Agent Skills: Security Skill

Security Skill

Comprehensive security domain for development environment validation, vulnerability scanning, and security compliance. Provides automated security workflows with intelligent routing based on user intent.

Auto-Activation Keywords

This skill activates automatically when you mention:

• General: security, secure, vulnerability, vulnerabilities
• Validation: validate security, security check, environment security
• Scanning: scan, security scan, vulnerability scan, dependency scan
• Encryption: encrypt, decrypt, GPG, PGP, secrets
• Signing: sign commits, SSH key, GPG key, signed commits
• Compliance: OWASP, audit, security audit, compliance
• Tools: safety, bandit, semgrep, security tools

Routing Logic

Based on user intent, this skill routes to appropriate workflows:

Environment Validation

Keywords: "validate security", "check security environment", "security setup", "GPG key", "SSH key"

• Simple validation → Use /security/validate-env workflow
• Comprehensive audit → Invoke security-auditor agent

Vulnerability Scanning

Keywords: "scan dependencies", "security scan", "vulnerability check", "safety check", "bandit"

• Dependency scanning → Use /security/scan workflow
• Full security audit → Invoke security-auditor agent

File Encryption

Keywords: "encrypt", "decrypt", "GPG encrypt", "protect secrets", "encrypt .env"

• File encryption/decryption → Use /security/encrypt workflow
• Secrets management review → Invoke security-auditor agent

Security Audit

Keywords: "security audit", "penetration test", "threat assessment", "vulnerability assessment"

• Complex security audit → Invoke security-auditor agent directly
• Compliance validation → Use workflows + agent for comprehensive review

Workflow Quick Reference

# Validate security environment
/security/validate-env [--verbose]

# Scan for vulnerabilities
/security/scan [--type=dependencies|code|all]

# Encrypt/decrypt files
/security/encrypt [file-path]
/security/decrypt [file-path]

Complex Task Delegation

For comprehensive security analysis, invoke the security-auditor agent via Task tool:

Use security-auditor agent when:
- Conducting full security audits across multiple components
- Threat modeling and risk assessment
- Compliance validation against security standards
- Penetration testing simulation
- Security architecture review

Supporting Context

• OWASP Top 10: See context/owasp-top-10.md for common vulnerabilities
• Security commands: See context/security-commands.md for comprehensive command reference
• Security standards: See /standards/security.md for project requirements

Integration Points

Agents

• security-auditor: Comprehensive security audits and vulnerability assessment
• test-engineer: Security testing integration (via testing/workflows/security.md)

MCP Tools

• mcp__zen-core__chat: Multi-model consensus for security decisions

Hooks

• Pre-commit: Security validation on file changes
• Post-tool-use: MCP usage tracking

Standards

• Security standards: /standards/security.md (GPG/SSH requirements, encrypted secrets)
• Git workflow: /standards/git-workflow.md (signed commits requirement)

Security Requirements Summary

Required for all projects:

• GPG key configured (for .env encryption)
• SSH key configured and loaded (for signed commits)
• Git signing enabled (commit.gpgsign = true)
• No secrets in repository (use encrypted .env files)
• Dependency scanning (safety check passes)
• Static analysis (bandit passes)

Git signing configuration:

# SSH signing (recommended)
git config --global gpg.format ssh
git config --global user.signingkey ~/.ssh/id_ed25519.pub
git config --global commit.gpgsign true

# Or GPG signing
git config --global user.signingkey <GPG_KEY_ID>
git config --global commit.gpgsign true

Dependency scanning:

# Check Python dependencies
poetry run safety check --full-report

# Static security analysis
poetry run bandit -r src

Secrets encryption:

# Encrypt .env file
gpg --symmetric --cipher-algo AES256 .env

# Decrypt .env file
gpg --decrypt .env.gpg > .env

Examples

Example 1: Validate security environment

User: "Can you check if my security environment is properly configured?"
→ Security skill auto-activates
→ Routes to /security/validate-env workflow
→ Validates GPG keys, SSH keys, Git signing, environment setup

Example 2: Scan for vulnerabilities

User: "Scan my project for security vulnerabilities"
→ Security skill auto-activates
→ Routes to /security/scan workflow
→ Runs safety check and bandit analysis
→ Reports vulnerabilities with remediation steps

Example 3: Comprehensive security audit

User: "Perform a comprehensive security audit of the authentication system"
→ Security skill auto-activates
→ Invokes security-auditor agent
→ Agent performs:
  - Threat modeling
  - Vulnerability assessment
  - Code review for security issues
  - Compliance validation
  - Risk-prioritized recommendations

Example 4: Encrypt sensitive file

User: "Encrypt my .env file with GPG"
→ Security skill auto-activates
→ Routes to /security/encrypt workflow
→ Encrypts file with AES256
→ Provides decryption instructions

This skill consolidates check-security-env skill and security commands into a unified security domain with intelligent routing.