lm-deluge 0.0.75__tar.gz → 0.0.76__tar.gz

{lm_deluge-0.0.75/src/lm_deluge.egg-info → lm_deluge-0.0.76}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: lm_deluge
-Version: 0.0.75
+Version: 0.0.76
 Summary: Python utility for using LLM API models.
 Author-email: Benjamin Anderson <ben@trytaylor.ai>
 Requires-Python: >=3.10

{lm_deluge-0.0.75 → lm_deluge-0.0.76}/pyproject.toml

@@ -3,7 +3,7 @@ requires = ["setuptools", "wheel"]
 
 [project]
 name = "lm_deluge"
-version = "0.0.75"
+version = "0.0.76"
 authors = [{ name = "Benjamin Anderson", email = "ben@trytaylor.ai" }]
 description = "Python utility for using LLM API models."
 readme = "README.md"

{lm_deluge-0.0.75 → lm_deluge-0.0.76}/src/lm_deluge/client.py

@@ -1,4 +1,5 @@
 import asyncio
+from dataclasses import dataclass
 from typing import (
     Any,
     AsyncGenerator,
@@ -37,6 +38,14 @@ from .request_context import RequestContext
 from .tracker import StatusTracker
 
 
+@dataclass
+class AgentLoopResponse:
+    """Wrapper for agent loop results to distinguish from single request results."""
+
+    conversation: Conversation
+    final_response: APIResponse
+
+
 # TODO: add optional max_input_tokens to client so we can reject long prompts to prevent abuse
 class _LLMClient(BaseModel):
     """
@@ -88,7 +97,9 @@ class _LLMClient(BaseModel):
     # Internal state for async task handling
     _next_task_id: int = PrivateAttr(default=0)
     _tasks: dict[int, asyncio.Task] = PrivateAttr(default_factory=dict)
-    _results: dict[int, APIResponse] = PrivateAttr(default_factory=dict)
+    _results: dict[int, APIResponse | AgentLoopResponse] = PrivateAttr(
+        default_factory=dict
+    )
     _tracker: StatusTracker | None = PrivateAttr(default=None)
     _capacity_lock: asyncio.Lock = PrivateAttr(default_factory=asyncio.Lock)
 
@@ -747,11 +758,11 @@ class _LLMClient(BaseModel):
     async def wait_for(self, task_id: int) -> APIResponse:
         task = self._tasks.get(task_id)
         if task:
-            return await task
-        res = self._results.get(task_id)
-        if res:
-            return res
+            result = await task
         else:
+            result = self._results.get(task_id)
+
+        if result is None:
             return APIResponse(
                 id=-1,
                 model_internal="",
@@ -762,6 +773,11 @@ class _LLMClient(BaseModel):
                 error_message="Task not found",
             )
 
+        assert isinstance(
+            result, APIResponse
+        ), f"Expected APIResponse, got {type(result)}. Use wait_for_agent_loop for agent loop tasks."
+        return result
+
     async def wait_for_all(
         self, task_ids: Sequence[int] | None = None
     ) -> list[APIResponse]:
@@ -797,6 +813,9 @@ class _LLMClient(BaseModel):
                 tid = tasks_map.pop(task)
                 task_result = self._results.get(tid, await task)
                 assert task_result
+                assert isinstance(
+                    task_result, APIResponse
+                ), f"Expected APIResponse, got {type(task_result)}. as_completed() only works with single requests, not agent loops."
                 yield tid, task_result
 
         while tasks_map:
@@ -807,6 +826,9 @@ class _LLMClient(BaseModel):
                 tid = tasks_map.pop(task)
                 task_result = self._results.get(tid, await task)
                 assert task_result
+                assert isinstance(
+                    task_result, APIResponse
+                ), f"Expected APIResponse, got {type(task_result)}. as_completed() only works with single requests, not agent loops."
                 yield tid, task_result
 
     async def stream(
@@ -828,24 +850,15 @@ class _LLMClient(BaseModel):
                     return self.postprocess(item)
                 return item
 
-    async def run_agent_loop(
+    async def _run_agent_loop_internal(
         self,
-        conversation: Prompt,
+        task_id: int,
+        conversation: Conversation,
         *,
         tools: list[Tool | dict | MCPServer] | None = None,
         max_rounds: int = 5,
-        show_progress: bool = False,
-    ) -> tuple[Conversation, APIResponse]:
-        """Run a simple agent loop until no more tool calls are returned.
-
-        The provided ``conversation`` will be mutated and returned alongside the
-        final ``APIResponse`` from the model. ``tools`` may include ``Tool``
-        instances or built‑in tool dictionaries.
-        """
-
-        if not isinstance(conversation, Conversation):
-            conversation = prompts_to_conversations([conversation])[0]
-            assert isinstance(conversation, Conversation)
+    ) -> AgentLoopResponse:
+        """Internal method to run agent loop and return wrapped result."""
 
         # Expand MCPServer objects to their constituent tools for tool execution
         expanded_tools: list[Tool] = []
@@ -898,7 +911,75 @@ class _LLMClient(BaseModel):
         if response is None:
             raise RuntimeError("model did not return a response")
 
-        return conversation, response
+        result = AgentLoopResponse(conversation=conversation, final_response=response)
+        self._results[task_id] = result
+        return result
+
+    def start_agent_loop_nowait(
+        self,
+        conversation: Prompt,
+        *,
+        tools: list[Tool | dict | MCPServer] | None = None,
+        max_rounds: int = 5,
+    ) -> int:
+        """Start an agent loop without waiting for it to complete.
+
+        Returns a task_id that can be used with wait_for_agent_loop().
+        """
+        if not isinstance(conversation, Conversation):
+            conversation = prompts_to_conversations([conversation])[0]
+            assert isinstance(conversation, Conversation)
+
+        task_id = self._next_task_id
+        self._next_task_id += 1
+
+        task = asyncio.create_task(
+            self._run_agent_loop_internal(
+                task_id, conversation, tools=tools, max_rounds=max_rounds
+            )
+        )
+        self._tasks[task_id] = task
+        return task_id
+
+    async def wait_for_agent_loop(
+        self, task_id: int
+    ) -> tuple[Conversation, APIResponse]:
+        """Wait for an agent loop task to complete.
+
+        Returns the conversation and final response from the agent loop.
+        """
+        task = self._tasks.get(task_id)
+        if task:
+            result = await task
+        else:
+            result = self._results.get(task_id)
+
+        if result is None:
+            raise RuntimeError(f"Agent loop task {task_id} not found")
+
+        assert isinstance(
+            result, AgentLoopResponse
+        ), f"Expected AgentLoopResponse, got {type(result)}"
+        return result.conversation, result.final_response
+
+    async def run_agent_loop(
+        self,
+        conversation: Prompt,
+        *,
+        tools: list[Tool | dict | MCPServer] | None = None,
+        max_rounds: int = 5,
+        show_progress: bool = False,
+    ) -> tuple[Conversation, APIResponse]:
+        """Run a simple agent loop until no more tool calls are returned.
+
+        The provided ``conversation`` will be mutated and returned alongside the
+        final ``APIResponse`` from the model. ``tools`` may include ``Tool``
+        instances or built‑in tool dictionaries.
+        """
+        task_id = self.start_agent_loop_nowait(
+            conversation, tools=tools, max_rounds=max_rounds
+        )
+        return await self.wait_for_agent_loop(task_id)
 
     def run_agent_loop_sync(
         self,

{lm_deluge-0.0.75 → lm_deluge-0.0.76}/src/lm_deluge/llm_tools/__init__.py

@@ -1,11 +1,18 @@
 from .extract import extract, extract_async
-from .translate import translate, translate_async
 from .score import score_llm
+from .subagents import SubAgentManager
+from .todos import TodoItem, TodoManager, TodoPriority, TodoStatus
+from .translate import translate, translate_async
 
 __all__ = [
     "extract",
     "extract_async",
+    "TodoItem",
+    "TodoManager",
+    "TodoPriority",
+    "TodoStatus",
     "translate",
     "translate_async",
     "score_llm",
+    "SubAgentManager",
 ]

lm_deluge-0.0.76/src/lm_deluge/llm_tools/subagents.py

@@ -0,0 +1,233 @@
+from lm_deluge.api_requests.base import APIResponse
+from lm_deluge.client import AgentLoopResponse, _LLMClient
+from lm_deluge.prompt import Conversation, prompts_to_conversations
+from lm_deluge.tool import Tool
+
+
+class SubAgentManager:
+    """Manages subagent tasks that can be spawned by a main LLM via tool calls.
+
+    The SubAgentManager exposes tools that allow a main LLM to delegate subtasks
+    to specialized or cheaper subagent models, saving context and improving efficiency.
+
+    Example:
+        >>> manager = SubAgentManager(
+        ...     client=LLMClient("gpt-4o-mini"),  # Subagent model
+        ...     tools=[search_tool, calculator_tool]  # Tools available to subagents
+        ... )
+        >>> main_client = LLMClient("gpt-4o")  # More expensive main model
+        >>> conv = Conversation.user("Research AI and calculate market size")
+        >>> # Main model can now call manager tools to spawn subagents
+        >>> conv, resp = await main_client.run_agent_loop(
+        ...     conv,
+        ...     tools=manager.get_tools()
+        ... )
+    """
+
+    def __init__(
+        self,
+        client: _LLMClient,
+        tools: list[Tool] | None = None,
+        max_rounds: int = 5,
+    ):
+        """Initialize the SubAgentManager.
+
+        Args:
+            client: LLMClient to use for subagent tasks
+            tools: Tools available to subagents (optional)
+            max_rounds: Maximum rounds for each subagent's agent loop
+        """
+        self.client = client
+        self.tools = tools or []
+        self.max_rounds = max_rounds
+        self.subagents: dict[int, dict] = {}
+
+    async def _start_subagent(self, task: str) -> int:
+        """Start a subagent with the given task.
+
+        Args:
+            task: The task description for the subagent
+
+        Returns:
+            Subagent task ID
+        """
+        conversation = prompts_to_conversations([task])[0]
+        assert isinstance(conversation, Conversation)
+
+        # Use agent loop nowait API to start the subagent
+        task_id = self.client.start_agent_loop_nowait(
+            conversation,
+            tools=self.tools,  # type: ignore
+            max_rounds=self.max_rounds,
+        )
+
+        # Track the subagent
+        self.subagents[task_id] = {
+            "status": "running",
+            "conversation": None,
+            "response": None,
+            "error": None,
+        }
+
+        return task_id
+
+    def _finalize_subagent_result(
+        self, agent_id: int, result: AgentLoopResponse
+    ) -> str:
+        """Update subagent tracking state from a finished agent loop."""
+        agent = self.subagents[agent_id]
+        agent["conversation"] = result.conversation
+        agent["response"] = result.final_response
+
+        if result.final_response.is_error:
+            agent["status"] = "error"
+            agent["error"] = result.final_response.error_message
+            return f"Error: {agent['error']}"
+
+        agent["status"] = "finished"
+        return result.final_response.completion or "Subagent finished with no output"
+
+    async def _check_subagent(self, agent_id: int) -> str:
+        """Check the status of a subagent.
+
+        Args:
+            agent_id: The subagent task ID
+
+        Returns:
+            Status string describing the subagent's state
+        """
+        if agent_id not in self.subagents:
+            return f"Error: Subagent {agent_id} not found"
+
+        agent = self.subagents[agent_id]
+        status = agent["status"]
+
+        if status == "finished":
+            response: APIResponse = agent["response"]
+            return response.completion or "Subagent finished with no output"
+        elif status == "error":
+            return f"Error: {agent['error']}"
+        else:
+            # Try to check if it's done
+            try:
+                # Check if the task exists in client's results
+                stored_result = self.client._results.get(agent_id)
+                if isinstance(stored_result, AgentLoopResponse):
+                    return self._finalize_subagent_result(agent_id, stored_result)
+
+                task = self.client._tasks.get(agent_id)
+                if task and task.done():
+                    try:
+                        task_result = task.result()
+                    except Exception as e:
+                        agent["status"] = "error"
+                        agent["error"] = str(e)
+                        return f"Error: {agent['error']}"
+
+                    if isinstance(task_result, AgentLoopResponse):
+                        return self._finalize_subagent_result(agent_id, task_result)
+
+                    agent["status"] = "error"
+                    agent["error"] = (
+                        f"Unexpected task result type: {type(task_result).__name__}"
+                    )
+                    return f"Error: {agent['error']}"
+
+                # Still running
+                return f"Subagent {agent_id} is still running. Call this tool again to check status."
+            except Exception as e:
+                agent["status"] = "error"
+                agent["error"] = str(e)
+                return f"Error checking subagent: {e}"
+
+    async def _wait_for_subagent(self, agent_id: int) -> str:
+        """Wait for a subagent to complete and return its output.
+
+        Args:
+            agent_id: The subagent task ID
+
+        Returns:
+            The subagent's final output
+        """
+        if agent_id not in self.subagents:
+            return f"Error: Subagent {agent_id} not found"
+
+        try:
+            # Use the wait_for_agent_loop API
+            conversation, response = await self.client.wait_for_agent_loop(agent_id)
+
+            agent = self.subagents[agent_id]
+            agent["conversation"] = conversation
+            agent["response"] = response
+
+            if response.is_error:
+                agent["status"] = "error"
+                agent["error"] = response.error_message
+                return f"Error: {response.error_message}"
+            else:
+                agent["status"] = "finished"
+                return response.completion or "Subagent finished with no output"
+        except Exception as e:
+            agent = self.subagents[agent_id]
+            agent["status"] = "error"
+            agent["error"] = str(e)
+            return f"Error waiting for subagent: {e}"
+
+    def get_tools(self) -> list[Tool]:
+        """Get the tools that allow a main LLM to control subagents.
+
+        Returns:
+            List of Tool objects for starting, checking, and waiting for subagents
+        """
+        start_tool = Tool(
+            name="start_subagent",
+            description=(
+                "Start a subagent to work on a subtask independently. "
+                "Use this to delegate complex subtasks or when you need to save context. "
+                "Returns the subagent's task ID which can be used to check its status."
+            ),
+            run=self._start_subagent,
+            parameters={
+                "task": {
+                    "type": "string",
+                    "description": "The task description for the subagent to work on",
+                }
+            },
+            required=["task"],
+        )
+
+        check_tool = Tool(
+            name="check_subagent",
+            description=(
+                "Check the status and output of a running subagent. "
+                "If the subagent is still running, you'll be told to check again later. "
+                "If finished, returns the subagent's final output."
+            ),
+            run=self._check_subagent,
+            parameters={
+                "agent_id": {
+                    "type": "integer",
+                    "description": "The task ID of the subagent to check",
+                }
+            },
+            required=["agent_id"],
+        )
+
+        wait_tool = Tool(
+            name="wait_for_subagent",
+            description=(
+                "Wait for a subagent to complete and return its output. "
+                "This will block until the subagent finishes. "
+                "Use check_subagent if you want to do other work while waiting."
+            ),
+            run=self._wait_for_subagent,
+            parameters={
+                "agent_id": {
+                    "type": "integer",
+                    "description": "The task ID of the subagent to wait for",
+                }
+            },
+            required=["agent_id"],
+        )
+
+        return [start_tool, check_tool, wait_tool]

lm_deluge-0.0.76/src/lm_deluge/llm_tools/todos.py

@@ -0,0 +1,342 @@
+# Adapted from https://github.com/sst/opencode - MIT License
+# MIT License
+# Copyright (c) 2025 opencode
+
+# Permission is hereby granted, free of charge, to any person obtaining a copy
+# of this software and associated documentation files (the "Software"), to deal
+# in the Software without restriction, including without limitation the rights
+# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+# copies of the Software, and to permit persons to whom the Software is
+# furnished to do so, subject to the following conditions:
+
+# The above copyright notice and this permission notice shall be included in all
+# copies or substantial portions of the Software.
+
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+# SOFTWARE.
+from __future__ import annotations
+
+import json
+import uuid
+from typing import Any, Literal, Sequence
+
+from pydantic import BaseModel, Field, field_validator
+
+from ..tool import Tool
+
+TODO_WRITE_DESCRIPTION = """Use this tool to create and manage a structured task list for your current coding session. This helps you track progress, organize complex tasks, and demonstrate thoroughness to the user.
+It also helps the user understand the progress of the task and overall progress of their requests.
+
+## When to Use This Tool
+Use this tool proactively in these scenarios:
+
+1. Complex multi-step tasks - When a task requires 3 or more distinct steps or actions
+2. Non-trivial and complex tasks - Tasks that require careful planning or multiple operations
+3. User explicitly requests todo list - When the user directly asks you to use the todo list
+4. User provides multiple tasks - When users provide a list of things to be done (numbered or comma-separated)
+5. After receiving new instructions - Immediately capture user requirements as todos. Feel free to edit the todo list based on new information.
+6. After completing a task - Mark it complete and add any new follow-up tasks
+7. When you start working on a new task, mark the todo as in_progress. Ideally you should only have one todo as in_progress at a time. Complete existing tasks before starting new ones.
+
+## When NOT to Use This Tool
+
+Skip using this tool when:
+1. There is only a single, straightforward task
+2. The task is trivial and tracking it provides no organizational benefit
+3. The task can be completed in less than 3 trivial steps
+4. The task is purely conversational or informational
+
+NOTE that you should not use this tool if there is only one trivial task to do. In this case you are better off just doing the task directly.
+
+## Examples of When to Use the Todo List
+
+<example>
+User: I want to add a dark mode toggle to the application settings. Make sure you run the tests and build when you're done!
+Assistant: I'll help add a dark mode toggle to your application settings. Let me create a todo list to track this implementation.
+*Creates todo list with the following items:*
+1. Create dark mode toggle component in Settings page
+2. Add dark mode state management (context/store)
+3. Implement CSS-in-JS styles for dark theme
+4. Update existing components to support theme switching
+5. Run tests and build process, addressing any failures or errors that occur
+*Begins working on the first task*
+
+<reasoning>
+The assistant used the todo list because:
+1. Adding dark mode is a multi-step feature requiring UI, state management, and styling changes
+2. The user explicitly requested tests and build be run afterward
+3. The assistant inferred that tests and build need to pass by adding "Ensure tests and build succeed" as the final task
+</reasoning>
+</example>
+
+<example>
+User: Help me rename the function getCwd to getCurrentWorkingDirectory across my project
+Assistant: Let me first search through your codebase to find all occurrences of 'getCwd'.
+*Uses grep or search tools to locate all instances of getCwd in the codebase*
+Assistant: I've found 15 instances of 'getCwd' across 8 different files. Let me create a todo list to track these changes.
+*Creates todo list with specific items for each file that needs updating*
+
+<reasoning>
+The assistant used the todo list because:
+1. First, the assistant searched to understand the scope of the task
+2. Upon finding multiple occurrences across different files, it determined this was a complex task with multiple steps
+3. The todo list helps ensure every instance is tracked and updated systematically
+4. This approach prevents missing any occurrences and maintains code consistency
+</reasoning>
+</example>
+
+
+<example>
+User: I need to implement these features for my e-commerce site: user registration, product catalog, shopping cart, and checkout flow.
+Assistant: I'll help implement these features. First, let's add all the features to the todo list.
+*Creates a todo list breaking down each feature into specific tasks based on the project architecture*
+Assistant: Let's start with user registration. This will involve creating database models, API endpoints, and frontend forms that integrate with your existing authentication system.
+
+<reasoning>
+The assistant used the todo list because:
+1. The user provided multiple complex features to implement in a comma separated list
+2. The todo list helps organize these large features into manageable tasks
+3. This approach allows for tracking progress across the entire implementation
+</reasoning>
+</example>
+
+<example>
+User: Can you help optimize my React application? It's rendering slowly and has performance issues.</user>
+Assistant: I'll help optimize your React application. First, let me examine your codebase to identify potential performance bottlenecks.
+*Reviews component structure, render patterns, state management, and data fetching*
+Assistant: After analyzing your codebase, I've identified several performance issues. Let me create a todo list to track our optimization efforts.
+*Creates todo list with items like: 1) Implement memoization for expensive calculations in ProductList, 2) Add virtualization for long lists in Dashboard, 3) Optimize image loading in Gallery component, 4) Fix state update loops in ShoppingCart, 5) Review bundle size and implement code splitting*
+Let's start by implementing memoization for the expensive calculations in your ProductList component.</assistant>
+
+<reasoning>
+The assistant used the todo list because:
+1. First, the assistant examined the codebase to identify specific performance issues
+2. Based on this analysis, it identified multiple optimization opportunities
+3. Performance optimization is a non-trivial task requiring multiple steps
+4. The todo list helps methodically track improvements across different components
+5. This systematic approach ensures all performance bottlenecks are addressed
+</reasoning>
+</example>
+
+## Examples of When NOT to Use the Todo List
+
+<example>
+User: How do I print 'Hello World' in Python?
+Assistant: In Python, you can print "Hello World" with this simple code:
+
+python
+print("Hello World")
+
+This will output the text "Hello World" to the console when executed.</assistant>
+
+<reasoning>
+The assistant did not use the todo list because this is a single, trivial task that can be completed in one step. There's no need to track multiple tasks or steps for such a straightforward request.
+</reasoning>
+</example>
+
+<example>
+User: What does the git status command do?
+Assistant: The git status command shows the current state of your working directory and staging area. It displays which changes have been staged, which haven't, and which files aren't being tracked by Git. This is a fundamental command for understanding the current state of your repository before making commits.
+
+<reasoning>
+The assistant did not use the todo list because this is an informational request with no actual coding task to complete. The user is simply asking for an explanation, not for the assistant to perform multiple steps or tasks.
+</reasoning>
+</example>
+
+<example>
+User: Can you add a comment to the calculateTotal function to explain what it does?
+Assistant: Sure, let me add a comment to the calculateTotal function to explain what it does.
+* Uses the Edit tool to add a comment to the calculateTotal function *
+
+<reasoning>
+The assistant did not use the todo list because this is a single, straightforward task confined to one location in the code. Adding a comment doesn't require tracking multiple steps or systematic organization.
+</reasoning>
+</example>
+
+<example>
+User: Run npm install for me and tell me what happens.
+Assistant: I'll run the npm install command for you.
+
+*Executes: npm install*
+
+The command completed successfully. Here's the output:
+[Output of npm install command]
+
+All dependencies have been installed according to your package.json file.
+
+<reasoning>
+The assistant did not use the todo list because this is a single command execution with immediate results. There are no multiple steps to track or organize, making the todo list unnecessary for this straightforward task.
+</reasoning>
+</example>
+
+## Task States and Management
+
+1. **Task States**: Use these states to track progress:
+   - pending: Task not yet started
+   - in_progress: Currently working on (limit to ONE task at a time)
+   - completed: Task finished successfully
+   - cancelled: Task no longer needed
+
+2. **Task Management**:
+   - Update task status in real-time as you work
+   - Mark tasks complete IMMEDIATELY after finishing (don't batch completions)
+   - Only have ONE task in_progress at any time
+   - Complete current tasks before starting new ones
+   - Cancel tasks that become irrelevant
+
+3. **Task Breakdown**:
+   - Create specific, actionable items
+   - Break complex tasks into smaller, manageable steps
+   - Use clear, descriptive task names
+
+When in doubt, use this tool. Being proactive with task management demonstrates attentiveness and ensures you complete all requirements successfully.
+"""
+
+TODO_READ_DESCRIPTION = "Use this tool to read your todo list"
+
+TodoStatus = Literal["pending", "in_progress", "completed", "cancelled"]
+TodoPriority = Literal["high", "medium", "low"]
+
+
+class TodoItem(BaseModel):
+    """Structured representation of a single todo entry."""
+
+    content: str = Field(description="Brief description of the task")
+    status: TodoStatus = Field(
+        default="pending",
+        description="Current status of the task: pending, in_progress, completed, cancelled",
+    )
+    priority: TodoPriority = Field(
+        default="medium",
+        description="Priority level of the task: high, medium, low",
+    )
+    id: str = Field(
+        default_factory=lambda: uuid.uuid4().hex,
+        description="Unique identifier for the todo item",
+    )
+
+    @field_validator("status", "priority", mode="before")
+    @classmethod
+    def _normalize_lower(cls, value: Any) -> Any:
+        if isinstance(value, str):
+            return value.strip().lower()
+        return value
+
+    def is_active(self) -> bool:
+        return self.status not in {"completed", "cancelled"}
+
+
+TodoLike = TodoItem | dict[str, Any]
+
+
+class TodoManager:
+    """Stateful todo scratchpad that exposes read/write tools."""
+
+    def __init__(
+        self,
+        todos: Sequence[TodoLike] | None = None,
+        *,
+        write_tool_name: str = "todowrite",
+        read_tool_name: str = "todoread",
+    ):
+        self.write_tool_name = write_tool_name
+        self.read_tool_name = read_tool_name
+        self._todos: list[TodoItem] = []
+        self._tools: list[Tool] | None = None
+
+        if todos:
+            self._todos = [self._coerce(todo) for todo in todos]
+
+    def _coerce(self, todo: TodoLike) -> TodoItem:
+        if isinstance(todo, TodoItem):
+            return todo
+        if isinstance(todo, dict):
+            return TodoItem(**todo)
+        raise TypeError("Todos must be TodoItem instances or dictionaries")
+
+    def _serialize(self) -> list[dict[str, Any]]:
+        return [todo.model_dump() for todo in self._todos]
+
+    def _pending_count(self) -> int:
+        return sum(1 for todo in self._todos if todo.is_active())
+
+    def _format_output(self) -> str:
+        payload = {
+            "title": f"{self._pending_count()} todos",
+            "todos": self._serialize(),
+        }
+        return json.dumps(payload, indent=2)
+
+    def _write_tool(self, todos: list[dict[str, Any]]) -> str:
+        self._todos = [self._coerce(todo) for todo in todos]
+        return self._format_output()
+
+    def _read_tool(self) -> str:
+        return self._format_output()
+
+    def get_todos(self) -> list[TodoItem]:
+        """Return a copy of the current todo list."""
+        return list(self._todos)
+
+    def get_tools(self) -> list[Tool]:
+        """Return Tool instances bound to this manager's state."""
+        if self._tools is not None:
+            return self._tools
+
+        todo_definition = {
+            "type": "object",
+            "properties": {
+                "content": {
+                    "type": "string",
+                    "description": "Brief description of the task",
+                },
+                "status": {
+                    "type": "string",
+                    "description": "Current status of the task: pending, in_progress, completed, cancelled",
+                    "enum": ["pending", "in_progress", "completed", "cancelled"],
+                },
+                "priority": {
+                    "type": "string",
+                    "description": "Priority level of the task: high, medium, low",
+                    "enum": ["high", "medium", "low"],
+                },
+                "id": {
+                    "type": "string",
+                    "description": "Unique identifier for the todo item",
+                },
+            },
+            "required": ["content", "status", "priority", "id"],
+        }
+
+        write_tool = Tool(
+            name=self.write_tool_name,
+            description=TODO_WRITE_DESCRIPTION,
+            parameters={
+                "todos": {
+                    "type": "array",
+                    "description": "The updated todo list",
+                    "items": {"$ref": "#/$defs/Todo"},
+                }
+            },
+            required=["todos"],
+            definitions={"Todo": todo_definition},
+            run=self._write_tool,
+        )
+
+        read_tool = Tool(
+            name=self.read_tool_name,
+            description=TODO_READ_DESCRIPTION,
+            parameters={},
+            run=self._read_tool,
+        )
+
+        self._tools = [write_tool, read_tool]
+        return self._tools
+
+
+__all__ = ["TodoManager", "TodoItem", "TodoStatus", "TodoPriority"]

{lm_deluge-0.0.75 → lm_deluge-0.0.76/src/lm_deluge.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: lm_deluge
-Version: 0.0.75
+Version: 0.0.76
 Summary: Python utility for using LLM API models.
 Author-email: Benjamin Anderson <ben@trytaylor.ai>
 Requires-Python: >=3.10

{lm_deluge-0.0.75 → lm_deluge-0.0.76}/src/lm_deluge.egg-info/SOURCES.txt

@@ -48,9 +48,12 @@ src/lm_deluge/built_in_tools/anthropic/editor.py
 src/lm_deluge/llm_tools/__init__.py
 src/lm_deluge/llm_tools/classify.py
 src/lm_deluge/llm_tools/extract.py
+src/lm_deluge/llm_tools/filesystem.py
 src/lm_deluge/llm_tools/locate.py
 src/lm_deluge/llm_tools/ocr.py
 src/lm_deluge/llm_tools/score.py
+src/lm_deluge/llm_tools/subagents.py
+src/lm_deluge/llm_tools/todos.py
 src/lm_deluge/llm_tools/translate.py
 src/lm_deluge/models/__init__.py
 src/lm_deluge/models/anthropic.py