ChatDBG is an open-source AI debugging assistant that integrates large language models into standard debuggers for C, C++, Python and Rust, so you can ask your debugger open-ended questions like “why is x null?” and get a root cause analysis with suggested fixes. Built by the PLASMA lab at UMass Amherst, it is Apache-2.0 licensed, has about 1,100 stars on GitHub, and is actively developed in 2026.
Debuggers tell you where a program stopped, not why it went wrong. ChatDBG closes that gap by letting an LLM take the wheel: connected to pdb, lldb or gdb, it inspects the failing state, issues debugger commands itself, explores stack frames and variables, and converses with you about what it finds. The result is an interactive diagnosis session: ChatDBG explains the error, identifies the root cause, and proposes concrete fixes, which its authors describe as the first debugger to automatically perform root cause analysis of this kind. The project comes from Emery Berger’s research group, known for tools that measurably improve developer productivity, and is documented in a research paper and talks. It installs from PyPI as a drop-in enhancement to the debuggers programmers already use daily.
ChatDBG is free and open source under the Apache-2.0 license. You only pay for the LLM API usage it makes on your behalf, or nothing when using it with a suitably configured local model.
ChatDBG is for C, C++, Rust and Python developers who lose time to cryptic crashes and state bugs, and for teams who want AI pair-debugging inside a real debugger instead of pasting code into a chatbot.