vigia-nids 0.2.0__tar.gz

vigia_nids-0.2.0/.dockerignore

@@ -0,0 +1,17 @@
+.venv
+.git
+.mypy_cache
+.pytest_cache
+.ruff_cache
+__pycache__
+*.pyc
+out
+data
+Datadeprueba
+tests
+benchmarks
+docs
+
+web/node_modules
+web/dist
+web/.gitignore

vigia_nids-0.2.0/.gitignore

@@ -0,0 +1,31 @@
+# Python
+__pycache__/
+*.py[cod]
+*.egg-info/
+build/
+dist/
+.venv/
+venv/
+
+# Herramientas
+.pytest_cache/
+.mypy_cache/
+.ruff_cache/
+.coverage
+htmlcov/
+graphify-out/
+
+# Datos y salidas: los datasets de seguridad nunca se versionan
+data/
+out/
+Datadeprueba/
+*.pcap
+*.pcapng
+*.parquet
+*.csv
+
+# Editor / sistema operativo
+.vscode/
+.idea/
+.DS_Store
+Thumbs.db

vigia_nids-0.2.0/CHANGELOG.md

@@ -0,0 +1,60 @@
+# Changelog
+
+Formato basado en [Keep a Changelog](https://keepachangelog.com/es-ES/1.1.0/),
+y este proyecto sigue [Versionado Semántico](https://semver.org/lang/es/).
+
+## [Unreleased]
+
+## [0.2.0] - 2026-09-29
+
+### Agregado
+
+- El paquete se publica en PyPI como `vigia-nids` (el nombre `vigia` estaba
+  tomado); el comando y el paquete importable siguen siendo `vigia`.
+- **Perfiles de datasets conocidos** (`vigia.profiles`, requisito R17):
+  `--profile cic-ids-2017` en `vigia audit` (y `vigia.load(profile=...)` en el
+  SDK) fija los roles de columna y expone IPs atacantes, ventanas de ataque y
+  errores conocidos documentados por el CIC, sin que un `--label-col` u otro
+  flag explícito pierda prioridad sobre el perfil.
+- **`dup.near`** (requisito R2, casi-duplicados): MinHash LSH (extra opcional
+  `ml`, agrega `datasketch`) para encontrar filas casi idénticas que
+  `dup.exact` no detecta por no ser copias byte a byte. Umbral configurable
+  con `near_duplicate_threshold`.
+- **API REST** (`vigia.api`, extra opcional `api`): FastAPI con endpoints para
+  subir datasets, lanzar auditorías, envenenamiento y deriva de forma
+  asíncrona (cola en proceso con `asyncio`, sin Celery ni Redis), consultar
+  estado y descargar el reporte HTML. Autenticación por `X-API-Key`, límite de
+  subida y de peticiones por minuto configurables por variable de entorno, y
+  borrado automático del dataset subido al terminar el trabajo.
+- Comando `vigia serve` para levantar la API con uvicorn.
+- **Panel web** (`web/`, React + Vite + TypeScript): subir un dataset, ver el
+  resultado (semáforo, hallazgos, checks no ejecutados), correcciones
+  sugeridas y comparación de deriva con PSI por columna. El semáforo gris
+  tiene la misma jerarquía visual que el rojo: nunca se confunde "no se pudo
+  revisar todo" con "está limpio". `docker compose up` sirve la API y el
+  panel juntos en un solo puerto.
+
+## [0.1.0] - 2026-09-22
+
+Primera versión funcional. Los tres módulos implementados y validados contra
+datos reales.
+
+### Agregado
+
+- **Módulo 1 — Auditor** (`vigia audit`): 15 checks que cubren los diez
+  requisitos P0 (duplicados, validez, fugas entre entrenamiento y prueba,
+  ruido de etiqueta), validados contra CIC-IDS2017 y contra Engelen et al.
+- **Módulo 2 — Envenenamiento** (`vigia poison`): 4 detectores, medidos sobre
+  tráfico real con ataques inyectados.
+- **Módulo 3 — Deriva** (`vigia drift`): 4 checks, medidos sobre los cinco
+  días de CIC-IDS2017.
+- 8 correcciones automáticas, con cuarentena de las filas afectadas.
+- Lectores de CSV, Parquet, Zeek, Suricata EVE y PCAP (R13 y R14).
+- CLI con 7 comandos y códigos de salida 0/1/2 para integrarse en CI.
+- Reporte HTML y JSON, con semáforo rojo/amarillo/verde/gris.
+- 188 tests, `ruff` y `mypy` limpios.
+- Documentación: arquitectura, checks, uso, deriva y envenenamiento en
+  [`docs/`](docs/).
+
+[0.2.0]: https://github.com/Joshua203-byte/Data/releases/tag/v0.2.0
+[0.1.0]: https://github.com/Joshua203-byte/Data/releases/tag/v0.1.0

vigia_nids-0.2.0/Dockerfile

@@ -0,0 +1,65 @@
+# Vigía — imagen única que sirve la API REST y el panel web
+# (docs/PLAN.md, fase 3.5).
+#
+# Un solo contenedor: FastAPI sirve la API bajo /api/v1 y /health, y el
+# build estático del panel (web/dist) montado en / vía StaticFiles
+# (ver src/vigia/api/app.py, `VIGIA_WEB_DIST`). Esto evita un segundo
+# contenedor con nginx solo para servir archivos estáticos -- para cuatro
+# pantallas no vale la pena el proxy aparte.
+#
+# `docker compose up` deja todo navegable en http://localhost:8000.
+
+# ---- Etapa 1: build del frontend --------------------------------------
+FROM node:22-slim AS web-build
+WORKDIR /web
+COPY web/package.json web/package-lock.json ./
+RUN npm ci
+COPY web/ ./
+# VITE_API_BASE_URL vacío: el panel y la API comparten origen en este
+# contenedor (mismo proceso, mismo puerto), así que las rutas relativas
+# alcanzan y no hace falta hardcodear un host.
+ENV VITE_API_BASE_URL=""
+RUN npm run build
+
+# ---- Etapa 2: runtime Python -------------------------------------------
+FROM python:3.12-slim AS runtime
+
+# Metadatos minimos; no instalamos herramientas de compilacion que no hacen
+# falta en runtime (pyarrow y demas traen wheels precompilados para slim).
+RUN apt-get update \
+    && apt-get install -y --no-install-recommends curl \
+    && rm -rf /var/lib/apt/lists/*
+
+# Usuario sin privilegios (docs/PLAN.md, 3.5): nunca corremos como root.
+RUN groupadd --system vigia && useradd --system --gid vigia --home-dir /app --shell /usr/sbin/nologin vigia
+
+WORKDIR /app
+
+# Instalar el paquete con el extra `api`. `parquet` va también: es el
+# formato recomendado para encadenar auditorías (docs/USO.md) y sin
+# pyarrow la API no podría leer .parquet subidos.
+COPY pyproject.toml README.md ./
+COPY src ./src
+RUN pip install --no-cache-dir ".[api,parquet]"
+
+# Panel estático ya compilado, en la ruta que `VIGIA_WEB_DIST` espera por
+# defecto (web/dist relativo al directorio de trabajo del proceso).
+COPY --from=web-build /web/dist ./web/dist
+
+# Directorio para datasets/uploads temporales (docs/PLAN.md, 3.5): un
+# volumen se monta acá desde docker-compose.yml. El propio storage.py usa
+# `tempfile.mkdtemp`, así que esto es principalmente para dejar un punto de
+# montaje estable y, si se quiere, apuntar TMPDIR ahí.
+RUN mkdir -p /data/uploads && chown -R vigia:vigia /app /data
+
+ENV TMPDIR=/data/uploads \
+    PYTHONUNBUFFERED=1
+
+USER vigia
+
+EXPOSE 8000
+
+HEALTHCHECK --interval=30s --timeout=5s --start-period=10s --retries=3 \
+    CMD curl -f http://localhost:8000/health || exit 1
+
+CMD ["uvicorn", "vigia.api.app:app", "--host", "0.0.0.0", "--port", "8000"]

vigia_nids-0.2.0/LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

vigia_nids-0.2.0/PKG-INFO

@@ -0,0 +1,254 @@
+Metadata-Version: 2.5
+Name: vigia-nids
+Version: 0.2.0
+Summary: Control de calidad continuo para los datos de modelos de ciberseguridad
+Project-URL: Homepage, https://github.com/Joshua203-byte/Data
+Project-URL: Documentation, https://github.com/Joshua203-byte/Data/tree/main/vigia/docs
+Project-URL: Issues, https://github.com/Joshua203-byte/Data/issues
+Project-URL: Changelog, https://github.com/Joshua203-byte/Data/blob/main/vigia/CHANGELOG.md
+Author: Joshua
+License: Apache-2.0
+License-File: LICENSE
+Keywords: data-quality,dataset,machine-learning,nids,security
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Security
+Requires-Python: >=3.11
+Requires-Dist: polars>=1.0
+Requires-Dist: pyyaml>=6.0
+Requires-Dist: rich>=13.0
+Requires-Dist: typer>=0.12
+Provides-Extra: api
+Requires-Dist: fastapi>=0.110; extra == 'api'
+Requires-Dist: python-multipart>=0.0.9; extra == 'api'
+Requires-Dist: uvicorn[standard]>=0.29; extra == 'api'
+Provides-Extra: dev
+Requires-Dist: httpx>=0.27; extra == 'dev'
+Requires-Dist: mypy>=1.10; extra == 'dev'
+Requires-Dist: psutil>=5.9; extra == 'dev'
+Requires-Dist: pytest-cov>=5.0; extra == 'dev'
+Requires-Dist: pytest>=8.0; extra == 'dev'
+Requires-Dist: ruff>=0.5; extra == 'dev'
+Provides-Extra: ml
+Requires-Dist: cleanlab>=2.6; extra == 'ml'
+Requires-Dist: datasketch>=2.0; extra == 'ml'
+Requires-Dist: lightgbm>=4.0; extra == 'ml'
+Requires-Dist: numpy>=1.26; extra == 'ml'
+Requires-Dist: scikit-learn>=1.4; extra == 'ml'
+Provides-Extra: parquet
+Requires-Dist: pyarrow>=15.0; extra == 'parquet'
+Provides-Extra: pcap
+Requires-Dist: nfstream>=6.5; extra == 'pcap'
+Requires-Dist: pandas>=2.0; extra == 'pcap'
+Description-Content-Type: text/markdown
+
+# Vigía
+
+**Control de calidad continuo para los datos de modelos de ciberseguridad.**
+
+Vigía revisa los datasets con los que se entrenan los detectores de intrusiones y
+responde una pregunta incómoda: *¿este 0.99 de F1 es real, o es un artefacto de
+los datos?*
+
+La investigación sobre los datasets más usados del área (CIC-IDS2017,
+CSE-CIC-IDS2018, UGR'16) documenta duplicados masivos, etiquetas incorrectas,
+fugas entre entrenamiento y prueba, y columnas que le delatan la respuesta al
+modelo. El resultado es siempre el mismo: **el modelo parece excelente en el
+laboratorio y falla en una red real.**
+
+> Estado: **v0.2.0 — los tres módulos implementados y validados contra datos
+> reales.** El auditor cubre los diez requisitos P0, más los lectores de Zeek,
+> Suricata EVE y PCAP (R13 y R14). El monitor de deriva está medido sobre los
+> cinco días de CIC-IDS2017, y los detectores de envenenamiento sobre ataques
+> inyectados en ese mismo tráfico y sobre el ruido de etiqueta que documentaron
+> Engelen et al. Los números están en
+> [`benchmarks/RESULTADOS.md`](benchmarks/RESULTADOS.md), incluidos los casos
+> donde el rendimiento es pobre.
+
+## Los tres módulos
+
+| Módulo | Comando | Qué responde |
+|---|---|---|
+| **1. Auditor** | `vigia audit` | ¿Este dataset tiene defectos que inflen las métricas? |
+| **2. Envenenamiento** | `vigia poison` | ¿Alguien insertó registros para manipular el modelo? |
+| **3. Deriva** | `vigia drift` | ¿El tráfico de hoy sigue pareciéndose al de entrenamiento? |
+
+24 checks, 8 correcciones automáticas y la comparación de rendimiento
+antes/después.
+
+## Instalación
+
+```bash
+pip install -e ".[dev]"      # desde el repositorio
+vigia --help
+```
+
+Requiere Python 3.11 o superior.
+
+## Uso
+
+```bash
+# Auditar un dataset (las columnas se detectan solas por nombre)
+vigia audit data/flows.csv --report out/
+
+# Indicar las columnas explícitamente
+vigia audit data/flows.parquet \
+  --label-col Label --split-col split --time-col Timestamp \
+  --report out/
+
+# Correr solo una categoría de checks
+vigia audit data/flows.csv --checks leak,dup
+
+# En CI: salir con código 1 si hay hallazgos críticos
+vigia audit data/train.csv --fail-on critical
+
+# Listar los checks disponibles
+vigia checks
+```
+
+Desde Python:
+
+```python
+import vigia
+
+ctx = vigia.load("data/flows.parquet", label_col="Label", split_col="split")
+report = vigia.audit(ctx)
+
+print(report.traffic_light())  # 'rojo' | 'amarillo' | 'verde' | 'gris'
+print(report.counts())  # {'critical': 2, 'high': 1, ...}
+
+from vigia.report import write_report
+
+write_report(report, "out/")  # report.html + report.json
+```
+
+## Qué revisa hoy
+
+| Check | Qué encuentra |
+|---|---|
+| `dup.exact` | Filas idénticas: el modelo se evalúa con ejemplos que ya vio |
+| `dup.class_ratio` | Clases formadas mayoritariamente por duplicados |
+| `dup.cross_split` | **Crítico.** La misma fila en entrenamiento y en prueba |
+| `leak.temporal` | El conjunto de prueba se solapa con el de entrenamiento en el tiempo |
+| `leak.host` | El mismo host atacante en ambos splits: el modelo memoriza la IP |
+| `shortcut.single_feature` | Una sola columna predice la etiqueta demasiado bien |
+| `shortcut.identifier` | IPs, IDs de flujo o timestamps usados como características |
+| `validity.nan_inf` | NaN e infinitos (CIC-IDS2017 los tiene en las columnas de tasas) |
+| `validity.constant` | Columnas con un solo valor |
+| `validity.impossible` | Puertos > 65535, duraciones negativas, bytes negativos |
+| `validity.column_names` | Nombres con espacios (`" Destination Port"`) o ambiguos |
+| `leak.session` | Flujos de la misma conexión repartidos entre splits |
+| `labels.conflict` | Filas idénticas con etiquetas distintas |
+| `labels.noise` | Etiquetas que contradicen al resto del dataset (cleanlab) |
+| `labels.taxonomy` | La misma clase escrita de varias formas |
+| `labels.imbalance` | Clases con tan pocos ejemplos que su métrica no dice nada |
+
+Cada hallazgo trae severidad, métrica, filas afectadas, ejemplos concretos y una
+recomendación. Un check que no puede correr (falta la columna de split, por
+ejemplo) no falla en silencio: aparece en la sección "checks no ejecutados" del
+reporte.
+
+### El semáforo
+
+| Color | Significa |
+|---|---|
+| **rojo** | Hay hallazgos críticos o altos |
+| **amarillo** | Hay hallazgos medios o bajos |
+| **verde** | Sin hallazgos y todos los checks se ejecutaron |
+| **gris** | Sin hallazgos, pero algunos checks no se ejecutaron |
+
+El gris existe porque "no encontré nada" y "está limpio" no son lo mismo. Si la
+columna de etiqueta o la de split no se detectan, los checks que se saltan son
+justamente los que buscan fuga y duplicados entre conjuntos: dar verde ahí sería
+confirmar que el dataset está bien cuando en realidad casi no se lo revisó. Si el
+semáforo da gris, especificá las columnas a mano con `--label-col` y `--split-col`.
+
+## Corregir
+
+Los hallazgos nombran su corrección. `vigia fix` las aplica:
+
+```bash
+# Ver las disponibles
+vigia fixes
+
+# Aplicarlas en cadena; el archivo original nunca se toca
+vigia fix data/flows.csv \
+  --apply drop_duplicates,drop_constant,drop_identifiers \
+  --out data/flows_corregido.parquet
+```
+
+Las filas con etiqueta dudosa **nunca se borran**: `quarantine_noise` las deja
+en un archivo aparte para que una persona las revise.
+
+## Medir cuánto del rendimiento era real
+
+```python
+from vigia.benchmark import compare
+
+c = compare(ctx, ["drop_identifiers"])
+print(c.summary())
+print(c.after.per_class_recall)
+```
+
+Entrena el mismo modelo antes y después de corregir. Una caída no significa
+que las correcciones empeoraron el modelo: significa que el número original
+estaba inflado. Sobre CIC-IDS2017, cambiar un split aleatorio por uno temporal
+lleva el recall de DoS Hulk, PortScan y DDoS de ~0,95 a **0,00** — ver
+[`benchmarks/RESULTADOS.md`](benchmarks/RESULTADOS.md).
+
+## Formatos de entrada
+
+| Formato | Extensión | Notas |
+|---|---|---|
+| CSV | `.csv`, `.txt` | Reintenta en cp1252 si no es UTF-8 |
+| Parquet | `.parquet`, `.pq` | Conserva los tipos |
+| Zeek | `.log` | TSV con cabecera o JSON Lines |
+| Suricata EVE | `.json`, `.eve` | `--event-type` elige flujos o alertas |
+| PCAP | `.pcap`, `.pcapng`, `.cap` | Con `pip install 'vigia-nids[pcap]'` |
+
+Archivo suelto o carpeta completa: los datasets CIC vienen partidos en un
+archivo por día, Vigía los concatena y conserva la procedencia en
+`__source_file`. Los PCAP se leen de a uno, porque extraer flujos tarda minutos.
+
+Detalles de cada formato en [`docs/USO.md`](docs/USO.md#formatos-de-entrada).
+
+## Desarrollo
+
+```bash
+pip install -e ".[dev,ml]"
+
+pytest                 # pruebas
+ruff check src tests   # lint
+ruff format src tests  # formato
+mypy src               # tipos
+```
+
+Agregar un check nuevo es una clase con cuatro atributos y un método. El
+registro es automático al importarse el módulo desde `vigia/checks/__init__.py`.
+Los detalles están en [`docs/ARQUITECTURA.md`](docs/ARQUITECTURA.md).
+
+## Documentación
+
+| Documento | Para qué |
+|---|---|
+| [`docs/CHECKS.md`](docs/CHECKS.md) | Qué significa cada hallazgo del auditor y qué hacer |
+| [`docs/ENVENENAMIENTO.md`](docs/ENVENENAMIENTO.md) | Módulo 2: los detectores, el simulador y qué tan bien funcionan |
+| [`docs/DERIVA.md`](docs/DERIVA.md) | Módulo 3: los checks, las métricas y cómo leerlos |
+| [`docs/USO.md`](docs/USO.md) | Referencia de la CLI, del SDK y del formato JSON |
+| [`docs/ARQUITECTURA.md`](docs/ARQUITECTURA.md) | Cómo está armado por dentro; cómo agregar un check |
+| [`benchmarks/RESULTADOS.md`](benchmarks/RESULTADOS.md) | Resultados verificados sobre NSL-KDD y CIC-IDS2017 |
+| [`docs/PLAN.md`](docs/PLAN.md) | Qué falta hasta la 1.0, en qué orden y cómo |
+
+## Documento de diseño
+
+El diseño completo — los tres módulos (auditor, envenenamiento, deriva), la
+arquitectura, el plan por fases y las referencias académicas — está en
+[`vigia.md`](../vigia.md).
+
+## Licencia
+
+Apache 2.0.