Skip to main content
Glama
CSOAI-ORG

GDPR Compliance for AI Systems MCP Server

breach_notification

Assess breach severity and determine GDPR notification requirements, including 72-hour rule, for personal data breaches. Generate required notification content for authorities and data subjects.

Instructions

Assess breach severity and notification requirements under GDPR Articles 33-34 (72-hour rule). Determines whether supervisory authority and data subject notification is required, and generates the notification content.

Args:
    breach_description: Description of the personal data breach
    data_categories_affected: Types of personal data affected
    number_of_records: Approximate number of records/individuals affected
    breach_type: Type of breach: "confidentiality" (unauthorized access), "integrity" (unauthorized alteration), "availability" (unauthorized loss of access)
    detection_timestamp: When the breach was detected (ISO format, or "now")
    ai_system_involved: Whether an AI system was involved in the breach
    caller: Caller identifier for rate limiting
    tier: Access tier (free/pro)

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
breach_descriptionYes
data_categories_affectedYes
number_of_recordsNo
breach_typeNoconfidentiality
detection_timestampNo
ai_system_involvedNo
callerNoanonymous
tierNofree
api_keyNo

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
resultYes

Implementation Reference

  • The main tool handler function for 'breach_notification'. Assesses breach severity under GDPR Articles 33-34 (72-hour rule), determines supervisory authority and data subject notification requirements, and generates notification content.
    @mcp.tool()
    def breach_notification(
        breach_description: str,
        data_categories_affected: list[str],
        number_of_records: int = 0,
        breach_type: str = "confidentiality",
        detection_timestamp: str = "",
        ai_system_involved: bool = False,
        caller: str = "anonymous",
        tier: str = "free", api_key: str = "") -> str:
        """Assess breach severity and notification requirements under GDPR Articles
        33-34 (72-hour rule). Determines whether supervisory authority and data
        subject notification is required, and generates the notification content.
    
        Args:
            breach_description: Description of the personal data breach
            data_categories_affected: Types of personal data affected
            number_of_records: Approximate number of records/individuals affected
            breach_type: Type of breach: "confidentiality" (unauthorized access), "integrity" (unauthorized alteration), "availability" (unauthorized loss of access)
            detection_timestamp: When the breach was detected (ISO format, or "now")
            ai_system_involved: Whether an AI system was involved in the breach
            caller: Caller identifier for rate limiting
            tier: Access tier (free/pro)
        """
        allowed, msg, tier = check_access(api_key)
        if not allowed:
            return {"error": msg, "upgrade_url": "https://meok.ai/pricing"}
        if err := _check_rate_limit(caller, tier):
            return {"error": err}
    
        if detection_timestamp and detection_timestamp != "now":
            try:
                detected = datetime.fromisoformat(detection_timestamp)
            except ValueError:
                detected = datetime.now(timezone.utc)
        else:
            detected = datetime.now(timezone.utc)
    
        deadline = detected + timedelta(hours=72)
    
        # Severity assessment
        special_categories = {"biometric", "health", "genetic", "financial", "criminal", "password", "credential"}
        has_special = any(any(sc in d.lower() for sc in special_categories) for d in data_categories_affected)
    
        severity_score = 0
        if has_special:
            severity_score += 3
        if number_of_records > 100000:
            severity_score += 3
        elif number_of_records > 10000:
            severity_score += 2
        elif number_of_records > 100:
            severity_score += 1
        if breach_type == "confidentiality":
            severity_score += 2
        elif breach_type == "integrity":
            severity_score += 2
        else:
            severity_score += 1
        if ai_system_involved:
            severity_score += 1
    
        if severity_score >= 7:
            severity = "CRITICAL"
            risk_to_rights = "high"
        elif severity_score >= 5:
            severity = "HIGH"
            risk_to_rights = "high"
        elif severity_score >= 3:
            severity = "MEDIUM"
            risk_to_rights = "some"
        else:
            severity = "LOW"
            risk_to_rights = "unlikely"
    
        notify_authority = risk_to_rights != "unlikely"
        notify_data_subjects = risk_to_rights == "high"
    
        result = {
            "assessment_type": "GDPR Breach Notification Assessment",
            "timestamp": datetime.now(timezone.utc).isoformat(),
            "breach": {
                "description": breach_description,
                "data_categories": data_categories_affected,
                "records_affected": number_of_records,
                "breach_type": breach_type,
                "detection_time": detected.isoformat(),
                "ai_involved": ai_system_involved,
            },
            "severity_assessment": {
                "severity": severity,
                "severity_score": severity_score,
                "risk_to_rights_and_freedoms": risk_to_rights,
                "special_categories_affected": has_special,
            },
            "notification_requirements": {
                "notify_supervisory_authority": {
                    "required": notify_authority,
                    "article": "Article 33",
                    "deadline": deadline.isoformat(),
                    "deadline_human": "72 hours from awareness of breach",
                    "content_required": [
                        "Nature of the breach including categories and approximate number of data subjects",
                        "Name and contact details of DPO or other contact point",
                        "Likely consequences of the breach",
                        "Measures taken or proposed to address the breach and mitigate effects",
                    ],
                },
                "notify_data_subjects": {
                    "required": notify_data_subjects,
                    "article": "Article 34",
                    "condition": "When breach is likely to result in HIGH risk to rights and freedoms",
                    "deadline": "Without undue delay",
                    "content_required": [
                        "Nature of the breach in clear and plain language",
                        "Name and contact details of DPO",
                        "Likely consequences of the breach",
                        "Measures taken or proposed to address the breach",
                    ],
                    "exceptions": [
                        "Data was encrypted/unintelligible to unauthorized persons",
                        "Subsequent measures ensure high risk is no longer likely",
                        "Disproportionate effort — public communication may substitute",
                    ],
                },
            },
            "ai_specific": (
                {
                    "ai_breach_considerations": [
                        "Assess whether model was compromised (poisoned, extracted, or manipulated)",
                        "Check for training data leakage through model memorization",
                        "Evaluate adversarial attack vectors that caused the breach",
                        "Consider model rollback to pre-breach checkpoint",
                        "Report under EU AI Act Article 62 (serious incident) if applicable",
                    ],
                }
                if ai_system_involved
                else None
            ),
            "immediate_actions": [
                "Contain the breach — stop unauthorized access/processing",
                "Preserve evidence for investigation",
                "Assess scope and impact",
                "Activate breach response team",
                f"Notify supervisory authority by {deadline.strftime('%Y-%m-%d %H:%M UTC')}" if notify_authority else "Document decision not to notify with rationale",
                "Notify affected data subjects without undue delay" if notify_data_subjects else None,
                "Update breach register (Art.33(5))",
            ],
        }
    
        result["immediate_actions"] = [a for a in result["immediate_actions"] if a]
    
        return result
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Without annotations, the description carries the burden of behavioral disclosure. It explains what the tool does (assess, determine, generate) and notes param roles (caller for rate limiting). However, it omits side effects (e.g., logging), authentication requirements, or rate limit specifics, leaving gaps in transparency.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is moderately concise with a clear paragraph explaining purpose and action, followed by an Args list. The front-loading of purpose is effective, though the Args list partially duplicates the schema. Overall, it is well-organized and not overly verbose.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the complexity of GDPR breach notification and the presence of an output schema, the description covers inputs and expected output (notification content). However, it could briefly mention output structure or any constraints (e.g., only works for specific jurisdictions) for fuller completeness.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so the description must compensate entirely. It provides clear, detailed explanations for each parameter, including possible values for 'breach_type' (confidentiality, integrity, availability). This adds substantial meaning beyond the bare schema, enabling correct invocation.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description specifies the tool's purpose: assessing breach severity and notification requirements under GDPR Articles 33-34, and generating notification content. It clearly differentiates from siblings like 'classify_processing' or 'dpia_generator' by focusing on breach-specific legal obligations.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explicitly states when to use the tool (for breach assessment under GDPR) and references the 72-hour rule. However, it does not mention when to avoid using it or suggest alternative sibling tools, which would enhance guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/CSOAI-ORG/gdpr-compliance-ai-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server