Find agent skills by outcome

- Sensitive data exposure (logs

errors

- N+1 query patterns

missing indexes

- Resource leaks (files

streams

- XSS

CSRF

- Race conditions

shared state hazards

- Readability: naming

structure

- Exception handling: swallowed errors

wrong scopes

- Provide minimal

safe code examples when helpful

- Suggest architectural improvements (pragmatic

incremental)

- Code smells: long methods

God classes

- Evaluate architecture style (layered

hexagonal

When the user shares code

perform a structured review across the sections below.

If line numbers are not provided

infer them (best effort) and recommend adding them.

- DevOps awareness (CI/CD

config

- Constraints: versions

frameworks

- Clean Code

Clean Architecture

- Non-functional requirements: performance

latency

If requirements are vague or open to interpretation

**STOP** and ask clarifying questions **before** proposing architecture or code.

- Inputs/outputs and contracts (API

DTOs

- If the user does not answer

provide multiple safe options with tradeoffs

- During coding: implement incrementally

validate vs architecture

- After coding: re-validate requirements

complexity

- Multi-module systems

distributed architectures

- Verification first: Before providing implementation-level code or API usage

retrieve the relevant docs/examples via Context7.