docs: consolidate vulnerabilities and threat profiles into one section

docs/docs/risks/custom-vulnerabilities.md

@@ -100,30 +100,75 @@ print(vuln.get_types())  # [<HIPAAComplianceType.PHI_DISCLOSURE>, ...]
 print(vuln.get_values())  # ['phi_disclosure', 'unauthorized_access']
 ```
 
-## Creating a Threat Profile (Optional)
+## Creating a Threat Profile
 
-To provide dataset and attack recommendations, create a threat profile:
+A threat profile is optional, but it's what lets an evaluation campaign auto-select datasets, attacks, objective, and metrics for your custom vulnerability instead of you wiring them up by hand each time — see [How Threat Profiles Work](./vulnerabilities.md#how-threat-profiles-work) for the shared `ThreatProfile` anatomy.
 
 ```python
 from hackagent.risks.profile_types import ThreatProfile
-from hackagent.risks.profile_helpers import ds, PRIMARY, STATIC_TEMPLATE_ONLY
+from hackagent.risks.profile_helpers import ds, PRIMARY, SECONDARY, STATIC_TEMPLATE_ONLY
 
 HIPAA_COMPLIANCE_PROFILE = ThreatProfile(
     vulnerability=HIPAACompliance,
     datasets=[
         ds(
-            "custom_hipaa_dataset",
+            "custom_hipaa_test_set",
             PRIMARY,
-            "Healthcare-specific test cases for PHI protection"
+            "Healthcare-specific scenarios testing PHI protection"
+        ),
+        ds(
+            "donotanswer",
+            SECONDARY,
+            "General refusal behavior baseline"
         ),
     ],
     attacks=STATIC_TEMPLATE_ONLY,
     objective="policy_violation",
-    metrics=["asr", "judge_score"],
+    metrics=["asr", "judge_score", "phi_leak_count"],
     description="Tests HIPAA compliance in healthcare AI systems.",
 )
+
+# Use it
+print(HIPAA_COMPLIANCE_PROFILE.name)  # "HIPAA Compliance"
+print(HIPAA_COMPLIANCE_PROFILE.dataset_presets)  # ['custom_hipaa_test_set', 'donotanswer']
+```
+
+### Profile Helpers
+
+The `profile_helpers` module provides utilities for building profiles:
+
+```python
+from hackagent.risks.profile_helpers import (
+    ds,                # Create DatasetRecommendation
+    PRIMARY,           # Relevance.PRIMARY
+    SECONDARY,         # Relevance.SECONDARY
+    STATIC_TEMPLATE_ONLY,  # Static Template-only attack list
+    JAILBREAK_ATTACKS, # Static Template + PAIR + AdvPrefix (secondary)
+    ALL_ATTACKS,       # Static Template + PAIR + AdvPrefix (all primary)
+)
+
+# Create a dataset recommendation
+dataset_rec = ds(
+    "advbench",
+    PRIMARY,
+    "Direct harmful behavior test cases"
+)
+
+# Use pre-built attack lists
+profile = ThreatProfile(
+    vulnerability=MyVuln,
+    datasets=[dataset_rec],
+    attacks=JAILBREAK_ATTACKS,
+    objective="jailbreak",
+    metrics=["asr"],
+)
 ```
 
+**Usage:**
+- **STATIC_TEMPLATE_ONLY** — Simple direct testing, no adversarial optimization
+- **JAILBREAK_ATTACKS** — Includes iterative refinement (PAIR) and gradient-based (AdvPrefix)
+- **ALL_ATTACKS** — Full attack suite for comprehensive adversarial testing
+
 ## BaseVulnerability Requirements
 
 When extending `BaseVulnerability`, you must provide:
@@ -364,6 +409,5 @@ If your custom vulnerability addresses a common threat, consider contributing it
 
 ## Learn More
 
-- **[Vulnerabilities](./vulnerabilities)** — Study the 13 built-in vulnerability implementations
-- **[Threat Profiles](./threat-profiles)** — Learn how to create threat profiles
+- **[Vulnerabilities](./vulnerabilities)** — Study the 13 built-in vulnerability implementations and their threat profiles
 - **[BaseVulnerability API](../hackagent/agent)** — Full API reference

docs/docs/risks/evaluation-campaigns.md

@@ -583,7 +583,6 @@ for profile in rag_profiles:
 
 ## Learn More
 
-- **[Vulnerabilities](./vulnerabilities)** — Complete reference for all 13 vulnerability classes
-- **[Threat Profiles](./threat-profiles)** — Detailed profile documentation
+- **[Vulnerabilities](./vulnerabilities)** — Complete reference for all 13 vulnerability classes, each with its threat profile
 - **[Datasets](/datasets)** — Available dataset presets
 - **[Attacks](/attacks)** — Attack techniques and configuration

docs/docs/risks/evaluation-campaigns/custom-campaigns.md

@@ -467,4 +467,4 @@ def send_slack_alert(campaign_name, results):
 - **[Quick Scan](./quick-scan)** - Fast vulnerability scanning
 - **[Comprehensive Audit](./comprehensive-audit)** - Full security assessment
 - **[Targeted Assessment](./targeted-assessment)** - Focus on specific attack surfaces
-- **[Threat Profiles](../threat-profiles)** - Pre-built vulnerability profiles
+- **[Vulnerabilities](../vulnerabilities)** - Pre-built vulnerability profiles

docs/docs/risks/index.mdx

@@ -51,7 +51,7 @@ The defined categories are not intended to be mutually exclusive, nor to form an
 ## What Is Implemented Today
 
 - **Implemented Risk Macro-Category**: Cybersecurity
-- **Implemented Risk Micro-Categories**: 13 built-in vulnerabilities, each with a ready-to-run threat profile — see [Vulnerability Categories](./risk-categories.mdx) for the full map and [Threat Profiles](./threat-profiles.md) for per-category datasets/attacks/metrics.
+- **Implemented Risk Micro-Categories**: 13 built-in vulnerabilities, each with a ready-to-run threat profile — see [Vulnerabilities](./vulnerabilities.md) for the full reference and per-category datasets/attacks/metrics.
 - **Documented Dedicated Scenario**: Indirect Injection (RAG context poisoning)
 - **Extensible**: define your own vulnerability categories — see [Custom Vulnerabilities](./custom-vulnerabilities.md)
 - Current quick flow: [Evaluation Campaign](./evaluation-campaigns.md)
@@ -63,14 +63,13 @@ The defined categories are not intended to be mutually exclusive, nor to form an
 | **Risk Profile** | Defines the scope of the assessment as one or more macro-categories (or micro-categories). |
 | **Evaluation Campaign** | Operationalizes the risk profile into an executable plan: datasets, attacks, objective, and metrics. |
 
-For the currently implemented configuration, see [Vulnerability Categories](./risk-categories.mdx), [Threat Profiles](./threat-profiles.md), and [Indirect Injection](./indirect-prompt-injection.md).
+For the currently implemented configuration, see [Vulnerabilities](./vulnerabilities.md) and [Indirect Injection](./indirect-prompt-injection.md).
 
 ## Documentation Guide
 
 | Page | Description |
 |------|-------------|
-| [Vulnerability Categories](./risk-categories.mdx) | The 13 built-in vulnerability classes, organized by attack surface layer |
-| [Threat Profiles](./threat-profiles.md) | Per-category recommended datasets, attack techniques, objective, and metrics |
+| [Vulnerabilities](./vulnerabilities.md) | Per-vulnerability reference, each with its threat profile — recommended datasets, attack techniques, objective, and metrics |
 | [Custom Vulnerabilities](./custom-vulnerabilities.md) | Extend `BaseVulnerability` to define your own categories |
 | [Indirect Injection](./indirect-prompt-injection.md) | Dedicated risk scenario for poisoned retrieved context in RAG pipelines |
 | [Evaluation Campaigns](./evaluation-campaigns.md) | Quick scans, comprehensive audits, targeted assessments, and custom campaigns |