mrdulasolutions/devlearn/devlearn-security

Claude Code and agent skill by mrdulasolutions/devlearn

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Teaches application security while fixing real issues: secrets, auth, XSS, injection, dependencies, HTTPS, error leakage. Use when user asks security review, auth hardening, "is this safe", secrets in code, OWASP, or before shipping sensitive features. Pairs with devlearn-pre-ship, devlearn-apis, devlearn-devops. Proactively suggest when diff touches auth, .env, user input, crypto, or payments. Voice triggers: "security review", "is this secure", "OWASP".

Install

npx vskill@latest install mrdulasolutions/devlearn/devlearn-security

SKILL.md

DevLearn: Security

Iron law

Never repeat secrets in lessons. Teach patterns; redact values. Rotate live leaked credentials immediately.

Voice

Follow ../shared/voice.md. Security scares vibers — start from what could go wrong for users, then fix.

Context

Not a pentest or compliance audit. Teach maintainable security habits while building: secrets out of git, validate on server, least privilege, dependency hygiene.

Lifecycle position: after pre-ship triage or when diff signals risk. See ../shared/ship-lifecycle.md.

Prerequisites

  • Running app or diff to review
  • devlearn-apis for auth/CORS/session details

Before you start

  1. Parse DEVLEARN.md lifecycle.security_pass
  2. Scan diff + .env* + auth routes + user input surfaces
  3. Redact any secret before quoting in chat or lessons

Phase 1: Triage

Scan for signals — full map in reference.md:

SignalRiskTeach
.env committedsecret leakenv + gitignore
API key in frontend bundlestolen keyserver-only
innerHTML + user textXSStextContent, sanitize
SQL string + user inputinjectionparameterized queries
No HTTPS in prodtamperingTLS
npm audit high/criticalsupply chainupdate/pin
Verbose errors in prodinfo leakgeneric user message
Missing auth on admin routesbroken access controlmiddleware

Output triage table: found / not found / n/a per category.

Phase 2: Fix + teach

Order: stop active leak → minimal fix → lesson block.

Viber lesson shape

What could go wrong for users → What we changed → Term (XSS, secret, token)

Seasoned decision shape

Decision → Alternatives (rate limit vs captcha) → Risk & verify

Append non-obvious choices to .devlearn/decisions.md when decisions: true.

One issue class per lesson block in vibe mode.

Phase 3: Pre-release security pass

Quick checklist (pairs with pre-ship):

  • [ ] No secrets in this commit or pasted in chat
  • [ ] Sensitive routes require auth
  • [ ] User input validated server-side
  • [ ] Prod errors don't expose stack traces to users
  • [ ] Dependencies audited (npm audit or equivalent)
  • [ ] HTTPS enforced on production URL
  • [ ] CORS not * with credentials (if applicable)

Document pass/fail in pre-ship checklist or standalone summary.

Phase 4: Auth teaching (on request)

ConceptPlain English
sessionserver remembers logged-in user via cookie
JWTsigned token client sends; server verifies signature
hash (bcrypt)password stored irreversibly
CORSbrowser cross-origin gate — not a substitute for auth
CSRFmalicious site triggers authenticated action — tokens/same-site help

Hand off HTTP/session implementation to devlearn-apis. Hand off pipeline secret storage to devlearn-devops.

Phase 5: Handoffs

SituationSuggest
Checklist complete, ready live/devlearn-deploy
Need CI audit gate/devlearn-devops
After deploy/devlearn-post-ship
User wants full release path/devlearn-pre-ship

Persona integration

PersonaTone
viberUser harm stories; one term; no fear mongering
seasonedThreat model lite; verify commands; tradeoffs
autodetectEscalate precision if user uses security vocabulary

Red flags — STOP

SituationAction
Live leaked credential in prodRotate first; then teach
Disable auth on prod "temporarily"Warn; dev-only pattern
Store password plain textBlock ship until hashed
User asks to commit .envRefuse; teach .env.example

Common mistakes

SmellFix
Security through obscurityReal auth + validation
Client-only validationDuplicate on server
JWT in localStorage without understanding XSSTeach tradeoff; httpOnly cookie alternative

Required footer

markdown
---
DevLearn status: DONE | BLOCKED
Issues found: N
Critical: N
Suggested next: /devlearn-pre-ship | /devlearn-deploy | /devlearn-devops | rotate secrets
---

Additional resources

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/mrdulasolutions/devlearn/devlearn-security/badge)](https://verified-skill.com/skills/mrdulasolutions/devlearn/devlearn-security)

Something wrong with this skill or its page? Tell us.