group-cases-v2

name	group-cases-v2
description	Use when executing advanced multi-factor case grouping and relationship clustering.
category	investigation
version	1.0.0
type	Skill
title	Skill: Group Cases v2
generated	{"by":"process:google-labs-jules","at":"2025-12-21T03:04:42.000Z"}

Runbook: Group Cases v2

Objective

To analyze a defined set of recent SOAR cases, identify logical groupings based on shared entities or alert types, prioritize these groups by potential impact or severity, and generate a summary report of the findings. This helps in understanding related security events and focusing investigative efforts.

Scope

This runbook covers:

• Retrieving a list of recent SOAR cases.
• Gathering detailed information (including alerts and key entities) for each case.
• Analyzing case data to identify logical groupings based on specified or observed criteria.
• Prioritizing the identified case groups.
• Optionally performing basic enrichment of key shared entities within high-priority groups.
• Generating a Markdown report summarizing the case groups, prioritization rationale, and key findings.

This runbook explicitly excludes:

• Deep investigation of individual cases within the groups (this would typically follow for high-priority groups).
• Containment or eradication actions.
• Automated closure of grouped cases (though findings might inform manual closure decisions).

Inputs

• (Optional) ${NUMBER_OF_CASES}: Number of recent cases to analyze (e.g., 5, 10). Defaults to a predefined number (e.g., 10) if not specified.
• (Optional) ${TIME_FRAME_HOURS}: Lookback period in hours for selecting cases. If not provided, NUMBER_OF_CASES will be the primary filter.
• (Optional) ${GROUPING_CRITERIA}: Specific criteria for grouping cases (e.g., "shared_hostname", "alert_type", "CVE", "malware_family"). If not provided, grouping will be based on observed similarities in entities and alert details.
• (Derived) ${CASE_LIST}: List of SOAR case IDs selected for analysis.
• (Derived) ${CASE_DETAILS_MAP}: A map or structure holding details (alerts, entities) for each case in ${CASE_LIST}.
• (Derived) ${CASE_GROUPS}: Identified groups of related case IDs.
• (Derived) ${PRIORITIZED_GROUPS}: The ${CASE_GROUPS} ordered by assessed priority.
• (Derived) ${ENRICHMENT_DATA_SUMMARY}: (Optional) Summary of enrichment for key entities in high-priority groups.

Outputs

• ${REPORT_FILE_PATH}: The full path to the generated Markdown summary report.
• ${REPORT_CONTENT}: The full Markdown content of the generated report.
• ${GROUPING_ANALYSIS_SUMMARY}: A brief textual summary of how cases were grouped and prioritized.

Tools

• secops-soar: list_cases, get_case_full_details, list_alerts_by_case, get_entities_by_alert_group_identifiers
• secops-mcp: lookup_entity
• gti-mcp: Relevant enrichment tools (e.g., get_ip_address_report, get_domain_report)
• write_to_file (Replaces write_report)

Workflow Steps & Diagram

1. List Cases: Retrieve recent cases using soar-mcp_list_cases, filtered by ${NUMBER_OF_CASES} or ${TIME_FRAME_HOURS}. Store in ${CASE_LIST}.
2. Gather Case Details: For each case ID in ${CASE_LIST}:
   • Use soar-mcp_get_case_full_details to get overall case information.
   • Use soar-mcp_list_alerts_by_case to get associated alerts.
   • Use soar-mcp_get_entities_by_alert_group_identifiers (if applicable, or parse entities from alerts/events) to extract key entities.
   • Store all details in ${CASE_DETAILS_MAP}.
3. Group Cases: Analyze entities and alert details across all cases in ${CASE_DETAILS_MAP}. Identify logical groups (${CASE_GROUPS}) based on ${GROUPING_CRITERIA} (if provided) or observed similarities (e.g., shared critical entities, common alert types, overlapping timeframes).
4. Prioritize Groups: Assess the priority of each group in ${CASE_GROUPS} based on factors like combined alert severity, number of cases in the group, criticality of shared entities, or potential impact. Store as ${PRIORITIZED_GROUPS}.
5. Enrich Key Entities (Optional): For high-priority groups in ${PRIORITIZED_GROUPS}, identify key shared entities. Perform basic enrichment on these entities using secops-mcp_lookup_entity and relevant gti-mcp tools. Store in ${ENRICHMENT_DATA_SUMMARY}.
6. Generate Summary Report: Create a Markdown report (${REPORT_CONTENT}) summarizing the ${PRIORITIZED_GROUPS}, the rationale for grouping and prioritization, and key findings (including ${ENRICHMENT_DATA_SUMMARY} if available). Use write_to_file to save the report to ${REPORT_FILE_PATH} (e.g., ./reports/case_grouping_report_${timestamp}.md).

ADK Graph-Based Workflow Diagram

graph TD
    START(["START"]) --> extract_group_v2_payload_node["1. extract_group_v2_payload_node<br/><i>(Extract Environment & Similarity Threshold Payload)</i>"]
    extract_group_v2_payload_node --> compute_v2_case_clusters_node["2. compute_v2_case_clusters_node<br/><i>(Compute Entity & Alert Similarity Clusters)</i>"]
    compute_v2_case_clusters_node --> group_cases_v2_router{"3. group_cases_v2_router<br/><i>(Event.actions.route)</i>"}

    group_cases_v2_router -- "MERGE_HIGH_SIMILARITY_CASES" --> handle_merge_high_similarity_branch["4a. handle_merge_high_similarity_branch<br/><i>(Consolidate High Similarity Case Clusters)</i>"]
    group_cases_v2_router -- "NO_MERGE_REQUIRED" --> handle_no_merge_required_branch["4b. handle_no_merge_required_branch<br/><i>(Document Case Groupings)</i>"]

    handle_merge_high_similarity_branch --> document_group_v2_report_node["5. document_group_v2_report_node<br/><i>(SOAR Comment & Report Summary)</i>"]
    handle_no_merge_required_branch --> document_group_v2_report_node

Sequence Diagram

sequenceDiagram
    participant Analyst/User
    participant AutomatedAgent as Automated Agent (MCP Client)
    participant SOAR as secops-soar
    participant SIEM as secops-mcp
    participant GTI as gti-mcp

    Analyst/User->>AutomatedAgent: Start Group Cases v2 Workflow\nInput: NUMBER_OF_CASES (opt), TIME_FRAME_HOURS (opt), GROUPING_CRITERIA (opt)

    %% Step 1: List Cases
    AutomatedAgent->>SOAR: list_cases(limit=NUMBER_OF_CASES, time_frame_hours=TIME_FRAME_HOURS)
    SOAR-->>AutomatedAgent: List of Cases (CASE_LIST: C1, C2...)

    %% Step 2: Gather Details
    Note over AutomatedAgent: Initialize CASE_DETAILS_MAP
    loop For each Case Ci in CASE_LIST
        AutomatedAgent->>SOAR: get_case_full_details(case_id=Ci)
        SOAR-->>AutomatedAgent: Details for Ci
        AutomatedAgent->>SOAR: list_alerts_by_case(case_id=Ci)
        SOAR-->>AutomatedAgent: Alerts for Ci
        AutomatedAgent->>SOAR: get_entities_by_alert_group_identifiers(case_id=Ci, ...)
        SOAR-->>AutomatedAgent: Entities for Ci
        Note over AutomatedAgent: Store all details in CASE_DETAILS_MAP[Ci]
    end

    %% Step 3 & 4: Group & Prioritize
    Note over AutomatedAgent: Analyze CASE_DETAILS_MAP based on GROUPING_CRITERIA or similarities.
    Note over AutomatedAgent: Form CASE_GROUPS (G1, G2...).
    Note over AutomatedAgent: Prioritize groups into PRIORITIZED_GROUPS.

    %% Step 5: Enrich (Optional)
    opt Enrich High Priority Groups
        Note over AutomatedAgent: Initialize ENRICHMENT_DATA_SUMMARY
        loop For each High Priority Group Gp in PRIORITIZED_GROUPS
            Note over AutomatedAgent: Identify key shared entities (Ep1, Ep2...)
            loop For each Entity Epi in Gp
                AutomatedAgent->>SIEM: lookup_entity(entity_value=Epi)
                SIEM-->>AutomatedAgent: SIEM Summary for Epi
                AutomatedAgent->>GTI: get_..._report(ioc=Epi) %% Appropriate GTI tool
                GTI-->>AutomatedAgent: GTI Enrichment for Epi
                Note over AutomatedAgent: Store in ENRICHMENT_DATA_SUMMARY
            end
        end
    end

    %% Step 6: Generate Report
    Note over AutomatedAgent: Synthesize findings into REPORT_CONTENT (Markdown)
    AutomatedAgent->>AutomatedAgent: write_to_file(path="./reports/case_grouping_report_${timestamp}.md", content=REPORT_CONTENT)
    Note over AutomatedAgent: Report file created (REPORT_FILE_PATH)
    Note over AutomatedAgent: Prepare GROUPING_ANALYSIS_SUMMARY

    AutomatedAgent->>Analyst/User: attempt_completion(result="Case grouping analysis complete. Report: REPORT_FILE_PATH. Summary: GROUPING_ANALYSIS_SUMMARY.")

Completion Criteria

• A list of recent SOAR cases (${CASE_LIST}) has been retrieved based on the specified criteria.
• Detailed information, including alerts and key entities, has been gathered for each case in the list and stored (${CASE_DETAILS_MAP}).
• Cases have been analyzed and grouped into logical clusters (${CASE_GROUPS}) based on defined or observed criteria.
• The identified case groups have been prioritized (${PRIORITIZED_GROUPS}).
• (Optional) Key shared entities within high-priority groups have undergone basic enrichment, and a summary (${ENRICHMENT_DATA_SUMMARY}) is available.
• A comprehensive Markdown summary report (${REPORT_CONTENT}), detailing the groups, prioritization, and key findings, has been generated and saved to ${REPORT_FILE_PATH}.
• A ${GROUPING_ANALYSIS_SUMMARY} is available.

Rubrics

The following rubric is used to evaluate the execution of this Triage/Response runbook by an LLM agent.

Grading Scale (0-100 Points)

Criteria	Points	Description
Context & Enrichment	25	Correctly extracted entities and enriched them with relevant context (GTI, SIEM).
Analysis & Decision	25	Analyzed the enriched data to make a sound decision (FP/TP, Escalate/Close).
Action Execution	20	Performed the required response actions (e.g., isolation, containment) correctly.
Documentation	15	Clearly documented findings and actions in the case/ticket.
Operational Artifacts	15	Produced required artifacts: Sequence diagram, execution metadata (date/cost), and summary.

Evaluation Criteria Details

1. Context & Enrichment (25 Points)

• 10 pts: Accurately extracted key entities (IPs, users, hashes) from the input.
• 15 pts: Performed necessary enrichment (e.g., enrich_ioc) to gather reputation and history.

2. Analysis & Decision (25 Points)

• 15 pts: Interpreted the context correctly to determine the nature of the alert.
• 10 pts: Reached a logical conclusion or next step (e.g., "Escalate to Tier 2" or "Isolate Host").

3. Action Execution (20 Points)

• 10 pts: Called the correct tools to perform response actions (if applicable) or investigative steps.
• 10 pts: Verified the success of actions or handled errors appropriately.

4. Documentation (15 Points)

• 15 pts: Posted a comprehensive comment or update to the SOAR case summarizing the triage.

5. Operational Artifacts (15 Points)

• 5 pts: Sequence Diagram: Produced a Mermaid sequence diagram visualizing the steps taken.
• 5 pts: Execution Metadata: Recorded the date, duration, and estimated token cost.
• 5 pts: Summary Report: Generated a concise summary of the actions and outcomes.