### Updated Epic 4 Documentation - Enhanced Story 4.3: Multi-video Analysis with Multi-Agent System - Three perspective agents (Technical, Business, User) - Synthesis agent for unified summaries - Integration with existing AI ecosystem - Increased effort from 28 to 40 hours - Enhanced Story 4.4: Custom Models & Enhanced Markdown Export - Executive summary generation (2-3 paragraphs) - Timestamped sections with [HH:MM:SS] format - Enhanced markdown structure with table of contents - Increased effort from 24 to 32 hours - Enhanced Story 4.6: RAG-Powered Video Chat with ChromaDB - ChromaDB vector database integration - RAG implementation using existing test patterns - Chat interface with timestamp source references - DeepSeek integration for AI responses ### Epic Effort Updates - Total Epic 4 effort: 126 → 146 hours - Remaining work: 72 → 92 hours - Implementation timeline extended to 4-5 weeks 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com> |
||
|---|---|---|
| .. | ||
| README.md | ||
| __init__.py | ||
| bulk_tool_caller.py | ||
| example.py | ||
README.md
Bulk Tool Caller
This module provides the BulkToolCaller class, which extends the MCPMixin to offer tools for performing multiple tool calls in a single request to a FastMCP server. This can be useful for optimizing interactions with the server by reducing the overhead of individual tool calls.
Usage
To use the BulkToolCaller, see the example example.py file. The BulkToolCaller can be instantiated and then registered with a FastMCP server URL. It provides methods to call multiple tools in bulk, either different tools or the same tool with different arguments.
Provided Tools
The BulkToolCaller provides the following tools:
call_tools_bulk
Calls multiple different tools registered on the MCP server in a single request.
-
Arguments:
tool_calls(list ofCallToolRequest): A list of objects, where each object specifies thetoolname andargumentsfor an individual tool call.continue_on_error(bool, optional): IfTrue, continue executing subsequent tool calls even if a previous one resulted in an error. Defaults toTrue.
-
Returns: A list of
CallToolRequestResultobjects, each containing the result (isError,content) and the originaltoolname andargumentsfor each call.
call_tool_bulk
Calls a single tool registered on the MCP server multiple times with different arguments in a single request.
-
Arguments:
tool(str): The name of the tool to call.tool_arguments(list of dict): A list of dictionaries, where each dictionary contains the arguments for an individual run of the tool.continue_on_error(bool, optional): IfTrue, continue executing subsequent tool calls even if a previous one resulted in an error. Defaults toTrue.
-
Returns: A list of
CallToolRequestResultobjects, each containing the result (isError,content) and the originaltoolname andargumentsfor each call.