Loading…...

Demo evidence · OWASP Juice Shop · overall risk 77/100

Public demo evidence · Web attack

We attacked
OWASP Juice Shop
in 1.0s.

Real attacks against OWASP Juice Shop — a deliberately weak web app we host. No customer data involved. This page is public and shareable.

run run_442cc9a7536a…profile standarddepth mediumcompleted 10/17/2026, 2:01:47 AM

4-dimensional defense score

higher detection & readiness = stronger posture

77/ 100

Overall riskCombined risk index across all axes (higher = riskier).

100/ 100

Attack exposure% of endpoints with exploitable findings.

59/ 100

Detection% of attacks that hit a detectable log signal.

25/ 100

Defense readinessHow close current defenses are to a tuned baseline.

22probes ran→9bypassed defenses→9Sigma rules emitted

Attack methodology

41%bypass rate9 of 22 probes bypassed defenses

Techniques with confirmed bypasses

A01A03A05A09

0Patches awaiting replay verification

Findings

9 total

critical×2high×6medium×1

Download Sigma rules (.yml)

mediumOWASP A05 — Confidential acquisitions document publicly readable→ https://juice-shop.sandbox.doesitdefend.ai/ftp/acquisitions.md
A05probea05-ftp-acquisitions-001
Probe a05-ftp-acquisitions-001 confirmed vulnerable. Indicator matched: status=200. Confidential acquisitions document publicly readable

TWO WAYS TO RUN YOUR OWN

Run it yourself.

Same engine, two paths. Pick the one that matches what you want to test.

FREE SANDBOX · ACCOUNT REQUIRED

Attack this same sandbox.

The Free plan covers the sandbox targets we host — Juice Shop and VulnBot. An account is required. Share a public demo evidence page with one link.

Run Juice Shop →Run VulnBot →

YOUR OWN TARGET · STARTER+

Point the engine at your web app.

STARTER+ EVIDENCE PACK

What paid evidence looks like.

Starter+ runs export a verifiable ZIP: SARIF for your tooling, CSV and JSON for triage, the raw attack traces behind every finding, a remediation playbook for LLM runs — and a one-page executive summary your leadership can read without opening a terminal.

manifest.json
Lists every file in the archive with its size and SHA-256 checksum, so anyone can verify integrity offline.
sarif · csv · json
Standard formats your scanner pipeline, SIEM and triage sheets already ingest — no vendor lock-in.
evidence/*
The actual request/response evidence captured during the simulation, each anchored to its producer checksum.
executive-summary.html
Scope, outcome, what to fix first and how to verify the pack — the page you forward to your boss.

Verify offline: shasum -a 256 <file>, compare against manifest.json.

See pricing

Sigma rule (auto-emitted) ▾

title: "OWASP A05 — Security Misconfiguration (directory listing / sensitive file exposure) — OWASP A05 — Confidential acquisitions document publicly readable"
id: "d906eb03-7618-4744-aa31-1fb337d0933e"
status: experimental
description: |
    Auto-generated starter rule from DoesItDefend finding fnd_4332622b59c4446e9bfec62b.
    
    Detection gap observed:
    Probe a05-ftp-acquisitions-001 confirmed vulnerable. Indicator matched: status=200. Confidential acquisitions document publicly readable
    
    Recommended defense:
    Disable directory listing (autoindex off on nginx, Options -Indexes on Apache). Move sensitive files out of the webroot. Apply deny-by-default policy to static files.
author: "DoesItDefend (auto-generated)"
references:
  - "doesitdefend://finding/fnd_4332622b59c4446e9bfec62b"
  - "https://attack.mitre.org/techniques/T1083/"
date: "2026-10-17"
tags:
  - attack.t1083
  - doesitdefend.severity.medium
logsource:
  product: webserver
  category: access
detection:
  selection:
    cs-uri-stem|contains:
      - /ftp
      - /.git
      - /.env
      - /encryptionkeys
    http_status: 200
  condition: selection
falsepositives:
  - "Intentionally-public asset directories"
level: medium

Starter rule — tune thresholds/filters for your environment before deploying

highOWASP A03 — Order-by SQL injection via sort param (error-based detection)→ https://juice-shop.sandbox.doesitdefend.ai/api/Products?sort=id;DROP TABLE Users--

A03probea03-sqli-order-by-001

Probe a03-sqli-order-by-001 confirmed vulnerable. Indicator matched: status=400. Order-by SQL injection via sort param (error-based detection)

Show the attack ▾

Request

GET /api/Products?sort=id;DROP TABLE Users--

Response

HTTP 400 — {&quot;message&quot;:&quot;Sorting not allowed on given attributes&quot;,&quot;errors&quot;:[&quot;id;DROP TABLE Users--&quot;]}

PATCHUse prepared statements (parameterized queries). Validate and sanitize input at the boundary. Apply context-aware output encoding to prevent XSS. Treat a WAF as a defensive layer, not the primary control.

Sigma rule (auto-emitted) ▾

title: "OWASP A03 — Injection (SQLi / XSS / NoSQLi) — OWASP A03 — Order-by SQL injection via sort param (error-based detection)"
id: "558b5424-c722-467d-a0d3-5e083a7fcc90"
status: experimental
description: |
    Auto-generated starter rule from DoesItDefend finding fnd_425f826246bf433b94341eab.
    
    Detection gap observed:
    Probe a03-sqli-order-by-001 confirmed vulnerable. Indicator matched: status=400. Order-by SQL injection via sort param (error-based detection)
    
    Recommended defense:
    Use prepared statements (parameterized queries). Validate and sanitize input at the boundary. Apply context-aware output encoding to prevent XSS. Treat a WAF as a defensive layer, not the primary control.
author: "DoesItDefend (auto-generated)"
references:
  - "doesitdefend://finding/fnd_425f826246bf433b94341eab"
  - "https://attack.mitre.org/techniques/T1190/"
date: "2026-10-17"
tags:
  - attack.t1190
  - doesitdefend.severity.high
logsource:
  product: webserver
  category: access
detection:
  keywords:
    - "' OR 1=1"
    - "UNION SELECT"
    - "DROP TABLE"
    - "<script>"
    - "javascript:"
    - "<img src=x onerror"
    - "<iframe src=\"javascript"
  condition: keywords
falsepositives:
  - "Penetration testing tools"
  - "Security research / bug bounty"
level: high

Starter rule — tune thresholds/filters for your environment before deploying

highOWASP A09 — Encryption keys directory listing leaks premium.key→ https://juice-shop.sandbox.doesitdefend.ai/encryptionkeys

A09probea09-secret-keys-001

Probe a09-secret-keys-001 confirmed vulnerable. Indicator matched: body~"premium.key". Encryption keys directory listing leaks premium.key

Show the attack ▾

Request

GET /encryptionkeys

Response

HTTP 200 —      listing directory /encryptionkeys * { margin: 0; padding: 0; outline: 0; } body { p…

PATCHNever expose keystore/secret directories via the webroot. Rotate compromised keys. Centralize secret management in Secret Manager/Vault.

Sigma rule (auto-emitted) ▾

title: "OWASP A09 — Logging/Security Logging Failure (key/secret directory exposure) — OWASP A09 — Encryption keys directory listing leaks premium.key"
id: "5a290c00-b879-45b7-8d7a-f51df63b491d"
status: experimental
description: |
    Auto-generated starter rule from DoesItDefend finding fnd_6547b01c36d6a1d5d4b94ca9.
    
    Detection gap observed:
    Probe a09-secret-keys-001 confirmed vulnerable. Indicator matched: body~"premium.key". Encryption keys directory listing leaks premium.key
    
    Recommended defense:
    Never expose keystore/secret directories via the webroot. Rotate compromised keys. Centralize secret management in Secret Manager/Vault.
author: "DoesItDefend (auto-generated)"
references:
  - "doesitdefend://finding/fnd_6547b01c36d6a1d5d4b94ca9"
  - "https://attack.mitre.org/techniques/T1552/001/"
date: "2026-10-17"
tags:
  - attack.t1552.001
  - doesitdefend.severity.high
logsource:
  product: webserver
  category: access
detection:
  response_keywords:
    - premium.key
    - private.key
    - id_rsa
    - .pem
  condition: response_keywords
falsepositives:
  - "Public example keys clearly marked as canaries"
level: high

Starter rule — tune thresholds/filters for your environment before deploying

highOWASP A05 — FTP directory listing exposed at /ftp→ https://juice-shop.sandbox.doesitdefend.ai/ftp

A05probea05-ftp-listing-001

Probe a05-ftp-listing-001 confirmed vulnerable. Indicator matched: body~"package.json". FTP directory listing exposed at /ftp

Show the attack ▾

Request

GET /ftp

Response

HTTP 200 —      listing directory /ftp * { margin: 0; padding: 0; outline: 0; } body { padding: 80p…

PATCHDisable directory listing (autoindex off on nginx, Options -Indexes on Apache). Move sensitive files out of the webroot. Apply deny-by-default policy to static files.

Sigma rule (auto-emitted) ▾

title: "OWASP A05 — Security Misconfiguration (directory listing / sensitive file exposure) — OWASP A05 — FTP directory listing exposed at /ftp"
id: "e7e01f4d-fa88-49e4-a333-9373b3057e3e"
status: experimental
description: |
    Auto-generated starter rule from DoesItDefend finding fnd_79939de458c2d65832fe5c21.
    
    Detection gap observed:
    Probe a05-ftp-listing-001 confirmed vulnerable. Indicator matched: body~"package.json". FTP directory listing exposed at /ftp
    
    Recommended defense:
    Disable directory listing (autoindex off on nginx, Options -Indexes on Apache). Move sensitive files out of the webroot. Apply deny-by-default policy to static files.
author: "DoesItDefend (auto-generated)"
references:
  - "doesitdefend://finding/fnd_79939de458c2d65832fe5c21"
  - "https://attack.mitre.org/techniques/T1083/"
date: "2026-10-17"
tags:
  - attack.t1083
  - doesitdefend.severity.high
logsource:
  product: webserver
  category: access
detection:
  selection:
    cs-uri-stem|contains:
      - /ftp
      - /.git
      - /.env
      - /encryptionkeys
    http_status: 200
  condition: selection
falsepositives:
  - "Intentionally-public asset directories"
level: medium

Starter rule — tune thresholds/filters for your environment before deploying

highOWASP A01 — Poison-null byte path traversal bypasses .md whitelist→ https://juice-shop.sandbox.doesitdefend.ai/ftp/package.json.bak%2500.md

A01probea01-pathtraversal-ftp-001

Probe a01-pathtraversal-ftp-001 confirmed vulnerable. Indicator matched: body~"dependencies". Poison-null byte path traversal bypasses .md whitelist

Show the attack ▾

Request

GET /ftp/package.json.bak%2500.md

Response

HTTP 200 — { &quot;name&quot;: &quot;juice-shop&quot;, &quot;version&quot;: &quot;6.2.0-SNAPSHOT&quot;, &quot;description&quot;: &quot;An intentionally insecure JavaScript Web Application&quot;, &quot;homepage&quot;: &quot;http://owasp-juice.shop&quot;, &quot;author&quot;: &quot;Björn Kimminich  (https://kimminich.de)&quot;, &quot;contributors&quot;: [ &quot;Björn…

PATCHEnforce per-resource authorization (not just authentication). Validate ownership via tenant_id/user_id before returning data. Add per-user rate limits on enumeration-prone endpoints.

Sigma rule (auto-emitted) ▾

title: "OWASP A01 — Broken Access Control (IDOR / admin-route exposure) — OWASP A01 — Poison-null byte path traversal bypasses .md whitelist"
id: "c699462a-7f0b-4e8a-85ef-d2ddbe0ee7f4"
status: experimental
description: |
    Auto-generated starter rule from DoesItDefend finding fnd_47014c49d28d1fccb5cfda33.
    
    Detection gap observed:
    Probe a01-pathtraversal-ftp-001 confirmed vulnerable. Indicator matched: body~"dependencies". Poison-null byte path traversal bypasses .md whitelist
    
    Recommended defense:
    Enforce per-resource authorization (not just authentication). Validate ownership via tenant_id/user_id before returning data. Add per-user rate limits on enumeration-prone endpoints.
author: "DoesItDefend (auto-generated)"
references:
  - "doesitdefend://finding/fnd_47014c49d28d1fccb5cfda33"
  - "https://attack.mitre.org/techniques/T1190/"
date: "2026-10-17"
tags:
  - attack.t1190
  - doesitdefend.severity.high
logsource:
  product: webserver
  category: access
detection:
  selection:
    cs-uri-stem|contains:
      - /api/Users
      - /rest/basket/
      - /api/Feedbacks/
      - /administration
      - /ftp/
  condition: selection
falsepositives:
  - "Legitimate admin access from allow-listed IPs"
level: high

Starter rule — tune thresholds/filters for your environment before deploying

high

OWASP A03 — Error-based UNION injection in product search: result-column mismatch forces an unhandled SQLite error into the response body, proving the payload reaches the statement builder verbatim

→ https://juice-shop.sandbox.doesitdefend.ai/rest/products/search?q=apple')) UNION SELECT id,email,password,role,deluxeToken,1,2,3 FROM Users--

A03probea03-sqli-search-001

Probe a03-sqli-search-001 confirmed vulnerable. Indicator matched: body~"SQLITE_ERROR". Error-based UNION injection in product search: result-column mismatch forces an unhandled SQLite error into the response body, proving the payload reaches the statement builder verbatim

Show the attack ▾

Request

GET /rest/products/search?q=apple&#x27;)) UNION SELECT id,email,password,role,deluxeToken,1,2,3 FROM Users--

Response

HTTP 500 —    Error: SQLITE_ERROR: SELECTs to the left and right of UNION do not have the same number of result columns * { margin: 0; padding: 0; outline: 0; } body { padding: 80px 100px; font: 13px &quot;Helvetica Neue&quot;, &quot;Luci…

Sigma rule (auto-emitted) ▾

title: "OWASP A03 — Injection (SQLi / XSS / NoSQLi) — OWASP A03 — Error-based UNION injection in product search: result-column mismatch forces an unhandled SQLite error into the response body, proving the payload reaches the statement builder verbatim"
id: "acf15471-53c8-4e1e-a285-7968577ae90c"
status: experimental
description: |
    Auto-generated starter rule from DoesItDefend finding fnd_98cf750876b07be7206b511a.
    
    Detection gap observed:
    Probe a03-sqli-search-001 confirmed vulnerable. Indicator matched: body~"SQLITE_ERROR". Error-based UNION injection in product search: result-column mismatch forces an unhandled SQLite error into the response body, proving the payload reaches the statement builder verbatim
    
    Recommended defense:
    Use prepared statements (parameterized queries). Validate and sanitize input at the boundary. Apply context-aware output encoding to prevent XSS. Treat a WAF as a defensive layer, not the primary control.
author: "DoesItDefend (auto-generated)"
references:
  - "doesitdefend://finding/fnd_98cf750876b07be7206b511a"
  - "https://attack.mitre.org/techniques/T1190/"
date: "2026-10-17"
tags:
  - attack.t1190
  - doesitdefend.severity.high
logsource:
  product: webserver
  category: access
detection:
  keywords:
    - "' OR 1=1"
    - "UNION SELECT"
    - "DROP TABLE"
    - "<script>"
    - "javascript:"
    - "<img src=x onerror"
    - "<iframe src=\"javascript"
  condition: keywords
falsepositives:
  - "Penetration testing tools"
  - "Security research / bug bounty"
level: high

Starter rule — tune thresholds/filters for your environment before deploying

highOWASP A01 — Administration UI route reachable without auth check on frontend→ https://juice-shop.sandbox.doesitdefend.ai/#/administration

A01probea01-admin-public-001

Probe a01-admin-public-001 confirmed vulnerable. Indicator matched: status=200. Administration UI route reachable without auth check on frontend

Show the attack ▾

Request

GET /#/administration

Response

HTTP 200 —      OWASP Juice Shop &lt;meta name=&quot;description&quot; cont…

PATCHEnforce per-resource authorization (not just authentication). Validate ownership via tenant_id/user_id before returning data. Add per-user rate limits on enumeration-prone endpoints.

Sigma rule (auto-emitted) ▾

title: "OWASP A01 — Broken Access Control (IDOR / admin-route exposure) — OWASP A01 — Administration UI route reachable without auth check on frontend"
id: "14fe93f1-6acc-4c0c-b922-a0044f3b2144"
status: experimental
description: |
    Auto-generated starter rule from DoesItDefend finding fnd_36ae87c68f8649fc3c34fbd9.
    
    Detection gap observed:
    Probe a01-admin-public-001 confirmed vulnerable. Indicator matched: status=200. Administration UI route reachable without auth check on frontend
    
    Recommended defense:
    Enforce per-resource authorization (not just authentication). Validate ownership via tenant_id/user_id before returning data. Add per-user rate limits on enumeration-prone endpoints.
author: "DoesItDefend (auto-generated)"
references:
  - "doesitdefend://finding/fnd_36ae87c68f8649fc3c34fbd9"
  - "https://attack.mitre.org/techniques/T1190/"
date: "2026-10-17"
tags:
  - attack.t1190
  - doesitdefend.severity.high
logsource:
  product: webserver
  category: access
detection:
  selection:
    cs-uri-stem|contains:
      - /api/Users
      - /rest/basket/
      - /api/Feedbacks/
      - /administration
      - /ftp/
  condition: selection
falsepositives:
  - "Legitimate admin access from allow-listed IPs"
level: high

Starter rule — tune thresholds/filters for your environment before deploying

criticalOWASP A03 — SQL injection in login email bypasses authentication (classic ' OR 1=1--)→ https://juice-shop.sandbox.doesitdefend.ai/rest/user/login

A03probea03-sqli-login-email-001

Probe a03-sqli-login-email-001 confirmed vulnerable. Indicator matched: body~"authentication". SQL injection in login email bypasses authentication (classic ' OR 1=1--)

Show the attack ▾

Request

POST /rest/user/login {&quot;email&quot;:&quot;&#x27; OR 1=1--&quot;,&quot;password&quot;:&quot;anything&quot;}

Response

HTTP 200 — {&quot;authentication&quot;:{&quot;token&quot;:&quot;eyJ0eXAiOiJKV1QiLCJhbGciOiJSUzI1NiJ9.eyJzdGF0dXMiOiJzdWNjZXNzIiwiZGF0YSI6eyJpZCI6MSwidXNlcm5hbWUiOiIiLCJlbWFpbCI6ImFkbWluQGp1aWNlLXNoLm9wIiwicGFzc3dvcmQiOiIwMTkyMDIzYTdiYmQ3MzI1MDUxNmYwNjlkZjE4YjUwMCIsInJvbGUiOiJhZG1pbiIsImRlbHV4ZVRva2VuIjo…

Sigma rule (auto-emitted) ▾

title: "OWASP A03 — Injection (SQLi / XSS / NoSQLi) — OWASP A03 — SQL injection in login email bypasses authentication (classic ' OR 1=1--)"
id: "59a31512-f215-411c-adf8-fbcb32a229d2"
status: experimental
description: |
    Auto-generated starter rule from DoesItDefend finding fnd_4df9523ae45a733ea774c535.
    
    Detection gap observed:
    Probe a03-sqli-login-email-001 confirmed vulnerable. Indicator matched: body~"authentication". SQL injection in login email bypasses authentication (classic ' OR 1=1--)
    
    Recommended defense:
    Use prepared statements (parameterized queries). Validate and sanitize input at the boundary. Apply context-aware output encoding to prevent XSS. Treat a WAF as a defensive layer, not the primary control.
author: "DoesItDefend (auto-generated)"
references:
  - "doesitdefend://finding/fnd_4df9523ae45a733ea774c535"
  - "https://attack.mitre.org/techniques/T1190/"
date: "2026-10-17"
tags:
  - attack.t1190
  - doesitdefend.severity.critical
logsource:
  product: webserver
  category: access
detection:
  keywords:
    - "' OR 1=1"
    - "UNION SELECT"
    - "DROP TABLE"
    - "<script>"
    - "javascript:"
    - "<img src=x onerror"
    - "<iframe src=\"javascript"
  condition: keywords
falsepositives:
  - "Penetration testing tools"
  - "Security research / bug bounty"
level: high

Starter rule — tune thresholds/filters for your environment before deploying

criticalOWASP A03 — SQL injection with comment terminator bypasses password check→ https://juice-shop.sandbox.doesitdefend.ai/rest/user/login

A03probea03-sqli-login-quote-001

Probe a03-sqli-login-quote-001 confirmed vulnerable. Indicator matched: body~"authentication". SQL injection with comment terminator bypasses password check

Show the attack ▾

Request

POST /rest/user/login {&quot;email&quot;:&quot;[EMAIL]&#x27;--&quot;,&quot;password&quot;:&quot;anything&quot;}

Response

HTTP 200 — {&quot;authentication&quot;:{&quot;token&quot;:&quot;eyJ0eXAiOiJKV1QiLCJhbGciOiJSUzI1NiJ9.eyJzdGF0dXMiOiJzdWNjZXNzIiwiZGF0YSI6eyJpZCI6MSwidXNlcm5hbWUiOiIiLCJlbWFpbCI6ImFkbWluQGp1aWNlLXNoLm9wIiwicGFzc3dvcmQiOiIwMTkyMDIzYTdiYmQ3MzI1MDUxNmYwNjlkZjE4YjUwMCIsInJvbGUiOiJhZG1pbiIsImRlbHV4ZVRva2VuIjo…

Sigma rule (auto-emitted) ▾

title: "OWASP A03 — Injection (SQLi / XSS / NoSQLi) — OWASP A03 — SQL injection with comment terminator bypasses password check"
id: "04a55d74-5d70-440b-afb1-d2d155dc3e72"
status: experimental
description: |
    Auto-generated starter rule from DoesItDefend finding fnd_de95686591a63621af9ee245.
    
    Detection gap observed:
    Probe a03-sqli-login-quote-001 confirmed vulnerable. Indicator matched: body~"authentication". SQL injection with comment terminator bypasses password check
    
    Recommended defense:
    Use prepared statements (parameterized queries). Validate and sanitize input at the boundary. Apply context-aware output encoding to prevent XSS. Treat a WAF as a defensive layer, not the primary control.
author: "DoesItDefend (auto-generated)"
references:
  - "doesitdefend://finding/fnd_de95686591a63621af9ee245"
  - "https://attack.mitre.org/techniques/T1190/"
date: "2026-10-17"
tags:
  - attack.t1190
  - doesitdefend.severity.critical
logsource:
  product: webserver
  category: access
detection:
  keywords:
    - "' OR 1=1"
    - "UNION SELECT"
    - "DROP TABLE"
    - "<script>"
    - "javascript:"
    - "<img src=x onerror"
    - "<iframe src=\"javascript"
  condition: keywords
falsepositives:
  - "Penetration testing tools"
  - "Security research / bug bounty"
level: high

Starter rule — tune thresholds/filters for your environment before deploying

Embed this badge

Show your LLM security score in your README, docs, or marketing page. The badge image is regenerated every 5 minutes from this run's data.

Markdown

[![LLM security badge](https://api.doesitdefend.ai/v1/badges/runs/run_442cc9a7536a601946338105.svg)](https://doesitdefend.ai/demo/runs/run_442cc9a7536a601946338105)

HTML

<a href="https://doesitdefend.ai/demo/runs/run_442cc9a7536a601946338105"><img src="https://api.doesitdefend.ai/v1/badges/runs/run_442cc9a7536a601946338105.svg" alt="LLM security badge"></a>

We attackedOWASP Juice Shopin 1.0s.

4-dimensional defense score

Findings

Run it yourself.

Attack this same sandbox.

Point the engine at your web app.

What paid evidence looks like.

Embed this badge

We attacked
OWASP Juice Shop
in 1.0s.