myatminlu
- 6 skills
- 0 followers
- 17 hours ago last updated
- ▌ Bug Fixing · myatminlu bundleA disciplined, architect-level workflow for diagnosing and fixing bugs in an existing codebase — reproduce, read the real code before touching anything, find the true root cause, design the minimal correct fix at the right layer, verify it, and hunt for clones of the same defect. Use this skill for ANY defect work — crashes, exceptions, wrong output, failing or flaky tests, regressions, race conditions, memory leaks, performance degradation, data corruption, integration failures, works-locally-but-not-in-prod problems, or any request phrased as fix, debug, broken, not working, error, or why does X happen. Use it even when the bug looks trivial or the user has already proposed a fix — especially then, because the proposed fix is usually a symptom-level patch. Do NOT use for greenfield feature work with no defect involved.
- ▌ Codebase Audit · myatminlu bundleForensic whole-codebase audit and staged refactor. Use this skill whenever the user wants a codebase reviewed, audited, cleaned up, refactored, or "made systematic" — and also whenever they mention duplicated code, duplicated features or services, copy-pasted modules, band-aid or quick-fix or hacky code, technical debt, legacy vs new implementations living side by side, dead code, code smells, or wants to verify that features, functions and services are actually wired together and working end to end. Trigger it for requests like "review my codebase", "find duplicate logic", "why is this code such a mess", "check if everything is connected properly", or "plan a refactor", even when the user does not say the word "audit". Do NOT use it for a single-file review, a single bug fix, or writing new features.
- ▌ Code Quality Check · myatminlu bundleRigorous, evidence-backed, read-only code quality review that produces severity-ranked findings, a scorecard, and a merge/ship verdict. Use this skill whenever the user asks for a code review, quality check, PR or diff review, "is this code good", "check my code", "review this file", a health check, a pre-merge gate, a second opinion on code that an AI agent or teammate wrote, or wants best-practice / standards compliance checked — including when they simply paste code and ask what is wrong with it. Also trigger on mentions of code smells, maintainability, readability, dead code, duplicated logic, band-aid fixes, error handling, security review, performance review, test quality, or "make this production ready". Use it even for a single file or a small diff. Do NOT use it to write the fix — produce the findings, then hand off to the implementation or bug-fixing workflow.
- ▌ Code Implementation · myatminlu bundleDisciplined, architect-grade workflow for implementing, fixing, or refactoring code in an existing codebase. Use this skill for ANY task that will change source code — new features, bug fixes, refactors, integrations, migrations, performance work, or "just a small change" — and especially when the user says something is broken, duplicated, patched-over, inconsistent, or not wired together properly. It forces the agent to read the real code before writing any, trace the root cause instead of patching symptoms, reuse what already exists instead of adding near-duplicate code, and verify the change end to end. Use it even when the fix looks obvious.
- ▌ Nestjs Dev Guidelines · myatminlu bundleProduction-grade NestJS backend standards for writing, reviewing, and evolving NestJS/Nest-style TypeScript services. Use when a repo has modules, controllers, providers/services, DTOs, guards, pipes, or Nest-oriented boundaries, or when designing NestJS APIs, modules, migrations, auth/RBAC, multi-tenant isolation, DDD layered/hexagonal architecture, ports/adapters, use cases, repository/query patterns, N+1/slow queries, validation, pagination, caching, error handling/exception filters, BullMQ/jobs, webhooks, uploads, decorators, provider scopes, dynamic modules, circular deps, health/readiness/shutdown, observability/logging/tracing, OpenTelemetry, testing, code review, modernization/version/runtime advice, or AI backend patterns like LLM gateways, SSE streaming, usage metering, and quotas. Verify volatile versions/commands/model IDs/APIs against official docs and the repo. Plain Express/Fastify/Hono/Koa without Nest boundaries: cross-cutting guidance only. Not for non-Node backends or pure frontend work.
- ▌ Project Structure Refactoring · myatminlu bundleCross-language project structure and structural refactoring standards: design clean folder/module layouts and safely restructure existing codebases. Use when restructuring or reorganizing a codebase, choosing or fixing a repo's folder structure, deciding where a file belongs, package-by-feature vs by-layer, reorganizing/moving files, planning a migration to a new layout, fixing messy/flat/deep folder trees, splitting a giant or god file into modules, monorepos (apps/packages, workspaces, Turborepo/Nx), Python src layout, Go cmd/internal layout, React/Next.js/Expo feature folders, barrel files, import path aliases, or updating Dockerfiles/CI/configs after file moves. Covers plan-approve-execute staged migrations with green builds, git mv history preservation, and boundary enforcement. Structural only: hand code-level refactoring (extract/rename/dedupe) to code-implementation or codebase-audit, NestJS layout specifics to nestjs-dev-guidelines. Principles apply to any language beyond the stacks documented.