The system prompt advertises the tool as list_summary_registry:
src/memorizz/memagent/core.py#L6967
memory_tools.append(
"• list_summary_registry — List all available summaries for this thread. "
)
However, the registered function is named list_summary_registry_tool:
src/memorizz/memagent/core.py#L1960
def list_summary_registry_tool() -> Dict[str, Any]:
It is registered without an explicit alias:
src/memorizz/memagent/core.py#L2162
self.tool_manager.add_tool(list_summary_registry_tool)
ToolManager uses func.name as the tool ID and name:
src/memorizz/memagent/managers/tool_manager.py#L172-L173
"_id": func.__name__,
"name": func.__name__,
Therefore, the OpenAI tool schema exposes list_summary_registry_tool, while the system prompt instructs the model to call list_summary_registry.
Note: This issue was identified by an automated testing tool for academic research and manually verified. If you have any concerns about this type of reporting, please let me know, and I will adjust my workflow accordingly.
The system prompt advertises the tool as
list_summary_registry:src/memorizz/memagent/core.py#L6967
However, the registered function is named
list_summary_registry_tool:src/memorizz/memagent/core.py#L1960
It is registered without an explicit alias:
src/memorizz/memagent/core.py#L2162
ToolManager uses func.name as the tool ID and name:
src/memorizz/memagent/managers/tool_manager.py#L172-L173
Therefore, the OpenAI tool schema exposes
list_summary_registry_tool, while the system prompt instructs the model to calllist_summary_registry.