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 pivot-on-ioc-gti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 | pivot-on-ioc-gti |
| description | Use when pivoting on GTI campaign identifiers, threat actors, and related infrastructure. |
| category | common |
| version | 1.0.0 |
| type | Skill |
| title | Skill: Common Step: Pivot on IOC using GTI Relationships |
| generated | {"by":"human:dandye","at":"2025-11-10T22:21:35.000Z"} |
Explore relationships connected to a specific Indicator of Compromise (IOC) within Google Threat Intelligence (GTI) to discover related entities (other IOCs, threats, etc.).
This sub-runbook executes the appropriate gti-mcp_get_entities_related_to_a_... tool based on the input IOC type and desired relationship(s). It returns the discovered related entities.
${IOC_VALUE}: The specific IOC value (e.g., "198.51.100.10", "evil-domain.com", "abcdef123456...", "http://bad.url/path").${IOC_TYPE}: The type of IOC ("IP Address", "Domain", "File Hash", "URL", "Collection"). Note: Added "Collection" type.${RELATIONSHIP_NAMES}: A list of relationship names to query (e.g., ["resolutions", "communicating_files"], ["malware_families", "attack_techniques"]). The available relationships depend on the ${IOC_TYPE}.${RELATED_ENTITIES}: A structured dictionary or list containing the entities found for each queried relationship.${PIVOT_STATUS}: Confirmation or status of the pivoting attempt(s).gti-mcp: get_entities_related_to_an_ip_address, get_entities_related_to_a_domain, get_entities_related_to_a_file, get_entities_related_to_an_url, get_entities_related_to_a_collection${IOC_VALUE}, ${IOC_TYPE}, and ${RELATIONSHIP_NAMES} from the calling runbook. Initialize ${RELATED_ENTITIES} as an empty structure.${IOC_TYPE}, select the correct gti-mcp tool (e.g., get_entities_related_to_an_ip_address for "IP Address").relationship in ${RELATIONSHIP_NAMES}.
ip_address, domain, hash, url, id) set to ${IOC_VALUE} and relationship_name=relationship.relationship key within ${RELATED_ENTITIES}.${PIVOT_STATUS} based on the success/failure of the API calls. Return ${RELATED_ENTITIES} and ${PIVOT_STATUS} to the calling runbook.sequenceDiagram
participant CallingRunbook
participant PivotOnIOC as pivot_on_ioc_gti.md (This Runbook)
participant GTI as gti-mcp
CallingRunbook->>PivotOnIOC: Execute GTI Pivot\nInput: IOC_VALUE, IOC_TYPE, RELATIONSHIP_NAMES
%% Step 2: Determine Tool
Note over PivotOnIOC: Select GTI tool based on IOC_TYPE
%% Step 3: Iterate Relationships
loop For each relationship R in RELATIONSHIP_NAMES
PivotOnIOC->>GTI: [Selected Tool](identifier=IOC_VALUE, relationship_name=R)
GTI-->>PivotOnIOC: Related Entities List for R
Note over PivotOnIOC: Store results in RELATED_ENTITIES[R]
end
%% Step 4: Return Results
Note over PivotOnIOC: Set PIVOT_STATUS
PivotOnIOC-->>CallingRunbook: Return Results:\nRELATED_ENTITIES,\nPIVOT_STATUS
The relevant GTI relationship queries have been attempted for the specified IOC. The discovered related entities (${RELATED_ENTITIES}) and status (${PIVOT_STATUS}) are available.