Skip to content

HOWTO: privacy-guardian — Data Masking Defense

docs/tools/HOWTO-PRIVACY-GUARDIAN.md

Standard: DSOM For My AI Protocol v6.1 | Secret Management Tools: privacy-guardian.ps1, privacy-guardian.sh, playbooks/dsom/privacy-scan.yml Platforms: Windows Native (T1), Linux/WSL2 (T2)


1. Purpose

privacy-guardian is the Security Sentinel of the DSOM ecosystem. Before any context manifest is shared with an external LLM (Gemini, Claude, GPT), this tool scans the file for common sensitive data patterns to prevent accidental data leaks or credential exposure.

Use it to detect: - Network Leaks: Plaintext IPv4 addresses. - Identity Leaks: Email addresses and local user home paths (e.g., C:\Users\Admin). - Cloud Secrets: AWS Access Keys, Google API Keys, GitHub Tokens, OpenAI Secrets, Slack Tokens. - Cryption Leaks: PEM-formatted Private Keys.

Location: - tools/privacy-guardian.ps1 (Windows) - tools/privacy-guardian.sh (Linux/WSL2) - playbooks/dsom/privacy-scan.yml (Ansible)


2. Prerequisites

Requirement Minimum Notes
Target File sod_manifest_*.txt Scans the today's manifest by default.
Tooling Regex Uses native grep (Linux) or Select-String (Windows).

3. Usage

3.1 Standard Scanning

# Windows (T1)

.\tools\privacy-guardian.ps1
# WSL2 / Linux (T2)

bash tools/privacy-guardian.sh

3.2 Ansible Sweep

ansible-playbook playbooks/dsom/privacy-scan.yml

4. Scan Patterns

The tool utilizes a library of high-fidelity regex patterns:

Pattern Type Scope
IPv4 ([0-9]{1,3}\.){3}[0-9]{1,3}
GitHub ghp_... (Tokens)
Google AIza... (API Keys)
OpenAI sk-... (API Keys)
Private Keys -----BEGIN ... PRIVATE KEY-----

5. Reading the Output & Status Codes

Label Meaning Action Needed
✅ SCAN COMPLETE No generic leaks found. Safe to upload to AI.
⚠️ POTENTIAL LEAK Found a match at a specific line. REDACT the manifest manually before upload.
❌ SCAN FAILED Security risk confirmed. DO NOT UPLOAD. Fix the source file or mask the manifest.

6. Security Advisory

[!IMPORTANT] This tool is a Heuristic Scanner. It detects patterns, not every possible secret.

  • Always perform a quick manual scroll-through of your manifest before uploading.
  • Never upload raw .env files or vault/secrets.yml to an external AI chat.
  • Utilize the Atomic Git Hygiene rule to ensure secrets never even enter your git history.

Document Purpose
HOWTO-REANIMATE.md Generator of the file this tool scans.
docs/REANIMATION-PROMPT-TEMPLATE.md Where the manifest is ultimately used.

Standard: DSOM For My AI Protocol v6.1 | Harisfazillah Jamel | LinuxMalaysia Document Version: v1.0 | Created: 2026-04-08


Deep State of Mind (DSOM) For My AI Protocol | Harisfazillah Jamel (LinuxMalaysia) | 2026-07-04 Standard: UK English | DBP-standard Bahasa Melayu Malaysia (Piawai) | GNU General Public License v3.0