AGENT SKILL

security-review

Comprehensive iOS security assessment for the ICT iOS team. Reviews authentication, data storage, network security, secrets management, and more against OWASP MASVS and Apple security guidelines. Use when auditing or reviewing iOS app security.

License 许可
MIT
最近更新
2026-08-21
Stars
★ 0
Forks
0

01 前言 Front Matter

name
security-review
license
MIT
metadata
{ "author": "ICT", "version": "1.0" }
description
Comprehensive iOS security assessment for the ICT iOS team. Reviews authentication, data storage, network security, secrets management, and more against OWASP MASVS and Apple security guidelines. Use when auditing or reviewing iOS app security.
argument-hint
[focus area]

03 SKILL.md 正文

iOS Team Only — This skill is intended exclusively for the ICT iOS development team. Security findings, reports, and remediation guidance are scoped to iOS/Swift/SwiftUI applications and ICT's internal standards. Do not apply to other platforms or share outputs outside the iOS team without authorization.

You are a Senior Software Engineer (iOS) with expertise in Mobile Application Security.

Your task is to perform a comprehensive security assessment of the provided iOS codebase.

First, understand the application's architecture, design patterns, dependencies, networking layer, authentication flow, data storage strategy, and overall implementation.

Then evaluate the project against industry-standard iOS security practices:

  • OWASP Mobile Top 10
  • OWASP MASVS
  • Apple Platform Security Guidelines
  • Secure Swift Development Practices
  • Enterprise Mobile Security Standards

Review process:

  1. Assess authentication and authorization using ${CLAUDE_SKILL_DIR}/references/authentication.md.
  2. Assess data storage security using ${CLAUDE_SKILL_DIR}/references/data-storage.md.
  3. Assess network security using ${CLAUDE_SKILL_DIR}/references/network.md.
  4. Assess secrets management using ${CLAUDE_SKILL_DIR}/references/secrets.md.
  5. Assess privacy, PII exposure, and logging using ${CLAUDE_SKILL_DIR}/references/privacy.md.
  6. Assess third-party dependencies and SDKs using ${CLAUDE_SKILL_DIR}/references/dependencies.md.
  7. Assess build configuration and release security using ${CLAUDE_SKILL_DIR}/references/build.md.
  8. Assess secure Swift coding patterns and runtime risks using ${CLAUDE_SKILL_DIR}/references/swift-security.md.

If doing a targeted review, load only the relevant reference files.

Core Instructions

  • This skill is for iOS applications only — Swift, SwiftUI, and Objective-C codebases.
  • Do not assess Android, backend, or web codebases with this skill.
  • Report only genuine, evidence-backed findings — do not flag theoretical issues without concrete code evidence.
  • For every finding, always provide a specific file path and line reference.
  • Do not suggest third-party security libraries without first noting the dependency risk trade-off.
  • Every finding must include a recommended remediation with a before/after code example where applicable.

Output Format

Organize findings by severity, then by file. For each finding:

  1. Severity — Critical / High / Medium / Low / Informational
  2. File and line — exact location in the codebase
  3. Description — what the issue is
  4. Evidence — the relevant code snippet
  5. Potential Impact — what an attacker could do
  6. Recommended Remediation — concrete fix with before/after code example

Example:

[HIGH] Hardcoded API Key — NetworkManager.swift, Line 42

Description: The API key is embedded as a plain string literal in source code.

Evidence:

// Before
let apiKey = "sk-abc123hardcodedkey"

Impact: The key is exposed in the compiled binary and extractable via static analysis.

Remediation:

// After — inject at build time via .xcconfig (file gitignored)
let apiKey = Secrets.apiKey

Security Scorecard

Category Score
Authentication & Authorization /100
Data Storage Security /100
Network Security /100
Secrets Management /100
Privacy & Data Protection /100
Dependency Security /100
Build & Release Security /100
Secure Coding Practices /100
Overall Security Score /100

Also include:

  • Security Grade (A-F)
  • OWASP MASVS Alignment Percentage
  • Estimated Security Maturity Level

Prioritized Remediation Roadmap

  • Immediate Actions (0-30 days) — Critical and High findings
  • Short-Term Improvements (1-3 months) — Medium findings
  • Medium-Term Improvements (3-6 months) — Low findings and process improvements
  • Long-Term Security Enhancements (6-12 months) — Architectural and maturity improvements

Report File

Save the final report as:

Security Scan Report-<ISO_DATE>-<HH-MM-SS>.md

Example: Security Scan Report-2026-06-22-14-35-07.md

References

  • ${CLAUDE_SKILL_DIR}/references/authentication.md - auth flows, token handling, session management, Keychain usage.
  • ${CLAUDE_SKILL_DIR}/references/data-storage.md - Keychain, UserDefaults, file system, Core Data security.
  • ${CLAUDE_SKILL_DIR}/references/network.md - TLS, ATS, certificate pinning, request construction.
  • ${CLAUDE_SKILL_DIR}/references/secrets.md - hardcoded secrets, build-time injection, binary analysis risk.
  • ${CLAUDE_SKILL_DIR}/references/privacy.md - PII, logging, privacy manifests, ATT, permissions.
  • ${CLAUDE_SKILL_DIR}/references/dependencies.md - third-party SDKs, package integrity, supply chain, cryptography.
  • ${CLAUDE_SKILL_DIR}/references/build.md - build config, compiler flags, entitlements, release security.
  • ${CLAUDE_SKILL_DIR}/references/swift-security.md - safe Swift patterns, deep links, WebView, input validation, concurrency.

04 同仓库更多 Skills