界面已本地化;技能介绍与说明文档按所选语言由 AI 自动翻译,代码、命令、路径和专业术语保留原文,以原作者文档为准。
探索技能
数据与研究 · SKILL.md

performing-osint-with-spiderfoot

performing-osint-with-spiderfoot

【数据与研究】Automate OSINT collection using SpiderFoot REST API and CLI for target profiling, module-based reconnaissance, and structured result analysis across 200+ data sources

独立 GitHub 快照 · 自动分类 · 以原作者文档为准

技能说明

GitHub

name: performing-osint-with-spiderfoot description: Automate OSINT collection using SpiderFoot REST API and CLI for target profiling, module-based reconnaissance, and structured result analysis across 200+ data sources domain: cybersecurity subdomain: threat-intelligence tags:

  • osint
  • spiderfoot
  • reconnaissance
  • threat-intelligence
  • attack-surface
  • target-profiling version: "1.0" author: mahipal license: Apache-2.0

Performing OSINT with SpiderFoot

Overview

SpiderFoot is an open-source OSINT automation tool with 200+ modules that integrates with data sources for threat intelligence and attack surface mapping. This skill uses the SpiderFoot REST API and CLI (sf.py/spiderfoot-cli) to create and manage scans, select modules by use case (footprint, investigate, passive), parse structured results for domains, IPs, email addresses, leaked credentials, and DNS records, and generate target intelligence profiles.

Prerequisites

  • SpiderFoot 4.0+ installed or SpiderFoot HX cloud account
  • Python 3.8+ with requests library
  • SpiderFoot server running on default port 5001
  • Optional: API keys for VirusTotal, Shodan, HaveIBeenPwned modules

Steps

  1. Connect to SpiderFoot REST API or use CLI interface
  2. Create a new scan with target specification (domain, IP, email, name)
  3. Select scan modules by use case (all, footprint, investigate, passive)
  4. Monitor scan progress via API polling
  5. Retrieve and parse scan results by data element type
  6. Extract key findings: subdomains, IPs, emails, leaked credentials
  7. Generate structured OSINT intelligence report

Expected Output

JSON report containing OSINT findings organized by data type (domains, IPs, emails, credentials, DNS records), module source attribution, and target profile summary with risk indicators.

我的收藏