Fuzz web targets with ffuf for content, parameter, and vhost discovery — FUZZ keyword placement in paths, query strings, POST bodies and Host headers, matcher/filter calibration (-mc/-fs/-ac), recursion, and when to switch to feroxbuster/gobuster/dirsearch. Load when a web target needs hidden paths,
v2026.10.14-ee0e0c53 instalações+3 em 7 dias★ 27 EstrelasMITsincronizado há 22 horas
Instalar
Ainda não tem a CLI da NotHoly? Instale primeiro.
$ curl -fsSL https://notholy.com/install.sh | sh
$ notholy skills install noorqureshi/tools-ffuf
Grava em ~/.notholy/skills depois de conferir a assinatura e o SHA-256.
AssinaturaAssinado, ed25519
m19WVPM3ga…lBO2yzAQ==A CLI confere o SHA-256 e a assinatura antes de gravar qualquer arquivo.
You have an in-scope web target (tradecraft-scope-roe) and need what the site map doesn't show:
hidden directories and files, undocumented GET/POST parameters, or name-based virtual hosts. This is
the active half of recon-content-discovery — wordlist-driven, noisy, and governed by roe.md rate
limits.
ffuf replaces the FUZZ keyword anywhere in a request — path, query, body, header — with each
wordlist entry, then classifies responses by status/size/word-count. Discovery is a filtering
problem: a wildcard or SPA route answers 200 to everything, so the skill is teaching ffuf what
"nothing found" looks like, then listing only the deviations.
Calibrate the noise floor — run once, note the junk response's status/size/words, then filter:
-mc 200,301,302 (match codes), -fs 1234 (hide size), -fc 403, or just -ac to auto-calibrate
against random canary words. Without this step every wildcarded app "finds" thousands of pages.
feroxbuster = recursion by default; gobuster = simple and steady (+ DNS mode); dirsearch = sane
defaults; wfuzz = multi-wordlist combos (-z file,a -z file,b).
Feed results back — discovered paths go to the Burp/web-* testing loop; discovered params
go to arjun for confirmation and then to the relevant injection skills.
Unfiltered output is fiction — a "1000 hits" run against a wildcard app is 1000 copies of the
same 200. Calibrate first (-ac or manual -fs/-fc/-fw), then trust the short list.
SPA catch-alls — any-path-200 with identical body size means a frontend router, not real
endpoints; filter by size and move on.
Rate limits and WAFs — default ffuf speed trips Cloudflare/ModSecurity and breaches many
bug-bounty RoEs; use -rate/-p (delay) and check roe.md before launching.
Recursion explodes — depth 3 on a big wordlist × every found dir can run for hours; cap with
-recursion-depth and scope recursion to interesting dirs.
Windows/curl quirk — when verifying hits with curl, [] in URLs (e.g. __proto__[test])
needs --globoff or curl misparses them as ranges.
Match the wordlist to the stack — raft/directory-list for general, CMS-specific lists after
fingerprinting (recon-techstack-fingerprinting), burp-parameter-names for params,
subdomains-top1million-5000 for vhosts.
Every reported path/param/vhost is manually re-requested (curl/Burp) and returns content that
differs from the calibrated baseline — a page of unverified scanner hits is a lead list, not a
result.