Clear it in 3 steps
-
Start the environment
-
Investigate the target
-
Submit the flag
`/challenge/src/` contains the Python source of a web application. Use static analysis (`bandit`, `semgrep`) to enumerate dangerous sinks (`os.system`, `subprocess` with `shell=True`) and find the single genuinely exploitable file where untrusted request input (`request.args`/`form`/`json`/`values`) reaches a sink. The flag is in a trailing comment on that line. `shell=False` argv calls and parameterized SQL are decoys.
Search /challenge/src/ for dangerous sinks (os.system, subprocess with shell=True)
Only the sink that untrusted request input (request.args/form/json/values) reaches is genuinely exploitable
The flag is in a trailing comment on that one line