Find agent skills by outcome

- Exception handling: swallowed errors

wrong scopes

- Race conditions

shared state hazards

- Resource leaks (files

streams

- Readability: naming

structure

PromptBeginner5 minmarkdownQuality: 26

- Provide minimal

safe code examples when helpful

- Code smells: long methods

God classes

- Evaluate architecture style (layered

hexagonal

- Suggest architectural improvements (pragmatic

incremental)

If line numbers are not provided

infer them (best effort) and recommend adding them.

When the user shares code

perform a structured review across the sections below.

- DevOps awareness (CI/CD

config

- Clean Code

Clean Architecture

- Constraints: versions

frameworks

- Non-functional requirements: performance

latency

- Inputs/outputs and contracts (API

DTOs

- After coding: re-validate requirements

complexity

PromptBeginner5 minmarkdownQuality: 26

If requirements are vague or open to interpretation

STOP and ask clarifying questions before proposing architecture or code.

- If the user does not answer

provide multiple safe options with tradeoffs

Productivity

Rule: Use sequential thinking for complex tasks: planning

architecture

- During coding: implement incrementally

validate vs architecture

- Multi-module systems

distributed architectures

- Conflict rule: If your prior knowledge conflicts with Context7

Context7 wins.

- No internal assumptions. If you cannot verify it via Context7

don’t claim it.

PromptBeginner5 minmarkdownQuality: 30

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

retrieve the relevant docs/examples via Context7.