strix-agent 0.1.16__tar.gz → 0.1.18__tar.gz

{strix_agent-0.1.16 → strix_agent-0.1.18}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: strix-agent
-Version: 0.1.16
+Version: 0.1.18
 Summary: Open-source AI Hackers for your apps
 License: Apache-2.0
 Keywords: cybersecurity,security,vulnerability,scanner,pentest,agent,ai,cli
@@ -129,12 +129,12 @@ strix --target api.your-app.com --instruction "Prioritize authentication and aut
 ### ⚙️ Configuration
 
 ```bash
-# Required
 export STRIX_LLM="openai/gpt-5"
 export LLM_API_KEY="your-api-key"
 
-# Recommended
-export PERPLEXITY_API_KEY="your-api-key"
+# Optional
+export LLM_API_BASE="your-api-base-url"  # if using a local model, e.g. Ollama, LMStudio
+export PERPLEXITY_API_KEY="your-api-key"  # for search capabilities
 ```
 
 [📚 View supported AI models](https://docs.litellm.ai/docs/providers)

{strix_agent-0.1.16 → strix_agent-0.1.18}/README.md

@@ -89,12 +89,12 @@ strix --target api.your-app.com --instruction "Prioritize authentication and aut
 ### ⚙️ Configuration
 
 ```bash
-# Required
 export STRIX_LLM="openai/gpt-5"
 export LLM_API_KEY="your-api-key"
 
-# Recommended
-export PERPLEXITY_API_KEY="your-api-key"
+# Optional
+export LLM_API_BASE="your-api-base-url"  # if using a local model, e.g. Ollama, LMStudio
+export PERPLEXITY_API_KEY="your-api-key"  # for search capabilities
 ```
 
 [📚 View supported AI models](https://docs.litellm.ai/docs/providers)

{strix_agent-0.1.16 → strix_agent-0.1.18}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "strix-agent"
-version = "0.1.16"
+version = "0.1.18"
 description = "Open-source AI Hackers for your apps"
 authors = ["Strix <hi@usestrix.com>"]
 readme = "README.md"

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/agents/base_agent.py

@@ -181,10 +181,21 @@ class BaseAgent(metaclass=AgentMeta):
                 continue
 
             except LLMRequestFailedError as e:
-                self.state.add_error(f"LLM request failed: {e}")
+                error_msg = str(e)
+                error_details = getattr(e, "details", None)
+                self.state.add_error(error_msg)
                 self.state.enter_waiting_state(llm_failed=True)
                 if tracer:
-                    tracer.update_agent_status(self.state.agent_id, "llm_failed")
+                    tracer.update_agent_status(self.state.agent_id, "llm_failed", error_msg)
+                    if error_details:
+                        tracer.log_tool_execution_start(
+                            self.state.agent_id,
+                            "llm_error_details",
+                            {"error": error_msg, "details": error_details},
+                        )
+                        tracer.update_tool_execution(
+                            tracer._next_execution_id - 1, "failed", error_details
+                        )
                 continue
 
             except (RuntimeError, ValueError, TypeError) as e:

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/cli/app.py

@@ -9,6 +9,8 @@ import threading
 from collections.abc import Callable
 from typing import Any, ClassVar
 
+from rich.markup import escape as rich_escape
+from rich.text import Text
 from textual import events, on
 from textual.app import App, ComposeResult
 from textual.binding import Binding
@@ -24,7 +26,7 @@ from strix.llm.config import LLMConfig
 
 
 def escape_markup(text: str) -> str:
-    return text.replace("[", "\\[").replace("]", "\\]")
+    return rich_escape(text)
 
 
 class ChatTextArea(TextArea):  # type: ignore[misc]
@@ -483,7 +485,7 @@ class StrixCLIApp(App):  # type: ignore[misc]
                 self._displayed_events = current_event_ids
 
         chat_display = self.query_one("#chat_display", Static)
-        chat_display.update(content)
+        self._update_static_content_safe(chat_display, content)
 
         chat_display.set_classes(css_class)
 
@@ -546,11 +548,18 @@ class StrixCLIApp(App):  # type: ignore[misc]
                 self._safe_widget_operation(keymap_indicator.update, "")
                 self._safe_widget_operation(status_display.remove_class, "hidden")
             elif status == "llm_failed":
-                self._safe_widget_operation(status_text.update, "[red]LLM request failed[/red]")
+                error_msg = agent_data.get("error_message", "")
+                display_msg = (
+                    f"[red]{escape_markup(error_msg)}[/red]"
+                    if error_msg
+                    else "[red]LLM request failed[/red]"
+                )
+                self._safe_widget_operation(status_text.update, display_msg)
                 self._safe_widget_operation(
                     keymap_indicator.update, "[dim]Send message to retry[/dim]"
                 )
                 self._safe_widget_operation(status_display.remove_class, "hidden")
+                self._stop_dot_animation()
             elif status == "waiting":
                 animated_text = self._get_animated_waiting_text(self.selected_agent_id)
                 self._safe_widget_operation(status_text.update, animated_text)
@@ -633,7 +642,7 @@ class StrixCLIApp(App):  # type: ignore[misc]
 
         for agent_id, agent_data in self.tracer.agents.items():
             status = agent_data.get("status", "running")
-            if status in ["running", "waiting", "llm_failed"]:
+            if status in ["running", "waiting"]:
                 has_active_agents = True
                 current_dots = self._agent_dot_states.get(agent_id, 0)
                 self._agent_dot_states[agent_id] = (current_dots + 1) % 4
@@ -644,7 +653,7 @@ class StrixCLIApp(App):  # type: ignore[misc]
             and self.selected_agent_id in self.tracer.agents
         ):
             selected_status = self.tracer.agents[self.selected_agent_id].get("status", "running")
-            if selected_status in ["running", "waiting", "llm_failed"]:
+            if selected_status in ["running", "waiting"]:
                 self._update_agent_status_display()
 
         if not has_active_agents:
@@ -652,7 +661,7 @@ class StrixCLIApp(App):  # type: ignore[misc]
             for agent_id in list(self._agent_dot_states.keys()):
                 if agent_id not in self.tracer.agents or self.tracer.agents[agent_id].get(
                     "status"
-                ) not in ["running", "waiting", "llm_failed"]:
+                ) not in ["running", "waiting"]:
                     del self._agent_dot_states[agent_id]
 
     def _gather_agent_events(self, agent_id: str) -> list[dict[str, Any]]:
@@ -900,6 +909,7 @@ class StrixCLIApp(App):  # type: ignore[misc]
             "reporting_action": "#ea580c",
             "scan_start_info": "#22c55e",
             "subagent_start_info": "#22c55e",
+            "llm_error_details": "#dc2626",
         }
 
         color = tool_colors.get(tool_name, "#737373")
@@ -911,6 +921,14 @@ class StrixCLIApp(App):  # type: ignore[misc]
         if renderer:
             widget = renderer.render(tool_data)
             content = str(widget.renderable)
+        elif tool_name == "llm_error_details":
+            lines = ["[red]✗ LLM Request Failed[/red]"]
+            if args.get("details"):
+                details = args["details"]
+                if len(details) > 300:
+                    details = details[:297] + "..."
+                lines.append(f"[dim]Details:[/dim] {escape_markup(details)}")
+            content = "\n".join(lines)
         else:
             status_icons = {
                 "running": "[yellow]●[/yellow]",
@@ -1127,6 +1145,19 @@ class StrixCLIApp(App):  # type: ignore[misc]
         else:
             return True
 
+    def _update_static_content_safe(self, widget: Static, content: str) -> None:
+        try:
+            widget.update(content)
+        except Exception:  # noqa: BLE001
+            try:
+                safe_text = Text.from_markup(content)
+                widget.update(safe_text)
+            except Exception:  # noqa: BLE001
+                import re
+
+                plain_text = re.sub(r"\[.*?\]", "", content)
+                widget.update(plain_text)
+
 
 async def run_strix_cli(args: argparse.Namespace) -> None:
     app = StrixCLIApp(args)

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/cli/main.py

@@ -48,8 +48,23 @@ def validate_environment() -> None:
     if not os.getenv("STRIX_LLM"):
         missing_required_vars.append("STRIX_LLM")
 
+    has_base_url = any(
+        [
+            os.getenv("LLM_API_BASE"),
+            os.getenv("OPENAI_API_BASE"),
+            os.getenv("LITELLM_BASE_URL"),
+            os.getenv("OLLAMA_API_BASE"),
+        ]
+    )
+
     if not os.getenv("LLM_API_KEY"):
-        missing_required_vars.append("LLM_API_KEY")
+        if not has_base_url:
+            missing_required_vars.append("LLM_API_KEY")
+        else:
+            missing_optional_vars.append("LLM_API_KEY")
+
+    if not has_base_url:
+        missing_optional_vars.append("LLM_API_BASE")
 
     if not os.getenv("PERPLEXITY_API_KEY"):
         missing_optional_vars.append("PERPLEXITY_API_KEY")
@@ -65,40 +80,72 @@ def validate_environment() -> None:
             error_text.append(" is not set\n", style="white")
 
         if missing_optional_vars:
-            error_text.append(
-                "\nOptional (but recommended) environment variables:\n", style="dim white"
-            )
+            error_text.append("\nOptional environment variables:\n", style="dim white")
             for var in missing_optional_vars:
                 error_text.append(f"• {var}", style="dim yellow")
                 error_text.append(" is not set\n", style="dim white")
 
         error_text.append("\nRequired environment variables:\n", style="white")
-        error_text.append("• ", style="white")
-        error_text.append("STRIX_LLM", style="bold cyan")
-        error_text.append(
-            " - Model name to use with litellm (e.g., 'openai/gpt-5')\n",
-            style="white",
-        )
-        error_text.append("• ", style="white")
-        error_text.append("LLM_API_KEY", style="bold cyan")
-        error_text.append(" - API key for the LLM provider\n", style="white")
+        for var in missing_required_vars:
+            if var == "STRIX_LLM":
+                error_text.append("• ", style="white")
+                error_text.append("STRIX_LLM", style="bold cyan")
+                error_text.append(
+                    " - Model name to use with litellm (e.g., 'openai/gpt-5')\n",
+                    style="white",
+                )
+            elif var == "LLM_API_KEY":
+                error_text.append("• ", style="white")
+                error_text.append("LLM_API_KEY", style="bold cyan")
+                error_text.append(
+                    " - API key for the LLM provider (required for cloud providers)\n",
+                    style="white",
+                )
 
         if missing_optional_vars:
             error_text.append("\nOptional environment variables:\n", style="white")
-            error_text.append("• ", style="white")
-            error_text.append("PERPLEXITY_API_KEY", style="bold cyan")
-            error_text.append(
-                " - API key for Perplexity AI web search (enables real-time research)\n",
-                style="white",
-            )
+            for var in missing_optional_vars:
+                if var == "LLM_API_KEY":
+                    error_text.append("• ", style="white")
+                    error_text.append("LLM_API_KEY", style="bold cyan")
+                    error_text.append(" - API key for the LLM provider\n", style="white")
+                elif var == "LLM_API_BASE":
+                    error_text.append("• ", style="white")
+                    error_text.append("LLM_API_BASE", style="bold cyan")
+                    error_text.append(
+                        " - Custom API base URL if using local models (e.g., Ollama, LMStudio)\n",
+                        style="white",
+                    )
+                elif var == "PERPLEXITY_API_KEY":
+                    error_text.append("• ", style="white")
+                    error_text.append("PERPLEXITY_API_KEY", style="bold cyan")
+                    error_text.append(
+                        " - API key for Perplexity AI web search (enables real-time research)\n",
+                        style="white",
+                    )
 
         error_text.append("\nExample setup:\n", style="white")
         error_text.append("export STRIX_LLM='openai/gpt-5'\n", style="dim white")
-        error_text.append("export LLM_API_KEY='your-api-key-here'\n", style="dim white")
+
+        if "LLM_API_KEY" in missing_required_vars:
+            error_text.append("export LLM_API_KEY='your-api-key-here'\n", style="dim white")
+
         if missing_optional_vars:
-            error_text.append(
-                "export PERPLEXITY_API_KEY='your-perplexity-key-here'", style="dim white"
-            )
+            for var in missing_optional_vars:
+                if var == "LLM_API_KEY":
+                    error_text.append(
+                        "export LLM_API_KEY='your-api-key-here'  # optional with local models\n",
+                        style="dim white",
+                    )
+                elif var == "LLM_API_BASE":
+                    error_text.append(
+                        "export LLM_API_BASE='http://localhost:11434'  # needed for local models only\n",
+                        style="dim white",
+                    )
+                elif var == "PERPLEXITY_API_KEY":
+                    error_text.append(
+                        "export PERPLEXITY_API_KEY='your-perplexity-key-here'\n", style="dim white"
+                    )
 
         panel = Panel(
             error_text,
@@ -152,6 +199,15 @@ async def warm_up_llm() -> None:
         if api_key:
             litellm.api_key = api_key
 
+        api_base = (
+            os.getenv("LLM_API_BASE")
+            or os.getenv("OPENAI_API_BASE")
+            or os.getenv("LITELLM_BASE_URL")
+            or os.getenv("OLLAMA_API_BASE")
+        )
+        if api_base:
+            litellm.api_base = api_base
+
         test_messages = [
             {"role": "system", "content": "You are a helpful assistant."},
             {"role": "user", "content": "Reply with just 'OK'."},

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/cli/tool_components/agents_graph_renderer.py

@@ -31,7 +31,7 @@ class CreateAgentRenderer(BaseToolRenderer):
         task = args.get("task", "")
         name = args.get("name", "Agent")
 
-        header = f"🤖 [bold #fbbf24]Creating {name}[/]"
+        header = f"🤖 [bold #fbbf24]Creating {cls.escape_markup(name)}[/]"
 
         if task:
             task_display = task[:400] + "..." if len(task) > 400 else task

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/cli/tool_components/base_renderer.py

@@ -1,6 +1,7 @@
 from abc import ABC, abstractmethod
 from typing import Any, ClassVar
 
+from rich.markup import escape as rich_escape
 from textual.widgets import Static
 
 
@@ -16,7 +17,7 @@ class BaseToolRenderer(ABC):
 
     @classmethod
     def escape_markup(cls, text: str) -> str:
-        return text.replace("[", "\\[").replace("]", "\\]")
+        return rich_escape(text)
 
     @classmethod
     def format_args(cls, args: dict[str, Any], max_length: int = 500) -> str:

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/cli/tool_components/browser_renderer.py

@@ -76,7 +76,7 @@ class BrowserRenderer(BaseToolRenderer):
                     "double_click": "double clicking",
                     "hover": "hovering",
                 }
-                message = action_words[action]
+                message = cls.escape_markup(action_words[action])
 
             return f"{browser_icon} [#06b6d4]{message}[/]"
 
@@ -97,9 +97,9 @@ class BrowserRenderer(BaseToolRenderer):
         }
 
         if action in simple_actions:
-            return f"{browser_icon} [#06b6d4]{simple_actions[action]}[/]"
+            return f"{browser_icon} [#06b6d4]{cls.escape_markup(simple_actions[action])}[/]"
 
-        return f"{browser_icon} [#06b6d4]{action}[/]"
+        return f"{browser_icon} [#06b6d4]{cls.escape_markup(action)}[/]"
 
     @classmethod
     def _format_url(cls, url: str) -> str:

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/cli/tool_components/proxy_renderer.py

@@ -64,7 +64,7 @@ class ViewRequestRenderer(BaseToolRenderer):
 
         part = args.get("part", "request")
 
-        header = f"👀 [bold #06b6d4]Viewing {part}[/]"
+        header = f"👀 [bold #06b6d4]Viewing {cls.escape_markup(part)}[/]"
 
         if result and isinstance(result, dict):
             if "content" in result:
@@ -107,7 +107,7 @@ class SendRequestRenderer(BaseToolRenderer):
         method = args.get("method", "GET")
         url = args.get("url", "")
 
-        header = f"📤 [bold #06b6d4]Sending {method}[/]"
+        header = f"📤 [bold #06b6d4]Sending {cls.escape_markup(method)}[/]"
 
         if result and isinstance(result, dict):
             status_code = result.get("status_code")

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/cli/tool_components/registry.py

@@ -54,7 +54,7 @@ def _render_default_tool_widget(tool_data: dict[str, Any]) -> Static:
 
     status_text = BaseToolRenderer.get_status_icon(status)
 
-    header = f"→ Using tool [bold blue]{tool_name}[/]"
+    header = f"→ Using tool [bold blue]{BaseToolRenderer.escape_markup(tool_name)}[/]"
     content_parts = [header]
 
     args_str = BaseToolRenderer.format_args(args)

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/cli/tool_components/reporting_renderer.py

@@ -27,7 +27,7 @@ class CreateVulnerabilityReportRenderer(BaseToolRenderer):
             if severity:
                 severity_color = cls._get_severity_color(severity.lower())
                 content_parts.append(
-                    f"  [dim]Severity: [{severity_color}]{severity.upper()}[/{severity_color}][/]"
+                    f"  [dim]Severity: [{severity_color}]{cls.escape_markup(severity.upper())}[/{severity_color}][/]"
                 )
 
             if content:

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/cli/tool_components/scan_info_renderer.py

@@ -28,12 +28,12 @@ class ScanStartInfoRenderer(BaseToolRenderer):
     @classmethod
     def _build_target_display(cls, target: dict[str, Any]) -> str:
         if target_url := target.get("target_url"):
-            return f"[bold #22c55e]{target_url}[/]"
+            return cls.escape_markup(str(target_url))
         if target_repo := target.get("target_repo"):
-            return f"[bold #22c55e]{target_repo}[/]"
+            return cls.escape_markup(str(target_repo))
         if target_path := target.get("target_path"):
-            return f"[bold #22c55e]{target_path}[/]"
-        return "[dim]unknown target[/dim]"
+            return cls.escape_markup(str(target_path))
+        return "unknown target"
 
 
 @register_tool_renderer
@@ -49,9 +49,11 @@ class SubagentStartInfoRenderer(BaseToolRenderer):
         name = args.get("name", "Unknown Agent")
         task = args.get("task", "")
 
-        content = f"🤖 Spawned subagent [bold #22c55e]{name}[/]"
+        name = cls.escape_markup(str(name))
+        content = f"🤖 Spawned subagent {name}"
         if task:
-            content += f"\n    Task: [dim]{task}[/dim]"
+            task = cls.escape_markup(str(task))
+            content += f"\n    Task: {task}"
 
         css_classes = cls.get_css_classes(status)
         return Static(content, classes=css_classes)

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/cli/tool_components/terminal_renderer.py

@@ -111,7 +111,7 @@ class TerminalRenderer(BaseToolRenderer):
         )
 
         if is_special:
-            return f"{terminal_icon} [#ef4444]{command}[/]"
+            return f"{terminal_icon} [#ef4444]{cls.escape_markup(command)}[/]"
 
         if is_input:
             formatted_command = cls._format_command_display(command)

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/cli/tracer.py

@@ -168,10 +168,14 @@ class Tracer:
             self.tool_executions[execution_id]["result"] = result
             self.tool_executions[execution_id]["completed_at"] = datetime.now(UTC).isoformat()
 
-    def update_agent_status(self, agent_id: str, status: str) -> None:
+    def update_agent_status(
+        self, agent_id: str, status: str, error_message: str | None = None
+    ) -> None:
         if agent_id in self.agents:
             self.agents[agent_id]["status"] = status
             self.agents[agent_id]["updated_at"] = datetime.now(UTC).isoformat()
+            if error_message:
+                self.agents[agent_id]["error_message"] = error_message
 
     def set_scan_config(self, config: dict[str, Any]) -> None:
         self.scan_config = config

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/llm/llm.py

@@ -28,9 +28,21 @@ api_key = os.getenv("LLM_API_KEY")
 if api_key:
     litellm.api_key = api_key
 
+api_base = (
+    os.getenv("LLM_API_BASE")
+    or os.getenv("OPENAI_API_BASE")
+    or os.getenv("LITELLM_BASE_URL")
+    or os.getenv("OLLAMA_API_BASE")
+)
+if api_base:
+    litellm.api_base = api_base
+
 
 class LLMRequestFailedError(Exception):
-    """Raised when LLM request fails after all retry attempts."""
+    def __init__(self, message: str, details: str | None = None):
+        super().__init__(message)
+        self.message = message
+        self.details = details
 
 
 MODELS_WITHOUT_STOP_WORDS = [
@@ -211,7 +223,7 @@ class LLM:
 
         return cached_messages
 
-    async def generate(
+    async def generate(  # noqa: PLR0912, PLR0915
         self,
         conversation_history: list[dict[str, Any]],
         scan_id: str | None = None,
@@ -255,8 +267,50 @@ class LLM:
                 tool_invocations=tool_invocations if tool_invocations else None,
             )
 
+        except litellm.RateLimitError as e:
+            raise LLMRequestFailedError("LLM request failed: Rate limit exceeded", str(e)) from e
+        except litellm.AuthenticationError as e:
+            raise LLMRequestFailedError("LLM request failed: Invalid API key", str(e)) from e
+        except litellm.NotFoundError as e:
+            raise LLMRequestFailedError("LLM request failed: Model not found", str(e)) from e
+        except litellm.ContextWindowExceededError as e:
+            raise LLMRequestFailedError("LLM request failed: Context too long", str(e)) from e
+        except litellm.ContentPolicyViolationError as e:
+            raise LLMRequestFailedError(
+                "LLM request failed: Content policy violation", str(e)
+            ) from e
+        except litellm.ServiceUnavailableError as e:
+            raise LLMRequestFailedError("LLM request failed: Service unavailable", str(e)) from e
+        except litellm.Timeout as e:
+            raise LLMRequestFailedError("LLM request failed: Request timed out", str(e)) from e
+        except litellm.UnprocessableEntityError as e:
+            raise LLMRequestFailedError("LLM request failed: Unprocessable entity", str(e)) from e
+        except litellm.InternalServerError as e:
+            raise LLMRequestFailedError("LLM request failed: Internal server error", str(e)) from e
+        except litellm.APIConnectionError as e:
+            raise LLMRequestFailedError("LLM request failed: Connection error", str(e)) from e
+        except litellm.UnsupportedParamsError as e:
+            raise LLMRequestFailedError("LLM request failed: Unsupported parameters", str(e)) from e
+        except litellm.BudgetExceededError as e:
+            raise LLMRequestFailedError("LLM request failed: Budget exceeded", str(e)) from e
+        except litellm.APIResponseValidationError as e:
+            raise LLMRequestFailedError(
+                "LLM request failed: Response validation error", str(e)
+            ) from e
+        except litellm.JSONSchemaValidationError as e:
+            raise LLMRequestFailedError(
+                "LLM request failed: JSON schema validation error", str(e)
+            ) from e
+        except litellm.InvalidRequestError as e:
+            raise LLMRequestFailedError("LLM request failed: Invalid request", str(e)) from e
+        except litellm.BadRequestError as e:
+            raise LLMRequestFailedError("LLM request failed: Bad request", str(e)) from e
+        except litellm.APIError as e:
+            raise LLMRequestFailedError("LLM request failed: API error", str(e)) from e
+        except litellm.OpenAIError as e:
+            raise LLMRequestFailedError("LLM request failed: OpenAI error", str(e)) from e
         except Exception as e:
-            raise LLMRequestFailedError("LLM request failed after all retry attempts") from e
+            raise LLMRequestFailedError(f"LLM request failed: {type(e).__name__}", str(e)) from e
 
     @property
     def usage_stats(self) -> dict[str, dict[str, int | float]]:

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/llm/request_queue.py

@@ -4,13 +4,27 @@ import threading
 import time
 from typing import Any
 
+import litellm
 from litellm import ModelResponse, completion
-from tenacity import retry, stop_after_attempt, wait_exponential
+from tenacity import retry, retry_if_exception, stop_after_attempt, wait_exponential
 
 
 logger = logging.getLogger(__name__)
 
 
+def should_retry_exception(exception: Exception) -> bool:
+    status_code = None
+
+    if hasattr(exception, "status_code"):
+        status_code = exception.status_code
+    elif hasattr(exception, "response") and hasattr(exception.response, "status_code"):
+        status_code = exception.response.status_code
+
+    if status_code is not None:
+        return bool(litellm._should_retry(status_code))
+    return True
+
+
 class LLMRequestQueue:
     def __init__(self, max_concurrent: int = 6, delay_between_requests: float = 1.0):
         self.max_concurrent = max_concurrent
@@ -40,6 +54,7 @@ class LLMRequestQueue:
     @retry(  # type: ignore[misc]
         stop=stop_after_attempt(5),
         wait=wait_exponential(multiplier=2, min=1, max=30),
+        retry=retry_if_exception(should_retry_exception),
         reraise=True,
     )
     async def _reliable_request(self, completion_args: dict[str, Any]) -> ModelResponse:

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/runtime/docker_runtime.py

@@ -320,7 +320,7 @@ class DockerRuntime(AbstractRuntime):
         import httpx
 
         try:
-            async with httpx.AsyncClient() as client:
+            async with httpx.AsyncClient(trust_env=False) as client:
                 response = await client.post(
                     f"{api_url}/register_agent",
                     params={"agent_id": agent_id},
@@ -337,7 +337,7 @@ class DockerRuntime(AbstractRuntime):
             container = self.client.containers.get(container_id)
             container.reload()
 
-            host = "localhost"
+            host = "127.0.0.1"
             if "DOCKER_HOST" in os.environ:
                 docker_host = os.environ["DOCKER_HOST"]
                 if "://" in docker_host:

{strix_agent-0.1.16 → strix_agent-0.1.18}/strix/tools/executor.py

@@ -62,7 +62,7 @@ async def _execute_tool_in_sandbox(tool_name: str, agent_state: Any, **kwargs: A
         "Content-Type": "application/json",
     }
 
-    async with httpx.AsyncClient() as client:
+    async with httpx.AsyncClient(trust_env=False) as client:
         try:
             response = await client.post(
                 request_url, json=request_data, headers=headers, timeout=None