angular-nest-ai-kit / agents
AGENTS.md2 starsChanged 4 months agoMITView on GitHub
Personas de IA desacopladas de qualquer projeto especifico, para monorepos que seguem o padrao Angular standalone + Signals + PrimeNG no frontend e NestJS + Prisma no backend. Copiar cada arquivo para .claude/agents/<nome>.md (Claude Code), .cursor/agents/ (Cursor, renomeando para .mdc) ou equivalente da ferramenta usada. Nao pular etapas para "ganhar tempo": o custo aparece no review. Todo agente foi escrito com estes placeholders. Trocar antes de usar: Os agentes dizem quem faz e com que criterio; as skills em skills/…
In this repository18 other files it gives its agents.
- ai-operating-protocolSkill
- angular-feature-architectureSkill
- angular-primeng-auditSkill
- angular-primeng-uiSkill
- angular-testing-patternsSkill
- commit-and-branch-conventionsSkill
- frontend-auth-bootstrapSkill
- generating-backend-moduleSkill
- generating-data-layerSkill
- generating-dumb-createSkill
- generating-dumb-editSkill
- generating-dumb-listSkill
- generating-smart-pageSkill
- monorepo-workspace-standardsSkill
- nestjs-api-standardsSkill
- prisma-migration-disciplineSkill
- rbac-security-auditorSkill
- shared-contracts-librarySkill
The file
Files
Agentes Genericos (Angular + NestJS)
Personas de IA desacopladas de qualquer projeto especifico, para monorepos que seguem o padrao Angular standalone + Signals + PrimeNG no frontend e NestJS + Prisma no backend.
Copiar cada arquivo para .claude/agents/<nome>.md (Claude Code), .cursor/agents/
(Cursor, renomeando para .mdc) ou equivalente da ferramenta usada.
Elenco
| Agente | Dominio | Aciona quando |
| architect | orquestracao | ponto de entrada; decide quem executa |
| schema-keeper | Prisma, migrations, queries | mudanca de schema, tabela nova, query lenta |
| backend-implementer | NestJS: modules, controllers, services, DTOs | endpoint ou regra de negocio nova |
| backend-quality-guard | specs backend, code review, Swagger | teste de service, revisao de PR backend |
| data-access-specialist | Angular: model, service, facade | camada de dados e estado do frontend |
| ui-component-engineer | Angular + PrimeNG: smart/dumb, templates | pagina, lista, dialog, refactor de template |
| frontend-qa-engineer | specs Angular (Jasmine/Karma) | teste de componente |
| security-specialist | RBAC, guards, permissoes, validacao | seguranca full-stack |
| monorepo-specialist | workspace, libs, build, cache, CI | configuracao de monorepo e dependencias |
| docs-master | Swagger, READMEs, docs tecnicas | documentacao desatualizada |
Fluxo full-stack
schema-keeper -> backend-implementer -> backend-quality-guard
-> data-access-specialist -> ui-component-engineer -> frontend-qa-engineer
-> security-specialist (valida ponta a ponta)Nao pular etapas para "ganhar tempo": o custo aparece no review.
Convencoes assumidas
Todo agente foi escrito com estes placeholders. Trocar antes de usar:
| Placeholder | Significado |
<frontend-app> / <backend-app> | nomes dos apps no workspace |
@org/shared | lib de contratos compartilhados (TypeScript puro) |
@org/auth-contracts | lib de contratos de autenticacao/permissao |
JwtAuthGuard / PermissionsGuard | guards de auth e autorizacao |
*hasPermission | diretiva estrutural de RBAC no template |
ApiResponse<T> | envelope padrao de resposta HTTP |
BaseService / BaseFacade | classes base do projeto |
Relacao com as skills
Os agentes dizem quem faz e com que criterio; as skills em skills/ dizem como gerar o codigo. Um agente sem as skills correspondentes ainda funciona, mas repete decisao de template a cada tarefa.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.