docs: set license badge to Apache-2.0 and document known limitations

Author: Oswaldo Montaño

README.md

@@ -7,7 +7,7 @@
 [![Go Version](https://img.shields.io/badge/go-1.24+-00ADD8?logo=go)](https://go.dev/)
 ![Tests](https://img.shields.io/badge/tests-185%20passing-success)
 ![Pipeline](https://img.shields.io/badge/pipeline-functional-success)
-[![License](https://img.shields.io/badge/license-TBD-blue)](LICENSE)
+[![License](https://img.shields.io/badge/license-Apache%202.0-blue)](LICENSE)
 
 **A Go CLI to orchestrate microservice component tests: spin up mocks, launch the app, run the tests and clean everything up — with a single command.**
 
@@ -216,6 +216,14 @@ go test -tags=integration ./internal/plugin/services/...
 
 ---
 
+## ⚠️ Known limitations
+
+- **App→mock networking on Linux (native `gtool test`).** The default pipeline runs the app as a bridge-network container while the mocks publish their ports on the host, so a containerized app cannot reach a mock at `localhost`. It needs a host gateway (`host.docker.internal`, automatic on Docker Desktop, manual on Linux) or a shared Docker network. The `--stable` / `--native` path avoids this by running the app as native processes that reach the mocks over `127.0.0.1` (mountebank runs on the host network).
+- **Newer mock plugins are lightly tested.** `redis`, `mongodb`, `mysql` and `minio` currently have unit tests only — no Docker integration tests yet. Treat them as experimental.
+- **Karate result is exit-code based.** Pass/fail comes from the launcher's exit code; there is no per-scenario parsing of the `karate-reports` output.
+
+---
+
 ## 📦 Status
 
 Functional end-to-end pipeline: 10 mock plugins, app launcher (Docker and native), Karate runner and orchestration with guaranteed teardown. 185 tests passing.

docs/readme/README.es.md

@@ -7,7 +7,7 @@
 [![Go Version](https://img.shields.io/badge/go-1.24+-00ADD8?logo=go)](https://go.dev/)
 ![Tests](https://img.shields.io/badge/tests-185%20passing-success)
 ![Pipeline](https://img.shields.io/badge/pipeline-functional-success)
-[![License](https://img.shields.io/badge/license-TBD-blue)](LICENSE)
+[![License](https://img.shields.io/badge/license-Apache%202.0-blue)](../../LICENSE)
 
 **CLI en Go para orquestar pruebas de componente de microservicios: levanta mocks, lanza la app, corre los tests y limpia todo — con un solo comando.**
 
@@ -216,6 +216,14 @@ go test -tags=integration ./internal/plugin/services/...
 
 ---
 
+## ⚠️ Limitaciones conocidas
+
+- **Networking app→mock en Linux (`gtool test` nativo).** El pipeline por defecto corre la app como contenedor en red bridge, mientras los mocks publican sus puertos en el host, así que una app en contenedor no alcanza un mock en `localhost`. Necesita un host gateway (`host.docker.internal`, automático en Docker Desktop, manual en Linux) o una red Docker compartida. El path `--stable` / `--native` lo evita corriendo la app como procesos nativos que llegan a los mocks por `127.0.0.1` (mountebank corre en la red del host).
+- **Los mocks nuevos están poco probados.** `redis`, `mongodb`, `mysql` y `minio` hoy solo tienen unit tests — aún sin integration tests con Docker. Considéralos experimentales.
+- **El resultado de Karate es por exit-code.** El pass/fail viene del exit code del launcher; no hay parsing por escenario del output de `karate-reports`.
+
+---
+
 ## 📦 Estado
 
 Pipeline funcional end-to-end: 10 plugins de mock, lanzador de app (Docker y nativo), runner Karate y orquestación con teardown garantizado. 185 tests en verde.