Security Engineering

Use when a SEED Unit has security constraints (authn/authz, tenant isolation, input validation, secrets, OWASP). Invoked by spec-driven-development / seed-unit when security constraints apply.

inapp-inc Updated

File contents

Security engineering

Ensure SEED Units and design artifacts include explicit security requirements, validation, and evidence.

Core rules

  • Authn/authz and tenant isolation are first-class when applicable.
  • Prefer deny by default; safe errors (no enumeration/leakage).
  • Require evidence: security tests, static checks, or reviewable proofs.
  • Close each touching slice with an OWASP Top 10 check (or N/A — <rationale> under Sprint Mode when not applicable).
  • If APIs change, ensure Docs/openapi.yaml reflects auth requirements.
  • Hard enforcement (manual PR operations): do not create/update/merge PRs.

Required outputs

  • Security constraints + AC + evidence on the SEED Unit block
  • Security decisions in design.md (threats, mitigations, residual risk)

Minimum checklist

  • Input validation & injection safety
  • Authn/authz correctness
  • Data isolation / least privilege
  • Secrets handling (env/config only)
  • Safe logging (no sensitive payloads)

inapp-inc/cursor-skills-foundry/tree/main/.cursor/skills/security-engineering commit f9e8169138

Frequently asked questions

npx skillmds@latest add inapp-inc/security-engineering