用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/dandye/adk_runbooks --skill prioritize-and-investigate-case命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
Use when evaluating threat detection opportunities (TDOs), generating synthetic UDM events, evaluating Chronicle rule coverage, and drafting YARA-L 2.0 rules.
Use when hunting for advanced persistent threat (APT) actor activity, tools, and infrastructure.
Use when hunting for lateral movement via PsExec, WMI, remote scheduled tasks, or WinRM.
正在显示 SKILL.md
| name | prioritize-and-investigate-case |
| description | Use when scoring case urgency, prioritizing the queue, and running core investigations. |
| category | investigation |
| version | 1.0.0 |
| type | Skill |
| title | Skill: Prioritize and Investigate a Case |
| generated | {"by":"process:google-labs-jules","at":"2026-08-20T02:00:00Z"} |
From a list of cases, identify cases of the highest severity and potential impact based on underlying alerts and detections. Get rule logic to validate the detections in the cases. After identifying the highest N priority cases -> Explain the entirety of the case to the analyst in the context of the underlying rule logic (explain the rule logic and how it applies to this case). Get entity context to determine if there are additional alerts, detections, or events that may not have been included in the case but are potentially applicable.
Use the tools:
common_steps/find_relevant_soar_case </run_books/common_steps/find_relevant_soar_case>graph TD
START(["START"]) --> extract_prioritization_payload_node["1. extract_prioritization_payload_node<br/><i>(Extract Case Payload)</i>"]
extract_prioritization_payload_node --> compute_case_risk_score_node["2. compute_case_risk_score_node<br/><i>(Calculate Risk Score & Priority)</i>"]
compute_case_risk_score_node --> case_risk_router{"3. case_risk_router<br/><i>(Event.actions.route)</i>"}
case_risk_router -- "IMMEDIATE_ESCALATION" --> handle_immediate_escalation_branch["4a. handle_immediate_escalation_branch<br/><i>(Elevate Priority & Escalate)</i>"]
case_risk_router -- "STANDARD_TRIAGE" --> handle_standard_triage_branch["4b. handle_standard_triage_branch<br/><i>(Standard Triage Pathway)</i>"]
handle_immediate_escalation_branch --> document_prioritization_report_node["5. document_prioritization_report_node<br/><i>(SOAR Comment & Report Summary)</i>"]
handle_standard_triage_branch --> document_prioritization_report_node
sequenceDiagram
participant User
participant AutomatedAgent as Automated Agent (MCP Client)
participant SOAR as secops-soar
participant SIEM as secops-mcp
participant FindCase as skills/common/find-relevant-soar-case/SKILL.md
User->>AutomatedAgent: Prioritize and investigate cases
AutomatedAgent->>SOAR: list_cases()
SOAR-->>AutomatedAgent: List of cases (C1, C2... Priority P1, P2...)
Note over AutomatedAgent: Analyze cases, identify high priority (e.g., Case X based on initial priority/alerts)
AutomatedAgent->>SOAR: get_case_full_details(case_id=X)
SOAR-->>AutomatedAgent: Full details for Case X (alerts, comments, etc.)
Note over AutomatedAgent: Confirm priority based on full details. May use change_case_priority if needed.
AutomatedAgent->>SOAR: list_alerts_by_case(case_id=X)
SOAR-->>AutomatedAgent: Alerts for Case X (A1, A2...)
Note over AutomatedAgent: Initialize ALL_CASE_ENTITIES = set()
loop For each Alert Ai in Case X
AutomatedAgent->>SOAR: list_events_by_alert(case_id=X, alert_id=Ai)
SOAR-->>AutomatedAgent: Events for Alert Ai (containing rule_id, entities E1, E2...)
Note over AutomatedAgent: Add E1, E2... to ALL_CASE_ENTITIES
Note over AutomatedAgent: Extract rule_id from event/alert data
AutomatedAgent->>SIEM: list_security_rules(rule_id=rule_id)
SIEM-->>AutomatedAgent: Rule logic/definition for rule_id
AutomatedAgent->>SIEM: list_rule_detections(rule_id=rule_id)
SIEM-->>AutomatedAgent: Detections associated with rule_id
Note over AutomatedAgent: Analyze events/detections against rule logic
loop For each relevant Entity Ej in Events
AutomatedAgent->>SIEM: lookup_entity(entity_value=Ej)
SIEM-->>AutomatedAgent: Entity context for Ej
AutomatedAgent->>SIEM: search_security_events(text="Events involving entity Ej", hours_back=...)
SIEM-->>AutomatedAgent: Broader UDM events for Ej
end
end
Note over AutomatedAgent: Check for related SOAR cases using all identified entities
AutomatedAgent->>FindCase: Execute(Input: SEARCH_TERMS=list(ALL_CASE_ENTITIES), CASE_STATUS_FILTER="Opened")
FindCase-->>AutomatedAgent: Results: RELATED_SOAR_CASES
Note over AutomatedAgent: Synthesize findings, correlate rule logic with events/entities, include related cases
AutomatedAgent->>SOAR: post_case_comment(case_id=X, comment="Investigation Summary: Case X involves rule [Rule Name] triggered by events [...]. Entities [...] investigated. Related Cases: ${RELATED_SOAR_CASES}. Findings: [...]")
SOAR-->>AutomatedAgent: Comment confirmation
AutomatedAgent->>AutomatedAgent: attempt_completion(result="Completed investigation for Case X. Summary posted as comment.")
The following rubric is used to evaluate the execution of this Triage/Response runbook by an LLM agent.
| 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. |
enrich_ioc) to gather reputation and history.