Prepare chunk text for indexing with contextual enrichment Args: chunk: Conversation chunk

Prepare chunk text for indexing with contextual enrichment Args: chunk: Conversation chunk

Process a user request Args: request: User request force_tools: Force tool usage regardless of setti

Add a document to the index

Add a step to the workflow

Call a tool and return the result.

Send a message using Ollama's native tool calling Args: message: User message use_tools: Whether to

Send a message to the model and handle tool calls in a ReAct loop Args: message: User message use_to

Clear all memories for this user - useful for testing

清空对话历史

Clear conversation history

Render a block of monospace code lines with background.

Perform complex problem reasoning with step-by-step thinking. Args: problem: Problem statement context: Optional context

Connect to the MCP server.

Create necessary directories if they don't exist

Create a new Excel workbook. Args: file_path: Path for new workbook Returns: Dictionary with result

Create test conversation chunks with substantial content

Create test conversation chunks with substantial content

Delete a file or directory. Args: file_path: Path to delete recursive: For directories, delete recur

Demonstrate the experience agent.

Demonstrate knowledge base indexing.

Demonstrate trajectory summarization.

Disconnect from the server.

Edit a file by searching and replacing content. Args: path: File path search: Text to search for

Ensure sample PDFs exist, create them if they don't Returns: bool: True if PDFs are available

Execute a shell command in current directory. Args: command: Command to execute timeout: Timeout in

Calculate expected win probability for model A against model B. Uses logistic function: P(A wins) = 1 / (1 + 10^((R_B - R_A)

Extract text from PDF file. Args: file_path: Path to PDF file page_range: Optional page range (e.g., "1-5" or "1,3,5")

Generate Python code based on task description. Args: task_description: Description of what the code should do requireme

Get the current board state. Returns: Dictionary with complete board state

Retrieve synonyms for a PubChem compound. Args: cid: PubChem Compound ID max_synonyms: Maximum number of synonyms (1-100

Get memory cards as formatted string for LLM context. Args: max_cards: Maximum number of cards to include (None

Get a document by ID.

Retrieve a document by ID (can be internal or external)

Get all legal moves in the current position. Returns: Dictionary with legal moves in UCI and SAN formats

Get the history of moves played in the current game. Returns: Dictionary with move history

Get document IDs containing the term

Get sample tasks for testing Returns: list: List of sample task dictionaries

Get the current number of items in the index.

Get statistics about the knowledge base. Returns: Dictionary containing knowledge base statistics

Get statistics about the knowledge base. Returns: Dictionary with statistics

Get tool instance by name

Create a GitHub Pull Request. Args: repo_name: Repository name (format: owner/repo) title: PR title

Guard and validate a proposed action or reasoning. Args: proposed_action: The action being proposed context: Context inf

Index a single file. Args: file_path: Path to the file doc_id: Optional document ID

Index a single file. Args: file_path: Path to the file doc_id: Optional document ID

List all available test cases.

Load configuration from environment variables.

Load tools from MCP servers This must be called in an async context after agent initialization.

Load test cases from the evaluation framework Args: category: Optional category filter (layer1, layer2, layer3)

Log a banner message

Log errors with context.

Print a colored section header

Print a formatted section header

Process a file based on its extension.

Process multimodal content based on extraction mode

Read contents of a file Args: file_path: Path to the file relative to root directory offset: Line nu

Reset the agent's trajectory and conversation history

Reset the agent's state

Start a new conversation session

Search the knowledge graph. Args: query: Search query top_k: Number of results to return

Search the index for relevant chunks using retrieval pipeline Args: query: Search query top_k: Numbe

Combine web search with code analysis This method demonstrates using both native tools together: 1. Search for infor

Search for files matching a pattern. Args: pattern: Glob pattern (e.g., "*.py", "test_*.txt") direct

Find structurally similar compounds. Args: cid: Reference compound CID similarity_threshold: Minimum similarity (0.0-1.0

Send an email notification. Args: to_email: Recipient email address subject: Email subject body: Email body content

Send a Telegram message. Args: message: Message text to send chat_id: Optional Telegram chat ID (uses default if not provided

Check if memory processing should be triggered based on conversation count Returns: True if processing should oc

Submit an agent's response and get evaluation result. Args: test_id: The test case ID agent_response

Convert to dictionary

Convert to dictionary format

Convert to dictionary for storage

Serialize workflow to JSON string

Tokenize text into words, numbers, and codes. Handles: - Words (preserving case for acronyms) - Numbers (404, 500, 3.

Train the agent for multiple episodes. Args: num_episodes: Number of episodes to train verbose: Whet

Update Q-value using the Q-learning update rule. Q(s,a) <- Q(s,a) + α[r + γ max Q(s',a') - Q(s,a)]

Validate required configuration Returns: True if configuration is valid

Verify code syntax and provide feedback. Args: code: The code to verify language: Programming langua

Build context for the LLM including task description and past experiences. This is the key to in-context learning.

Simple check if answer meets criteria

Clean extracted PDF text.

Convert tool registry to Ollama's expected format

Create a basic chunk dictionary

Create a chunk with metadata

Create a chunk with metadata

Display the result of a request Args: result: Result dictionary from agent

Display a single evaluation result.