القائمة
Pre-deploy security assessment — validates against OWASP Top 10 and project standards
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Convert PDF/EPUB library to Markdown and generate Obsidian MOC notes
Hook-based compaction suggestions at logical task boundaries
Context window management — track spend, decide when to compact, preserve state
Session-start orientation — loads context, surfaces learnings, confirms registry
Quality and semantic review — catches what automated tools miss
Planner → Architect → Critic deliberation loop — produces a formally validated ADR
| name | security-gate |
| description | Pre-deploy security assessment — validates against OWASP Top 10 and project standards |
| version | 0.2.0 |
| level | 3 |
| triggers | ["security check","security gate","pre-deploy review","before deploying","security audit"] |
| context_files | ["context/security-standards.md"] |
| steps | [{"name":"Scope","description":"Identify files and modules in scope. List them explicitly."},{"name":"Injection Scan","description":"Check all database queries, shell calls, template rendering for injection vectors"},{"name":"Secrets Scan","description":"Check for hardcoded credentials, tokens, keys, passwords in any literal form"},{"name":"Auth Check","description":"Verify authentication and authorization are enforced — not just at login"},{"name":"Input Validation","description":"Verify all external inputs are validated at system boundaries"},{"name":"Output Sanitization","description":"Verify error messages and logs do not leak sensitive data"},{"name":"Dependency Review","description":"Flag unpinned dependencies and known CVEs"},{"name":"Standards Comparison","description":"Compare findings against context/security-standards.md"},{"name":"Verdict","description":"PASS / CONDITIONAL PASS (with required fixes) / FAIL (blocking issues found)"}] |
Pre-deployment security assessment. Run before code reaches production or a shared environment.
Without an explicit security gate, Claude applies security review inconsistently — thorough when it seems relevant, skipped when under pressure to finish. The most dangerous vulnerabilities are in code that looked safe at a glance.
For each category, actively look for it — do not assume absence without checking:
A01 — Broken Access Control
A02 — Cryptographic Failures
A03 — Injection
A04 — Insecure Design
A05 — Security Misconfiguration
A06 — Vulnerable Components
A07 — Authentication Failures
A08 — Integrity Failures
A09 — Logging Failures
A10 — SSRF
PASS: No issues found in scope. State scope explicitly.
CONDITIONAL PASS:
Issue: [description]
Location: [file:line]
Severity: LOW/MEDIUM
Fix: [specific remediation]
Blocking: No — fix before next release
FAIL:
Issue: [description]
Location: [file:line]
Category: [OWASP category]
Attack vector: [how it could be exploited]
Fix: [specific remediation]
Blocking: YES — do not deploy
Do not issue a PASS without checking every category. A fast PASS is a false PASS.
Do not scope-creep into architecture review. Report the finding and the fix category; do not redesign the system.
Do not combine security-gate with code-review. They are separate passes for a reason.