Skip to content
Back to search

ai-runbooks

This repository provides security operations runbooks and role-based guides for AI-assisted cybersecurity workflows.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/dandye--ai-runbooks/badge.svg)](/listings/dandye--ai-runbooks)

HTML

<a href="/listings/dandye--ai-runbooks"><img src="/api/listings/dandye--ai-runbooks/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

1. Clone the repository:

   git clone <repository-url>
   cd dandye_ai_runbooks

2. The symlinks should already be set up. Verify they exist:

   ls -la .claude/rules_bank
   ls -la .clinerules/rules_bank
   ls -la .gemini/rules_bank

3. If symlinks are missing, create them manually:

   # For Claude Code
   mkdir -p .claude
   ln -s ../rules_bank .claude/rules_bank

   # For Cline
   mkdir -p .clinerules
   ln -s ../rules_bank .clinerules/rules_bank

   # For Gemini CLI
   mkdir -p .gemini
   ln -s ../rules_bank .gemini/rules_bank

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

readme.md · README · 241 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Agentic SOC Runbooks
  2. This repository provides security operations runbooks and role-based guides for AI-assisted cybersecurity workflows.
  3. ## Overview
  4. Built by and for enterprise security teams, this project standardizes incident response procedures and threat analysis workflows across multiple AI platforms including Claude, Gemini, and Cline. It equips LLMs with specialized context and cognitive tools to perform complex security investigations with the same rigor and consistency as experienced analysts.
  5. ## Key Components
  6. ### Personas
  7. Security role definitions including:
  8. - SOC Analysts (Tier 1-3)
  9. - Threat Hunters
  10. - Incident Responders
  11. - CTI Researchers
  12. - Security Engineers
  13. - And more...
  14. ### Runbooks
  15. Step-by-step procedural guides for:
  16. - Alert triage
  17. - IOC enrichment and investigation
  18. - Threat hunting
  19. - Incident response
  20. - Case management
  21. - And many more security operations tasks
  22. ### Incident Response Plans (IRPs)
  23. Comprehensive end-to-end strategies for:
  24. - Malware incidents
  25. - Phishing attacks
  26. - Ransomware
  27. - Compromised accounts
  28. ## Repository Structure
  29. ```
  30. ai_runbooks/
  31. ├── rules_bank/ # Master source directory for all content
  32. │ ├── personas/ # Security role definitions
  33. │ ├── run_books/ # Procedural guides for security tasks
  34. │ │ ├── common_steps/ # Reusable procedure components
  35. │ │ ├── irps/ # Incident Response Plans
  36. │ │ └── guidelines/ # Best practices and documentation
  37. │ └── [other documentation] # Configuration and reference files
  38. ├── reports/ # Generated security reports
  39. ├── mcp-security/ # MCP server implementations
  40. └── [configuration files] # Various setup and config scripts
  41. ```
  42. ## Usage
  43. 1. **For Security Teams**: Use this repository to standardize and document your security operations procedures for AI-assisted workflows.
  44. 2. **For Developers**:
  45. - Edit content only in the `rules_bank/` directory
  46. - Never edit files through the symlinks in AI tool directories
  47. - Use the provided Python scripts for configuration if needed
  48. 3. **For AI Assistants**: The symlinked `rules_bank` directory in each tool's configuration folder provides access to all personas, runbooks, and documentation.
  49. ## Integration
  50. This project is designed to work with:
  51. - Chronicle SIEM
  52. - SOAR platforms
  53. - Google Threat Intelligence (GTI)
  54. - Security Command Center (SCC)
  55. - Various MCP (Model Context Protocol) tools
  56. ## Contributing
  57. See `CONTRIBUTING` file for guidelines on contributing to this project.
  58. ## License
  59. This project is licensed under the Apache License 2.0 - see the `LICENSE` file for details.
  60. ---
  61. ## AI Tool Integration
  62. This repository is designed to work with three AI coding assistants:
  63. ### Claude Code (claude.ai/code)
  64. - Uses the `.claude/` directory for configuration
  65. - Reads context from `.claude/rules_bank/` symlink
  66. - See `CLAUDE.md` for Claude-specific guidance
  67. ### Cline
  68. - Uses the `.clinerules/` directory for configuration
  69. - Reads context from `.clinerules/rules_bank/` symlink
  70. - Follows the same runbook and persona system
  71. ### Gemini CLI
  72. - Uses the `.gemini/` directory for configuration
  73. - Reads context from `.gemini/rules_bank/` symlink
  74. - See `GEMINI.md` for Gemini-specific guidance
  75. - Compatible with the same documentation structure
  76. ### Quick Start for AI Tools
  77. 1. Clone the repository:
  78. ```bash
  79. git clone <repository-url>
  80. cd dandye_ai_runbooks
  81. ```
  82. 2. The symlinks should already be set up. Verify they exist:
  83. ```bash
  84. ls -la .claude/rules_bank
  85. ls -la .clinerules/rules_bank
  86. ls -la .gemini/rules_bank
  87. ```
  88. 3. If symlinks are missing, create them manually:
  89. ```bash
  90. # For Claude Code
  91. mkdir -p .claude
  92. ln -s ../rules_bank .claude/rules_bank
  93. # For Cline
  94. mkdir -p .clinerules
  95. ln -s ../rules_bank .clinerules/rules_bank
  96. # For Gemini CLI
  97. mkdir -p .gemini
  98. ln -s ../rules_bank .gemini/rules_bank
  99. ```
  100. ### Using with AI Tools
  101. Each AI tool will automatically read the context from its respective directory:
  102. - **Claude Code**: Reads from `.claude/`
  103. - **Cline**: Reads from `.clinerules/`
  104. - **Gemini CLI**: Reads from `.gemini/`
  105. All three tools access the same content through symlinks, ensuring consistency.
  106. ### Symlink Architecture
  107. The project uses a symlink-based system to share the same content across all three AI tools:
  108. 1. **Master Content**: All documentation, personas, and runbooks are maintained in `rules_bank/`
  109. 2. **Symlinks**: Each AI tool's configuration directory contains a symlink to `rules_bank/`
  110. 3. **Benefits**:
  111. - Single source of truth for all content
  112. - Updates automatically propagate to all AI tools
  113. - No duplication or synchronization needed
  114. ### Configuration Scripts
  115. - `set_persona_rules.py`: Configure active personas (legacy - may need updates for new structure)
  116. - `symlink_common_steps.py`: Manage common steps symlinks (legacy - may need updates)
  117. - `configure_mcp_tools.py`: Configure MCP (Model Context Protocol) tools for Claude Desktop
  118. ## Example Runbook Invocation Prompts
  119. Use these realistic prompts to invoke various security runbooks. All examples use real data extracted from the `reports/` directory:
  120. ### Alert Triage
  121. ```
  122. Please execute the alert triage runbook for SOAR case 2194.
  123. ```
  124. The case involves repeated authentication failures for user F.K. on host A.S.L from IP 10.1.0.4. Check for duplicate cases and perform initial entity enrichment.
  125. ### IOC Enrichment
  126. ```
  127. Run the basic IOC enrichment runbook for the suspicious domain scarfponcho.com.
  128. ```
  129. Threat intelligence analysis and SIEM correlation for a domain which was flagged in our environment.
  130. ```
  131. Execute IOC enrichment for the file hash de96a6e69944335375dc1ac238336066889d9ffc7d73628ef4fe1b1b160ab32c found on host WINS-D19.
  132. ```
  133. This was associated with PowerShell obfuscation activity by user LISAWALKER.
  134. ### Case Investigation
  135. ```
  136. Perform a comprehensive case investigation for SOAR case 3480.
  137. ```
  138. This is a critical priority case involving PowerShell attacks on WINS-D19, potential data exfiltration via BigQuery, and GCP exploitation attempts. Focus on the 17 associated alerts.
  139. ### Detection Rule Analysis
  140. ```
  141. Analyze detection rule ru_99d1f620-3fe2-41d5-918e-0e1bd2402065 (ursnif_malware_dns).
  142. ```
  143. I need rule validation, tuning recommendations, and analysis of recent detections for this high-severity rule.
  144. ### Threat Hunting
  145. ```
  146. Execute the IOC threat hunt runbook for IP addresses 46.8.9.205, 46.8.9.206, and 46.8.9.207.
  147. ```
  148. These IPs are associated with the malicious domain scarfponcho.com and need comprehensive SIEM analysis.
  149. ```
  150. Run the advanced threat hunting runbook focusing on lateral movement techniques (PSExec, WMI) around host WINS-D19.
  151. ```
  152. This follows the PowerShell attack activity discovered in case 3480.
  153. ### Malware Triage
  154. ```
  155. Perform malware triage for the sample with hash de96a6e69944335375dc1ac238336066889d9ffc7d73628ef4fe1b1b160ab32c.
  156. ```
  157. This was involved in the PowerShell obfuscation incident on WINS-D19.
  158. ### Suspicious Login Analysis
  159. ```
  160. Execute the suspicious login triage runbook for authentication events involving user F.K. on host A.S.L from source IP 10.1.0.4.
  161. ```
  162. This is related to case 2194 with multiple failed followed by successful logins.
  163. ### Incident Response
  164. ```
  165. Initiate the compromised user account IRP for user LISAWALKER'
  166. ```
  167. Evidence includes PowerShell execution with obfuscation techniques on host WINS-D19 and potential privilege escalation attempts.
  168. ### Cloud Security Analysis
  169. ```
  170. Run the cloud vulnerability triage runbook for the BigQuery exfiltration alerts in case 3480.
  171. ```
  172. Focus on the DLP context and Google Drive data movement patterns detected by SCC.
  173. ### Endpoint Isolation
  174. ```
  175. Execute the basic endpoint triage and isolation runbook for host WINS-D19.
  176. ```
  177. This system was compromised in case 3480 with PowerShell attacks and potential malware execution.
  178. ### Deep Dive Analysis
  179. ```
  180. Perform a deep dive IOC analysis on domain scarfponcho.com using GTI collections.
  181. ```
  182. Include malware family associations, campaign attribution, and infrastructure analysis.
  183. ### Summary
  184. These examples demonstrate how to invoke runbooks with specific, actionable intelligence that mirrors real security operations scenarios.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.