VoltAgent/voltagent

AI Summary: Debug a bug in the VoltAgent framework where structured output generation using `generateObject` and tool calling do not work simultaneously. The issue seems to stem from incompatibility between the `generateObject` function and the `addTools` method, potentially related to the Vercel AI SDK. The task involves investigating the interaction between these features, identifying the root cause of the failure, and proposing a solution or workaround.

AI Summary: The task is to improve the VoltAgent framework's resilience to errors during MCP tool execution. Currently, a timeout or error in an MCP tool call crashes the entire agent. The solution requires modifying the framework to gracefully handle tool failures, allowing the agent to continue execution or attempt recovery strategies instead of crashing. This involves capturing and returning tool errors to the LLM, enabling the LLM to make informed decisions based on the failure.

AI Summary: Implement client-side tool call functionality in the VoltAgent framework for scenarios where a server-side execution function is unavailable. This involves enabling the framework to pass tool calls directly to the frontend, leveraging existing frontend capabilities for tool usage and potentially integrating with standards like the Model Context Protocol (MCP).

AI Summary: Implement a configurable timeout setting for the MCP client within the VoltAgent framework. This allows users to adjust the timeout period for long-running MCP tool executions, preventing indefinite hangs.

AI Summary: Update the VoltAgent documentation to include `user_id` and `conversation_id` columns in the SQL schema for the `voltagent_memory_agent_history` table used by the SupabaseMemory. The provided SQL statement needs to be added or corrected within the relevant section of the documentation.

AI Summary: The task is to debug and fix a bug in the VoltAgent framework where the UserContext object is not included in the response of the `.generateText` method. This requires examining the VoltAgent codebase, identifying where the UserContext is lost during processing, and modifying the code to include it in the final response object.

AI Summary: The task is to refactor the VoltAgent TypeScript framework to make all hardcoded default values configurable via a central configuration file (e.g., `defaults.ts`), allowing overrides through environment variables and command-line arguments. This involves identifying all hardcoded defaults, moving them to the configuration file, and updating the code to use the configurable values.

AI Summary: The task is to debug and fix a regression bug in the `@voltagent/postgres` package (version 0.1.5). The bug causes the PostgresStorage to fail initialization due to a malformed SQL query in the `CREATE TABLE` statement. The root cause is identified as incorrect quoting of the default value for the `level` column. The solution involves changing double quotes to single quotes around the default value "INFO" in the SQL query.