| name | python-debugpy |
| description | Debug Python: pdb REPL + debugpy remote (DAP). |
| allowed-tools | ["bash","read_file"] |
| enabled | true |
| related-skills | ["systematic-debugging","node-inspect-debugger"] |
| license | MIT |
| author | Adapted from hermes-agent (Nous Research, MIT) |
Python Debugger (pdb + debugpy)
Overview
Three tools, picked by situation:
| Tool | When |
|---|
breakpoint() + pdb | Local, interactive, simplest. Add breakpoint() in source, run, get REPL. |
python -m pdb | Launch script under pdb with no source edits. |
debugpy | Remote / headless / attach to running process. DAP, scriptable. |
Start with breakpoint(). It's the cheapest thing that works.
When to Use
- A test fails and the traceback doesn't reveal why a value is wrong
- You need to step through a function and watch a collection mutate
- A long-running process misbehaves and you can't restart it
- Post-mortem: an exception fired and you want to inspect locals at crash site
- A subprocess is the actual bug site
Don't use for: things print() / logging.debug solve in under a minute,
or things pytest -vv --tb=long --showlocals already reveals.
pdb Quick Reference
Inside any pdb prompt ((Pdb)):
| Command | Action |
|---|
h / h cmd | help |
n | next line (step over) |
s | step into |
r | return from current function |
c | continue |
unt N | continue until line N |
j N | jump to line N (same function only) |
b N | set breakpoint at line N |
b file:N | set breakpoint in another file |
b func | set breakpoint at function |
cl N | clear breakpoint N |
l | list 11 lines around current |
ll | list whole function |
w / where | show call stack |
u / d | move up/down stack frame |
p expr | print expression |
pp expr | pretty-print expression |
a | print args of current function |
args | same as a |
display expr | watch expression (re-eval each step) |
interact | drop into interactive Python REPL |
Using breakpoint()
def process(data):
result = transform(data)
breakpoint()
return result
Run normally:
python script.py
Using python -m pdb (no source edits)
python -m pdb script.py
Post-mortem debugging
Drop into pdb at the exact exception site:
import pdb, traceback
try:
main()
except Exception:
traceback.print_exc()
pdb.post_mortem()
Or:
python -m pdb -c continue script.py
debugpy (remote attach)
For long-running processes or headless debugging:
pip install debugpy
python -m debugpy --listen 5678 --wait-for-client script.py
import debugpy
debugpy.listen(5678)
print("Waiting for debugger on port 5678...")
debugpy.wait_for_client()
Attach from another terminal (DAP client):
python -m debugpy --connect localhost:5678 --set-breakpoint script.py:42
Or use any DAP-compatible editor (VS Code, Neovim) to attach to port 5678.
Common Debugging Patterns
Watch a variable change
(Pdb) display my_list
Conditional breakpoint
breakpoint() if condition else None
(Pdb) b 42, x > 100
Inspect a running subprocess
import debugpy
debugpy.listen(5679)
debugpy.wait_for_client()
Pitfalls
- breakpoint() in production: remove before commit. Use
breakpoint() only
for local debugging.
- pdb + multiprocessing: child processes don't inherit the pdb prompt.
Use
debugpy for multi-process debugging.
- pdb + asyncio: pdb works but async stack traces can be confusing. Use
w (where) to see the full async call stack.
- Windows + pdb: some pdb features work differently on Windows.
python -m pdb is more reliable than breakpoint() in some Windows terminals.
- debugpy port conflicts: default port 5678 may be taken. Use
--listen 5679.