A small, opinionated toolkit for the early phases of an authorized network engagement: fast service discovery and signal-focused host triage. Built and field-tested during OSCP preparation — designed to be quieter and higher-signal than the usual all-in-one enumeration scripts.
⚠️ For authorized security testing, CTFs, and lab environments only. Scanning or running these tools against systems you do not own or have explicit written permission to test is illegal.
| Tool | Runs on | Purpose |
|---|---|---|
recon/recon.sh |
Attacker (Linux) | Three-phase nmap orchestration: full TCP sweep + parallel UDP + targeted service/version scan |
triage/oscp-linux-triage.sh |
Target (Linux) | Detection-only local triage — credential & misconfiguration hunting with a severity rubric |
triage/oscp-windows-triage.ps1 |
Target (Windows) | PowerShell credential hunter with the same signal-first philosophy |
Automates the nmap workflow you'd otherwise drive by hand across three terminals:
- UDP top-20 in the background (non-blocking), with a version/script follow-up on anything truly open.
- Full TCP port sweep (
-p- --min-rate=1000 -T4) in the foreground. - Targeted
-Ascan on exactly the ports found in step 2 — version detection, default scripts, OS guess.
./recon/recon.sh 10.10.10.10 # full run -> ./recon-10.10.10.10/
./recon/recon.sh --quick 10.10.10.10 # skip UDP
./recon/recon.sh 10.10.10.10 out-dir # custom output directoryDesign notes
- Parses discovered ports precisely (open TCP only) and pipes them straight into the targeted scan — no copy/paste between steps.
- Primes
sudoonce so the backgrounded UDP scan doesn't race the foreground scan for the password prompt; degrades cleanly underNOPASSWD. - Writes
-oN/-oAoutputs to a per-target directory so results are easy to grep later.
Validated end-to-end against a live Windows domain controller: a full 65,535-port sweep plus targeted service scan completed in ~5 minutes, and an independent higher-retry scan confirmed zero missed ports.
A deliberate alternative to linPEAS / winPEAS, built on two principles:
- Detection, not prescription. They report what was found — files, credentials, misconfigurations — against a severity rubric. They do not print pages of speculative "you could try…" attack steps.
- Signal over volume. Tuned to surface credentials and high-value findings first, with a false-positive filter so you aren't wading through framework boilerplate and package noise.
# Linux target
./triage/oscp-linux-triage.sh # CRIT/HIGH + system enum to stdout, full report to /tmp
./triage/oscp-linux-triage.sh --full # also print MED / FLAG findings# Windows target (PowerShell)
.\triage\oscp-windows-triage.ps1
.\triage\oscp-windows-triage.ps1 -FullSeverity model (both scripts):
| Level | Meaning |
|---|---|
CRITICAL |
bait-named file with secret content, private keys, or credentials found in another user's profile |
HIGH |
unambiguous secret formats (tokens, key=value credential assignments, URL-embedded creds) |
MED |
weaker indicators worth a look |
FLAG |
interesting filenames/locations to eyeball manually |
stdout carries CRITICAL + HIGH + system enumeration so findings survive truncation over
wmiexec / SMB shells; the complete report (including MED / FLAG) is written to a file.