feat: wrap get_examples tool output in XML tags with numbered IDs

feat: implement caching for tool calls to avoid redundant executions

docs: improve tool descriptions for better LLM understanding and usage guidance

feat: add count parameter to get_examples tool with default value of 5

refactor: update function descriptions for clarity and specificity

feat: add get_examples tool for retrieving text snippets

feat: add get_template tool for template suggestions

feat: disable parallel tool calls in LLM client configuration

fix: prevent infinite loops in API tool calls by tracking tool usage

feat: add keep_alive parameter to llm client configuration

fix: use configured language code when loading command-specific system prompt appendages

refactor: exclude content field from assistant message in tool calls

fix: preserve original tool call IDs when processing assistant messages

docs: clarify assistant message handling in tool calls

refactor: restructure namespace and streamline API call handling

fix: ensure assistant role is set when handling tool calls

refactor: reposition getAvailableTools method and adjust system prompt concatenation

refactor: extract tool definition to getAvailableTools method

feat: add tool call handling and recursive API call support

fix: correct regex pattern to remove content between <tool_call> and </tool_call> tags

feat: add tool call handling to LLM client API response processing

feat: add tool calling support with get_document function

refactor: update LLM metadata tags and fix template function name

fix: implement missing findTemplate method in action_plugin_dokullm class

feat: add find_template endpoint to dokullm action handler