Red-Team NDA, OPSEC & PII Security Auditor — Zero-Leak Gate

    1

    Autonomous multi-round 6-perspective adversarial NDA, OPSEC, PII & build-output Red-Team audit by Dương Phúc Khánh (khanhdp.com).

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    About this skill

    When the user invokes /op-nda or /nda-audit (or asks for a multi-agent NDA / OPSEC cross-review), execute this end-to-end 4-Phase, 3-Round Adversarial Red-Team Swarm without stopping until every lexical, metadata, infrastructure, PII, and semantic business-logic leak is eliminated, hardened into pre-commit verification gates, and deployed live (SUCCESS ✓).

    1. Dynamic Ground-Truth NDA Extraction:
      • Before spawning subagents, run a fast local discovery script across ~/ directory names and ~/profile/journal/daily/*.md to extract the live ground-truth inventory of:
        • Private repository and folder names (~/client-erp, ~/fintech-core, ~/internal-gateway, ~/quant-engine, etc.).
        • Internal Jira/Linear ticket prefixes (CORP-*, FIN-*, INFRA-*, SEC-*, CORE-101, etc.).
        • Private/client domains (*.internal-client.com, *.staging-corp.dev, api.private-fintech.io, etc.).
    2. Single-Batch Dispatch of 6 Adversarial Red-Team Subagents (Model: "flash"):
      • Combine maximum-parallel-subagents + doubt-driven-development (adversarial framing: "Assume previous sanitization missed subtle, camelCase, hyphenated, or semantic leaks; find issues, do not validate").
      • Dispatch all 6 subagents concurrently in Read-Only Forensics Mode across ~/kido-blog, ~/kido-chronicles, and ~/profile/web:
    Subagent RoleAdversarial PersonaTarget Vectors & Scope
    #redteam-1-corporate-ndaCorporate Legal & NDA CounselAll exact, substring, camelCase, and hyphenated variants of employers, clients, internal products (<EmployerName>, <ClientBrand>, <InternalProduct>, <ProprietaryCodename>), and Vietnamese banks/brokers/payment gateways (Techcombank, Vietcombank, MBBank, BIDV, VPBank, TPBank, ACB, Sacombank, SSI, VNDirect, Vietcap, DNSE, FireAnt, CafeF, Vietstock, Casso, SePay, PayOS).
    #redteam-2-git-tickets-commitsGit Forensics & Issue Tracker InvestigatorJira/Linear ticket IDs (\b[A-Z]{2,10}-\d{1,5}\b), raw commit SHA hashes (\b[0-9a-f]{7,40}\b), prose quoting commits (commit ngày, commits across, git log, PR #, initial commit), and internal file/directory paths inside code blocks.
    #redteam-3-pii-people-identityHR Privacy & PII InvestigatorAny real names, nicknames, handles, email addresses, Slack channel/user IDs (C09*, U09*), Telegram links, or Discord webhooks of teammates, employees, bosses, or customers (<ColleagueName>, <InternalHandle>, <ManagerName>, <ClientContact>, etc.). Only the author (Dương Phúc Khánh / Kido) and famous public computer scientists may appear.
    #redteam-4-infra-sec-ipsOffensive Infra & Secrets PentesterAll IPv4/IPv6 addresses, VPS/Cloud/CDN providers (Contabo, OVH, Hetzner, DigitalOcean, Vultr, Linode, UpCloud, BunnyCDN, b-cdn.net), workstation/VPS paths (~/, /opt/, /var/www/), internal Docker container names, DB connection strings, JWTs, API keys, and non-canonical external domains.
    #redteam-5-slugs-seo-metadataTechnical SEO & Slug Hygiene AuditorAll URL slugs (<slug>.md), frontmatter (title, description, category, tags, metrics, relatedEpisodes), catalog.json, sitemap*.xml, and llms.txt. Hunts for date-suffixed slugs (-YYYY-MM-DD.md), truncated trailing words from word-count slicing (e.g. -12, -multi, -job, -read), typos, or awkward regex-replacement artifacts (shortlink-core, AlphaBank, CORE-283).
    #redteam-6-semantic-business-logicCompetitive Intelligence AnalystDeep semantic reading of high-risk articles (Anti-detect browsers, Bank scrapers/webhooks, Shortlink monetization, Stock/FinTech pipelines, POS/HR ERPs) to detect proprietary database schemas, enum values, or business rules that could identify an employer/client even after brand names are removed.
    1. Synthesize the Consolidated NDA & OPSEC Risk Matrix:
      • Classify every finding from the 6 Red-Team subagents by severity:
        • 🔴 CRITICAL: Direct employer/client/project name, real colleague/employee PII, real IP/secret, or unmistakable proprietary business workflow.
        • 🟠 HIGH: Raw git commit/ticket artifact (PWLP-*, CORE-283, commit ngày ...), bank/VPS provider name, or internal repo path.
        • 🟡 MEDIUM: Truncated URL slug (-12, -multi), awkward mechanical replacement phrase, or splintered category taxonomy.
    2. Parallel Root-Cause Remediation:
      • Lexical & Metadata Layer: Fix all flagged filenames (git mv), frontmatter fields, and prose tokens across both VI and EN editions simultaneously (maintaining 100% 1-to-1 parity).
      • Semantic Layer: Rewrite any section that exposes sensitive domain mechanics (e.g. browser fingerprint spoofing, specific bank portal scraping, internal payroll/attendance rules) into universal, white-hat Big-Tech Distributed Systems patterns (e.g., Chromium Sandboxed Worker Isolation, ISO-8583 / OpenBanking Idempotent Webhook Ingestion, Distributed Token-Bucket Rate Limiting).

    Open docs

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 3 days ago

    • Free with an Agensi account

    Requires

    Git

    Permissions

    Terminal / Shell
    Network Access
    Listed3 days ago

    What's inside

    Frequently Asked Questions

    Autonomous multi-round 6-perspective adversarial NDA, OPSEC, PII & build-output Red-Team audit by Dương Phúc Khánh (khanhdp.com).

    This skill uses the open SKILL.md standard. Use it through an MCP-compatible agent connected to Agensi, and check the compatibility information on this page for confirmed agents and required capabilities.

    Connect your AI agent to Agensi through MCP, then ask it to use Red-Team NDA, OPSEC & PII Security Auditor — Zero-Leak Gate on your task. Your agent loads the latest approved version each time. Paid skills require a purchase or active subscription. Skill files are not available for download; your licence does not allow you to save, copy or export any part of a skill.

    New in Security

    Free