Liveness/readiness probe. Unauthenticated so kubelet can call it.

Initialize the ticket queue with generated tickets

Startup: warm the standby pool and start the idle reaper. Shutdown: stop the reaper and close Redis.

List all custom skills in the workspace. Args: client: Anthropic client instance with Skills beta Returns: List of skil

List all versions of a skill. Args: client: Anthropic client instance skill_id: ID of the skill Returns: List o

Calls the model with the given prompt and returns the response. Args: prompt (str): The user prompt to send to the model. sy

Load brand guidelines from JSON file Args: filepath: Path to JSON file containing brand guidelines Returns: BrandGuidel

Build an in-memory ZIP of a small repository with a stale doc. The repo is about 37 files laid out as microservices under `services/`, but `A

Mark ticket as processed and ready for team review. Args: ticket_id: The ticket ID Returns: Confirmation and summary of

Merge configuration with overrides. BUG: The default dict is shared across all calls!

Merge configs properly.

Fixture providing parsed notebook cells.

Fixture providing loaded notebook data.

Fixture providing the notebook execution timeout.

Perform comprehensive ratio analysis with interpretations. Args: ratios: Calculated financial ratios industry: Industry sect

Standby pool size — useful for monitoring.

Forward a turn to the session's agent pod and stream SSE back. Same path and shape as ``hosting/server.py`` (Tier 1/2), so client code writte

Print activity with enhanced subagent visibility. Shows: - Main agent tool usage with 🤖 - Subagent invocations with 🚀 and subagent n

Print activity to console

Print context management information from response. Args: response: API response to analyze Returns: Tuple of (context_

Print a formatted summary of file download results. Args: results: List of download results from download_all_files() Example:

Print the final agent result and cost information. Args: messages: List of conversation messages model: Optional model ident

Print a formatted summary of a skill. Args: skill_info: Dictionary with skill information

Process data with optional filters. BUG: filters list is shared across calls!

Process a single file.

Fixture providing the project root directory.

Add custom command-line options for notebook testing.

Reset failure tracking at start of test run.

Generate parameterized tests for notebooks.

Collect failures grouped by notebook.

Print a summary of failures grouped by notebook.

Fixture providing the notebook registry.

Detect content type and render to HTML. This is the main content routing function that dispatches to the appropriate renderer based on c

Reset processor state. BUG: No locking - if called during processing, causes corruption.

Route a ticket to the appropriate support team. Args: ticket_id: The ticket ID team: Team name (billing-team, tech-support,

Run a complete conversation loop until Claude stops using tools. Args: client: Anthropic client instance model: Model to use

Analyze multiple scenarios with different variable combinations. Args: scenarios: Dictionary of scenarios with variable

Search the knowledge base for relevant information. Args: category: The category to search (billing, technical, account) que

Send a query to the Chief of Staff agent with all features integrated. Args: prompt: The query to send (can include slash commands l

Send a query to the observability agent with MCP server support. Args: prompt: The query to send activity_handler: Callback

Perform two-way sensitivity analysis on valuation. Args: variable1: First variable to test ('wacc', 'growth', 'margin')

Create temporary directory for each test.

Set the priority level for a ticket. Args: ticket_id: The ticket ID priority: Priority level Returns: Confirmat

Show all captured responses.

Simple helper to visualize a single response. Args: response: Claude API response (Message object or dict)

Start a new conversation session (memory persists).

Stream a session's events and print until the agent finishes its turn. This is the streaming pattern from notebook 00, lifted into a helper s

One-line-per-block summary of a Messages API response, for transcript display.

Clean up temporary directory after each test.

Test that all code cells have been executed.

Test that API keys are loaded from environment variables.

Test that cells were executed in sequential order.

Test clearing all memory.

Test creating a file.

Test creating a file in a subdirectory (auto-creates dirs).

Test that create overwrites existing files.

Test that create only allows text file extensions.

Test that root /memories directory cannot be deleted.

Test deleting a directory.

Test deleting a file.

Test deleting a nonexistent path.

Test that dependencies are documented (pip install or markdown mention).

Test that execution counts start from 1 (fresh kernel).

Test that the notebook has at least one cell.

Test that the notebook has at least one code cell.

Test that the notebook has kernel specification.

Test that the notebook starts with a title (markdown heading).

Test inserting at line 0 (beginning).

Test inserting at the end.

Test inserting in the middle.

Test insert with invalid line number.

Test error handling for missing parameters.

Test that notebook uses a supported nbformat version.

Test that no deprecated Claude models are used.