Edit: Team Agent System Prompts

deci committed over 1 year ago

Edit: Code Exe Tool - The get_terminal_output function will now detect if the last 128 characters repeat 5 times consecutively and will stop execution if this pattern occurs, indicating a likely infinite loop. Now, when a repeating output pattern is detected: An error message will be printed to the console, indicating the loop and that the session is being reset. The log will be updated. self.reset_terminal(session=session) will be called to reset the specific session where the loop occurred. A detailed message will be returned to the agent, including the captured output, the loop detection confirmation, a notification that the session was reset, and advice to review the problematic code/command.

deci committed over 1 year ago

Edit: Code Exe Md prompt to further discourse line by line edits, multiple seperate f.write calls, complex string manipulation. encouraged reading, implement fix, and check syntax. context a little bulky but working for now.

deci committed over 1 year ago

Added: Task edit/update/delete methods, auto dependency true/false toggle, team leader planning phase, and task document handling, team leader integrate results task document handling *allows for persistent development between teams by leveraging team leader as an investigative task manager before and after team agent task execution cycles*

deci committed over 1 year ago

Edit: revised problem solving/bug fixing guidance to discourage common indentation-error prone methods.

deci committed over 1 year ago

Fix: dirty_jason.py the _parse_number method will gracefully handle cases where the parsed string is just '+', '-', or empty, returning it as a string instead of raising a ValueError.

deci committed over 1 year ago

Edit: prevent the FAISS/KeyError (or any extension error) from crashing agent loop at the agent.py level. The call to self.call_extensions("message_loop_prompts", ...) in prepare_prompt is now wrapped in a try/except block.

deci committed over 1 year ago

Edit: Code Exe Md to discourage multi-line f.write() calls. (prevent UI errors during code execution tool)

deci committed over 1 year ago

Edit: get_terminal_output function in code_execution_tool.py is now enhanced to detect common shell prompt patterns (like (venv) ...$, root@...:~#, etc.) using regex. When such a prompt is detected in the output, the function will immediately break and return the output, rather than waiting for the full timeout.

deci committed over 1 year ago

Added: Image Generation & Music Generation -both support for CPU based (debian slim) or gpu (CUDA based) images. -both utilize .sh script that creates and maintains a deticated instruments_venv for dependencies and include heartbeats that prevent terminal passback (api waste). -code_exe tool timeouts have been changes to a rolling window, if idle for 10 seconds it passes back, else there is no max exe time if a process is responsive.

deci committed over 1 year ago

Added: Team Agent Tool Edit: Prompts for Team Agent support Code Exe tool, input tool, and instructions to support individual terminal reset and better terminal management for complex coding tasks.

deci committed over 1 year ago

Added: Cuda Dockerfile. Edit: Initialize.sh to fix perms issue by adding -f force.

deci committed over 1 year ago

update gitignore docker/run/agent-zero**

deci committed over 1 year ago

Update README.md

frdel committed over 1 year ago

Update build.txt

frdel committed over 1 year ago

Update pre_install.sh

frdel committed over 1 year ago

npm amd64 conflict fix

frdel committed over 1 year ago

auto embeddings

frdel committed over 1 year ago

Merge branch 'testing' into main

frdel committed over 1 year ago

Update README.md

frdel committed over 1 year ago