FzKuji

GUI automation via visual perception. Screenshot → detect → click → verify. Use when asked to operate a desktop app, interact with a VM, or complete an OSWorld benchmark task.

Run an autonomous agent on a complex task with quality evaluation. Clarifies upfront, then works fully autonomously until the result meets the standard. Triggers: 'deep work', 'work on this seriously', 'high quality task', 'write a paper', 'complex task', 'do this properly', 'professional quality'.

一个测试的agentic函数，用于测试我们框架的功能，无实际用途

Fix a broken agentic function using LLM analysis. Reads source code and error history, then rewrites the function. Triggers: 'fix this function', 'fix function', 'this function is broken', 'repair', 'rewrite function', 'function doesn't work'.

Analyze text sentiment using LLM. Returns positive, negative, or neutral. Triggers: 'sentiment', 'is this positive', 'analyze mood', 'what tone'.

List files and folders in a directory with sizes. Triggers: 'list files', 'show directory', 'what files are in', 'ls'.

Generate a complete runnable Python app with LLM runtime, agentic functions, argparse, and main(). Triggers: 'create an app', 'build an app', 'generate a script', 'make a tool', 'scaffold an application'.

Generate a Python function from a natural language description using Agentic Programming. The LLM writes the code, the framework validates and sandboxes it. Triggers: 'create a function', 'generate a function', 'write a function that', 'make a function', 'I need a function'.

Run a saved agentic function by name with arguments. Use 'openprogram list' to see available functions. Triggers: 'run function', 'execute function', 'call function', 'use function', 'run sentiment', 'run extract'.