跳到正文
cccskills
免费在 GitHub 公开

content-discovery

Fuzz directories, files, and parameters using ffuf, feroxbuster, arjun, paramspider, and x8. Use when the user has live hosts and needs to find hidden routes, backup files, exposed configs, and untracked URL parameters.

中文简介正在准备中。当前显示原文说明。

集成工具WordPress

实测评分 未实测

查看安装方法

包含的文件(1)

  • SKILL.md9.1 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Content Discovery & Parameter Mining

The hidden /admin/v2/internal/ route is the bug. Find it.

When to invoke

Trigger phrases:

  • "fuzz directories"
  • "find hidden paths"
  • "discover parameters"
  • "ffuf this"
  • "directory brute force"

Two-stage approach

STAGE 1: Path discovery        STAGE 2: Parameter mining
┌─────────────────┐            ┌─────────────────┐
│ ffuf / ferox    │            │ arjun           │
│ Find hidden     │  ────►     │ paramspider     │
│ paths, files    │            │ x8              │
│                 │            │ Find params on  │
└─────────────────┘            │ found endpoints │
                               └─────────────────┘

Stage 1: Directory & file fuzzing

A. ffuf — fast, scriptable

TARGET="https://app.target.com"
OUT="loot/$(echo $TARGET | sed 's|https://||')/fuzz"
mkdir -p "$OUT"

# Standard run
ffuf -u "$TARGET/FUZZ" \
    -w ~/tools/SecLists/Discovery/Web-Content/raft-large-directories.txt \
    -c -mc 200,204,301,302,307,401,403 \
    -fc 404 \
    -t 50 \
    -o "$OUT/ffuf-dirs.json" -of json

# Filter common false positives (size-based)
ffuf -u "$TARGET/FUZZ" -w wordlist.txt -fs 1234,5678 -c

# Extensions for files
ffuf -u "$TARGET/FUZZ" \
    -w ~/tools/SecLists/Discovery/Web-Content/raft-medium-files.txt \
    -e .php,.bak,.old,.zip,.tar.gz,.sql,.env,.json,.yml,.yaml,.config \
    -c -mc 200,204,301,302,307,401,403 \
    -o "$OUT/ffuf-files.json" -of json

B. feroxbuster — recursive, robust

feroxbuster -u "$TARGET" \
    -w ~/tools/SecLists/Discovery/Web-Content/raft-large-directories.txt \
    -r \
    -s 200,204,301,302,401,403 \
    -t 50 \
    -d 3 \
    -o "$OUT/ferox.txt"

# With extensions
feroxbuster -u "$TARGET" \
    -w wordlist.txt \
    -x .php,.bak,.zip,.env,.json \
    -r -d 2 -t 30 \
    -o "$OUT/ferox-ext.txt"

C. dirsearch — Python alternative

dirsearch -u "$TARGET" \
    -w ~/tools/SecLists/Discovery/Web-Content/raft-large-directories.txt \
    -e bak,old,zip,env,sql \
    -t 30 \
    -x 404 \
    -o "$OUT/dirsearch.txt"

Wordlist strategy (CRITICAL)

Default wordlists find what everyone finds. Layer wordlists for maximum coverage:

# Layer 1: standard
~/tools/SecLists/Discovery/Web-Content/raft-large-directories.txt

# Layer 2: assetnote (modern, big)
~/tools/assetnote-wordlists/data/manual/httparchive_directories_1m.txt

# Layer 3: tech-specific (after fingerprinting)
# If WordPress:
~/tools/SecLists/Discovery/Web-Content/CMS/wordpress.txt
# If Spring Boot:
~/tools/SecLists/Discovery/Web-Content/spring-boot.txt
# If API discovery:
~/tools/SecLists/Discovery/Web-Content/api/api-endpoints-mazen160.txt

# Layer 4: target-specific (from JS-mined endpoints)
cat loot/target/js/endpoints.txt | unfurl paths | sort -u > target-custom.txt
ffuf -u "$TARGET/FUZZ" -w target-custom.txt -c

High-value paths to always include

Always test these even if not in wordlist:

HIGH_VALUE=(
    /.env
    /.git/config
    /.git/HEAD
    /.git/logs/HEAD
    /.DS_Store
    /server-status
    /server-info
    /swagger.json
    /swagger-ui/
    /api-docs
    /openapi.json
    /openapi.yaml
    /actuator           # Spring Boot
    /actuator/env       # Spring Boot CRITICAL
    /actuator/heapdump  # Spring Boot CRITICAL
    /actuator/health
    /metrics
    /debug
    /trace
    /phpinfo.php
    /info.php
    /wp-config.php.bak
    /robots.txt
    /sitemap.xml
    /security.txt
    /.well-known/security.txt
    /backup.zip
    /backup.tar.gz
    /db.sql
    /database.sql
    /dump.sql
    /api/v1
    /api/v2
    /api/v3
    /graphql
    /graphiql
    /__graphql
    /api/graphql
    /v1/graphql
    /admin
    /administrator
    /admin.php
    /wp-admin
    /console
    /jolokia            # Spring Boot mgmt
    /env
    /heapdump
    /threaddump
    /loggers
    /h2-console         # H2 DB console
)

for path in "${HIGH_VALUE[@]}"; do
    code=$(curl -s -o /dev/null -w "%{http_code}" "$TARGET$path")
    if [[ "$code" != "404" ]]; then
        echo "[$code] $TARGET$path"
    fi
done

Stage 2: Parameter discovery

A. Arjun — fast HTTP parameter discovery

# Single endpoint
arjun -u "$TARGET/api/v3/search" -m GET -oJ "$OUT/arjun-search.json"

# Multiple endpoints
arjun -i endpoints-live.txt -m GET -oJ "$OUT/arjun-bulk.json"

# POST endpoints (try JSON body)
arjun -u "$TARGET/api/v3/user" -m JSON -oJ "$OUT/arjun-user-post.json"

# Headers — arjun can probe header parameters too
arjun -u "$TARGET/api/v3/order" --headers -oJ "$OUT/arjun-headers.json"

B. paramspider — mine historical params

paramspider -d target.com --level high -o "$OUT/paramspider.txt"

# Output is URL with FUZZ in each param position — pipe to ffuf or scanners

C. x8 — modern, fast, Rust-based

# Install: cargo install x8
x8 -u "$TARGET/api/v3/user" -w ~/tools/SecLists/Discovery/Web-Content/burp-parameter-names.txt

D. Manual discovery — from JS bundles

# From js-analysis output, extract parameter-like strings
grep -hoE '"[a-zA-Z_][a-zA-Z_0-9]{2,30}"\s*:' loot/target/js/files/*.js | \
    sed 's/[":]//g' | sort -u > "$OUT/params-from-js.txt"

# Use as ffuf input
ffuf -u "$TARGET/api/v3/search?FUZZ=test" -w "$OUT/params-from-js.txt" -c -mc 200 -fc 404

Recursion strategy

Found /admin/? Recurse. Found /api/? Recurse.

# feroxbuster does this natively with -r and -d
feroxbuster -u "$TARGET" -w wordlist.txt -r -d 4 -t 30

# ffuf needs scripting — but you can chain:
ffuf -u "$TARGET/FUZZ" -w wordlist.txt -mc 200,401,403 -o run1.json
# Extract found dirs from run1, then:
cat run1.json | jq -r '.results[] | select(.status==200 or .status==403) | .url' | \
    while read dir; do
        ffuf -u "${dir}/FUZZ" -w wordlist.txt -mc 200,401,403 -o "$OUT/$(basename $dir).json"
    done

VHost / subdomain discovery via fuzzing

If you suspect more apps behind same IP:

# Fuzz Host header
ffuf -u "$TARGET" -H "Host: FUZZ.target.com" \
    -w ~/tools/SecLists/Discovery/DNS/subdomains-top1million-110000.txt \
    -fs 1234 \
    -mc 200,301,302

API enumeration with kiterunner

Modern API discovery (uses semantic patterns, not just brute):

# kiterunner uses massive seed lists of real API routes
kr scan "$TARGET" -A=apiroutes-240528 -o "$OUT/kr.txt"

# Or against the live list
kr scan -A=apiroutes-240528 endpoints-live.txt -o "$OUT/kr.txt"

Output template

loot/target.com/fuzz/
├── ffuf-dirs.json
├── ffuf-files.json
├── ferox.txt
├── dirsearch.txt
├── arjun-bulk.json
├── paramspider.txt
├── params-from-js.txt
├── kr.txt
└── interesting/
    ├── 403-endpoints.txt        ← potential auth bypass
    ├── 401-endpoints.txt        ← protected, may have leaks
    ├── 500-endpoints.txt        ← error pages → info disclosure
    └── files-exposed.txt        ← .env, .git, swagger, etc.

Triaging results

After fuzzing, focus on:

  1. 5xx responses — server errors often leak stack traces, hostnames, paths.
  2. 403 responses — try path canonicalization tricks (//admin, /admin/./, /admin..;/).
  3. 401 responses — protected → can you bypass? See [[auth-bypass]].
  4. 200 responses on unexpected paths — may be admin/internal.
  5. Different content-length on common endpoints — fingerprints variants.

Cross-references

  • [[asset-discovery]] — provides live host list
  • [[js-analysis]] — feeds custom param/path lists
  • [[fingerprinting]] — choose tech-specific wordlists
  • [[auth-bypass]] — handle 401/403 findings
  • [[idor-hunting]] — test parameter discoveries

Common pitfalls

  1. Default wordlist only. Misses tech-specific paths.
  2. No size filter. WAF returns same-size 200 for everything → false positives.
  3. Ignoring 403/401. Often the most valuable findings.
  4. Not throttling. 200 threads → blocked or banned.
  5. Skipping extensions. .bak, .old, .zip are gold mines.

Speed vs accuracy

ModeThreadsWordlist sizeTimeCoverage
Stealth1010k30 minMedium
Standard50100k1 hrHigh
Aggressive2001M+4+ hrMaximum

For most BB programs, stealth or standard. Aggressive can trigger WAF / rate limits.

When fuzzing is forbidden

Some programs say "no automated scanning". You can still:

  • Manually test high-value paths (the list above)
  • Use JS-mined endpoints (passive discovery)
  • Use historical URLs (Wayback / gau)
  • Read the program's docs / blog for API hints

Document everything — if scope is unclear, ask the manager.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Probe subdomain list with httpx/dnsx/naabu to identify live hosts, open ports, status codes, titles, and basic fingerprints. Use after subdomain-enum produces a sub list. Output feeds into fingerprinting and content discovery.

中文简介正在准备中。当前显示原文说明。

0xGhostCAT/claude-ai-cyber-security-skills442026年6月3日 更新

Hunt full Account Takeover (ATO) by chaining lower-severity findings — IDOR-to-email-disclosure, password-reset-token-theft, OAuth-redirect-uri abuse, response-manipulation, cookie scope, CRLF in Set-Cookie, race-condition on signup. Use when the user wants to escalate small findings to ATO or thinks an ATO path exists.

中文简介正在准备中。当前显示原文说明。

0xGhostCAT/claude-ai-cyber-security-skills442026年6月3日 更新

Hunt authentication and authorization bypasses — forced browsing, path traversal in auth, header injection, JWT downgrade, OAuth state confusion, role manipulation. Use when an endpoint returns 401/403 but appears reachable.

中文简介正在准备中。当前显示原文说明。

0xGhostCAT/claude-ai-cyber-security-skills442026年6月3日 更新

Write Bugcrowd-format reports — VRT (Vulnerability Rating Taxonomy) mapping, P1-P5 severity, impact-led writing, and Bugcrowd-specific submission etiquette. Use AFTER triage-validation passes for Bugcrowd programs.

中文简介正在准备中。当前显示原文说明。

0xGhostCAT/claude-ai-cyber-security-skills442026年6月3日 更新

Hunt business logic flaws — race conditions, workflow state bypass, parameter pollution affecting logic, coupon/discount abuse, refund logic, multi-step process bypass, integer overflow, negative quantities, double-spend. Use when looking for non-CVE bugs that require understanding the app's intended workflow.

中文简介正在准备中。当前显示原文说明。

0xGhostCAT/claude-ai-cyber-security-skills442026年6月3日 更新

Hunt web cache poisoning, CDN cache poisoning, and cache deception attacks. Use when target uses a CDN (Cloudflare, Akamai, Fastly, AWS CloudFront) or has Varnish/nginx caches and you want to identify cache key flaws that let attackers serve malicious content to other users.

中文简介正在准备中。当前显示原文说明。

0xGhostCAT/claude-ai-cyber-security-skills442026年6月3日 更新

查看 0xGhostCAT 的全部技能

报告此技能的问题