llm_usage": {"agents": [{"agent_id": "root", "model": "grok/grok-4"}]},
-        },
-    )
-    # An explicit field is authoritative and never overwritten by the slug.
-    assert read_run_summary(run_dir)["subscription_provider"] == "ChatGPT"
-
-
-def test_read_run_summary_ignores_api_key_runs(tmp_path: Path) -> None:
-    run_dir = _write_record(
-        tmp_path,
-        "api-key-run",
-        {
-            "auth_mode": "api_key",
-            "llm_usage": {"agents": [{"agent_id": "root", "model": "openai/gpt-5.4"}]},
-        },
-    )
-    assert "subscription_provider" not in read_run_summary(run_dir)
-
-
 def test_read_missing_artifacts_return_defaults(tmp_path: Path) -> None:
    run_dir = _make_run(tmp_path, "empty", status="running", end_time=None)
    assert read_vulnerabilities(run_dir) == []
@@ -2378,7 +2378,7 @@ wheels = [

 [[package]]
 name = "strix-agent"
-version = "1.5.1"
+version = "1.5.3"
 source = { editable = "." }
 dependencies = [
    { name = "caido-sdk-client" },
Author	SHA1	Message	Date
Ahmed AllamandAhmed Allam	7cc9fa9faa	chore: release v1.5.3	2026-08-10 21:28:52 +03:00
devin-ai-integration[bot]andGitHub	174c16fa26	fix(llm): send OpenRouter app attribution on the request itself (#1045 )	2026-08-10 11:24:02 -07:00
Ahmed AllamandAhmed Allam	94a2586aaa	fix(container): write the browser profile as root	2026-08-10 10:08:17 +03:00
Ahmed AllamandAhmed Allam	372e27fa17	chore(container): drop explanatory comment	2026-08-10 09:54:49 +03:00
Ahmed AllamandAhmed Allam	ad727edd66	fix(container): keep the browser env alive where image ENV is dropped	2026-08-10 09:54:49 +03:00
devin-ai-integration[bot]GitHub Ahmed Allam	7b3c8f9b74	fix(container): reclaim abandoned browser sessions (#1034 ) Co-authored-by: Ahmed Allam <ahmed39652003@gmail.com>	2026-08-09 16:57:51 -07:00
Ahmed AllamandAhmed Allam	ae07af6159	chore: drop explanatory comment	2026-08-09 15:44:16 +03:00
Ahmed AllamandAhmed Allam	649a2e2140	fix(llm): omit parallel_tool_calls on tool-less requests	2026-08-09 15:44:16 +03:00
Ahmed AllamandAhmed Allam	597aae6715	chore: release v1.5.2	2026-08-09 04:29:34 +03:00
Ahmed AllamandGitHub	06b158d1fa	fix(runner): settle child agents before closing sessions at wind-down (#1025 )	2026-08-08 18:17:58 -07:00
Ahmed AllamandGitHub	c29eb73c7f	fix(tools): coerce an empty-string list/dict argument to an empty container (#1024 )	2026-08-08 16:58:30 -07:00
Ahmed AllamandGitHub	72833b8e43	fix(runner): resume after a user interrupt instead of failing (#1023 )	2026-08-08 16:44:12 -07:00
Ahmed AllamandGitHub	1117ba6d4a	fix(sessions): open a sqlite connection per operation, not per thread (#1022 )	2026-08-08 16:20:01 -07:00
Ahmed AllamandGitHub	53e4658d88	fix(todo): stop a todo plan failing on priority or duplicates (#1021 )	2026-08-08 15:18:48 -07:00
Ahmed Allam GitHub Claude Opus 4.8	58df71d3db	fix(agents): let an agent wait on what it already said (#1020 ) * let an agent wait on what it already said An agent that answers in plain text is nudged to call a tool, and the only tool that hands control back takes a required message. So it says the same thing twice: once as text the user has already read, once as the argument it had to supply to stop. Seen on a run whose whole instruction was "hi" - a greeting, then the same greeting again through respond_to_user. message is optional now. The nudge arms the tool with the text that was delivered and says not to repeat it, so an agent that has said its piece can park on it with an empty call. Anything it does want to add it passes normally. Parking still cannot leave the user on silence: an empty call is refused unless something was actually said, and the arming is single use - execution clears it as soon as a turn ends any other way. The interactive prompt now also says to answer and stop in one respond_to_user call, which is what avoids the nudge in the first place. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * drop the worked example from the interactive prompt "the user greeted you, asked something you can answer outright, or you need a decision" was the run I had been reading, written into a rule that holds whatever the reason. The rule is that replying and stopping is one call; listing occasions only invites the model to check whether this is one of them. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * drop the arming flag; an empty message just waits Passing the delivered text from execution into the tool, and refusing an empty call without it, was machinery guarding against an agent parking having said nothing. That leaves the user looking at "waiting for your reply" with a cursor in front of them - they type. It does not need a mechanism. What is left is the default on message, and the nudge saying the text already landed. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * only offer waiting on words that were written The nudge told every agent its text had already been delivered, but it fires whenever a turn leaves the agent running, and a turn can end with no tool call and no text at all - _final_output_preview has carried <none> and <empty> branches all along. An agent that said nothing was being invited to wait on an answer the user never received, leaving them at a bare prompt. It now reads the turn: waiting on what was said is offered only when something was, and otherwise the agent is told plainly that the user has read nothing and to send its message. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * leave the continuation nudge alone Rewording it meant asserting from the outside whether the agent had spoken, and the nudge fires whenever a turn leaves the agent running - text or no text. The agent knows which it did without being told, so the guidance belongs in its prompt, where the condition is its own to read. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * say it in the nudge, where the agent is reading An agent stranded by the nudge reasons off the nudge. Told only to call respond_to_user, it supplies a message, and since it has just answered in plain text that message is the same answer again. The system prompt saying otherwise sits thousands of tokens earlier and loses. The clause goes on the line the agent acts on: call respond_to_user, with no message if it has already said it. That reads true whatever the turn did, including one that produced no text, because the agent is the one who knows which — nothing here has to work it out from the outside. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>	2026-08-09 00:57:16 +03:00
Ahmed Allam GitHub Claude Opus 4.8	0b9e029a5d	test(tui): correct the nudge the internal-turn test asserts (#1016 ) * correct the nudge the internal-turn test asserts The test expected "ended the autonomous Strix run", which strix.core.execution does not inject; it says "ended the autonomous run". The classifier was right and the test was not, so the suite failed on main while the behaviour it guards was fine. The sentence is written inline in another module and copied by hand into the classifier and again into the test, which is how it drifted. A second test now reads it back out of that module's source, joining the adjacent string literals its line wrapping leaves behind, and fails if either nudge is no longer injected verbatim. Reworded one and it reports which nudge went missing and what a resumed scan would do about it. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * read the nudges out of what the module can inject, not out of its text Searching the source accepted the sentence anywhere in the file, so a stale copy left behind in a comment would have kept the guard passing after the message it guards had changed - the drift it exists to catch. Parsing the module instead limits it to strings the code can actually inject. Comments never reach the tree, docstrings are dropped as description rather than behaviour, and adjacent literals are joined during parsing, which the line wrapping needed and the regex was only approximating. Checked by rewording the message and leaving the old wording in a comment: the guard fails, where searching the text passed. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>	2026-08-08 22:34:45 +03:00
Ahmed Allam GitHub Claude Opus 4.8	b260a4ee38	fix(tui): make the mount prompt clickable, and skip the mount instead of abandoning the scan (#1015 ) * make the working-directory prompt answer the mouse Its Confirm and Cancel were drawn as buttons and did nothing when clicked: the modal mouse handler had a case for every dialog except this one, so a click fell through and the scan sat waiting on an answer the user believed they had given. Only the keyboard could answer it. The prompt is docked in a corner rather than centered, so it also needs its own bounds; the centered ones every other dialog uses would have put the buttons in the wrong place. Those bounds now come from the same placement cornerOverlay draws with. Two returns that hand back the model alongside a call that mutates it are now sequenced explicitly. They work, but only because the compiler happens to evaluate the call first, and one of them is what puts the prompt back in the composer when the mount is declined. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * skip the mount instead of abandoning the scan Declining the working-directory prompt threw the whole launch away and dropped back to the start screen, which is a lot to lose for answering one question about one directory. The two answers are now about the directory alone: mount it, or run without it. The prompt is the whole of the input either way. The buttons say which is which - Mount and Skip rather than Confirm and Cancel - and the prompt says what skipping costs. A run with neither target nor directory is a real run, so two things follow it. It can be resumed: its instruction is what drives it, and that is in the run record. And it tells the agent plainly that it has neither, because an agent given no scope goes looking for the one it assumes it was meant to have. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>	2026-08-08 22:34:02 +03:00