Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Backend

ASecurity

Backend decision boundaries. Constrain decisions, not implementations. Use for APIs, services, database access, auth, and server-side logic.

6 stars
0 votes
0 copies
0 views
Added 9/28/2026
ai-agentsrustrefactoringapidatabasebackendsecurity

Works with

cliapi

Security Analysis

A100/100

Scanned 9/28/2026

Install to Claude Code

$npx -y skills add vorlaxen-labs/Agent-Skills --skill backend --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Backend?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Backend
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/vorlaxen-labs-backend/badge)](https://www.skillsdirectory.com/skills/vorlaxen-labs-backend)

More formats (shields.io, HTML) on the badges page.

Files
SKILL.md
---
name: web-backend
description: >-
  Backend decision boundaries. Constrain decisions, not implementations.
  Use for APIs, services, database access, auth, and server-side logic.
---
# Core Principle

**Constrain decisions, not implementations.**

These rules add backend **decision boundaries** on top of global standards.

They do not prescribe frameworks, ORMs, folder layouts, or service patterns. Follow the project.

---

# Hard Rules — Read First

1. **Never hallucinate.** Do not invent routes, handlers, services, models, tables, columns, or API contracts. Read the codebase first.
2. **Match risk to action.** Stop and ask for schema, auth flow, or API shape — not for placement or patterns already in the codebase. Read first, then proceed or ask.
3. **Do not change the project's backend stack** unless the user explicitly asks.
4. **These rules constrain decisions only.** Implementation style comes from the existing codebase.

---

# Decision: Stack & Architecture

**Forbidden without explicit user request:**

- Introducing a new framework, ORM, or validation library
- Introducing a service layer if the project does not use one
- Moving business logic out of handlers if the project keeps it there
- Imposing a different architectural pattern than the project uses

**Required:** read existing handlers, services, and models before writing anything.

---

# Decision: API Contracts

**Forbidden without explicit user request:**

- Breaking changes to existing API contracts
- Renaming or removing fields used by existing consumers
- Inventing endpoints, fields, or response shapes

**Required before any contract change:** read schemas, types, consumers, and validation.

**Required outcome:** clients never receive raw database errors, stack traces, or internal details.

---

# Decision: Input & Auth

**Required outcomes:**

- Untrusted input is validated at the API boundary
- Client-provided auth or ownership claims are verified server-side

**Forbidden:** introducing a second validation or auth system.

**Not your decision:** which validation or auth library to use — follow the project.

---

# Decision: Database

**Forbidden:**

- Guessing table names, columns, or relationships
- Inventing schema
- Schema changes beyond what the task requires

**Required:** read existing schema and models first.

**If names or relationships are unknown:** stop and read, or ask.

**Required outcomes for multi-step mutations:**

- Data remains consistent on failure
- Duplicate-sensitive operations are handled safely when the domain requires it

**Not your decision:** whether to use transactions, idempotency keys, or specific query patterns — follow the project unless the task demands a specific fix.

---

# Decision: Error Handling

**Required outcomes:**

- Errors are never silently swallowed
- Success is never returned on failure
- Client responses do not leak sensitive internals
- Useful context is preserved for logging

**Not your decision:** specific error types or response format — follow the project.

---

# Decision: Infrastructure & Background Systems

**Forbidden unless the task explicitly requires it or the user approves:**

- Adding queues, caches, distributed locks, or background jobs
- Adding Redis, BullMQ, or similar infrastructure
- Adding caching without a defined invalidation strategy

**If the task needs infrastructure beyond current scope:** stop, describe, ask, wait.

---

# Decision: Security

**Required:** consider security at external boundaries.

**Forbidden:** adding security infrastructure unrelated to the current feature.

**Not your decision:** specific security tooling — follow the project.

---

# Decision: Scope

**Forbidden without approval:**

- Refactoring unrelated services
- Migrating the backend to a new pattern or stack

**If unsure whether a change is in scope:** stop and ask.

Attribution

vorlaxen-labsvorlaxen-labs
View sourceMore from vorlaxen-labs →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1074701 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

695601 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

691 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →