feat: funciones Python para core, cybersecurity, datascience y finance

Agrega funciones Python reutilizables organizadas por dominio:
- core: composicion funcional (pipe, compose, map, filter, reduce, etc.)
- cybersecurity: analisis de amenazas y puertos
- datascience: estadisticas y deteccion de outliers
- finance: indicadores tecnicos y analisis financiero

python/functions/core/__init__.py

@@ -0,0 +1,43 @@
+"""Core functional programming utilities."""
+
+from .core import (
+    all_of,
+    any_of,
+    chunk,
+    compose,
+    drop,
+    filter_list,
+    find,
+    find_index,
+    flat_map,
+    flatten,
+    group_by,
+    map_list,
+    partition,
+    pipe,
+    reduce_list,
+    take,
+    unique,
+    zip_with,
+)
+
+__all__ = [
+    "all_of",
+    "any_of",
+    "chunk",
+    "compose",
+    "drop",
+    "filter_list",
+    "find",
+    "find_index",
+    "flat_map",
+    "flatten",
+    "group_by",
+    "map_list",
+    "partition",
+    "pipe",
+    "reduce_list",
+    "take",
+    "unique",
+    "zip_with",
+]

python/functions/core/all_of.md

@@ -0,0 +1,32 @@
+---
+name: all_of
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def all_of(xs: list, pred: callable) -> bool"
+description: "Retorna True si todos los elementos de la lista cumplen el predicado."
+tags: [list, functional, predicate, all, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = all_of([2, 4, 6], lambda n: n % 2 == 0)
+# True
+```
+
+## Notas
+
+Funcion pura. Retorna True para lista vacia (vacuamente verdadero). Cortocircuita al primer False.

python/functions/core/any_of.md

@@ -0,0 +1,32 @@
+---
+name: any_of
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def any_of(xs: list, pred: callable) -> bool"
+description: "Retorna True si al menos un elemento de la lista cumple el predicado."
+tags: [list, functional, predicate, any, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = any_of([1, 3, 5, 4], lambda n: n % 2 == 0)
+# True
+```
+
+## Notas
+
+Funcion pura. Retorna False para lista vacia. Cortocircuita al primer True.

python/functions/core/chunk.md

@@ -0,0 +1,32 @@
+---
+name: chunk
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def chunk(xs: list, size: int) -> list"
+description: "Divide una lista en sublistas de tamanio fijo. El ultimo chunk puede ser menor."
+tags: [list, functional, chunk, partition, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = chunk([1, 2, 3, 4, 5], 2)
+# [[1, 2], [3, 4], [5]]
+```
+
+## Notas
+
+Funcion pura. Si size <= 0 retorna lista vacia.

python/functions/core/compose.md

@@ -0,0 +1,33 @@
+---
+name: compose
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def compose(*fns) -> callable"
+description: "Compone funciones de derecha a izquierda. compose(f, g)(x) == f(g(x))."
+tags: [functional, compose, composition, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+double_then_str = compose(str, lambda n: n * 2)
+result = double_then_str(5)
+# "10"
+```
+
+## Notas
+
+Funcion pura. Composicion matematica: la ultima funcion se aplica primero. Inverso de pipe.

python/functions/core/core.py

@@ -0,0 +1,135 @@
+"""Core functional programming utilities — pure functions for list/collection operations."""
+
+from functools import reduce as _reduce
+from typing import Any, Callable, Dict, List, Tuple
+
+
+def filter_list(xs: list, pred: Callable) -> list:
+    """Filter list by predicate. Does not mutate the original."""
+    return [x for x in xs if pred(x)]
+
+
+def map_list(xs: list, fn: Callable) -> list:
+    """Map function over list. Does not mutate the original."""
+    return [fn(x) for x in xs]
+
+
+def reduce_list(xs: list, initial: Any, fn: Callable) -> Any:
+    """Reduce list with accumulator. fn(acc, x) -> acc."""
+    return _reduce(fn, xs, initial)
+
+
+def flat_map(xs: list, fn: Callable) -> list:
+    """Map function over list then flatten one level."""
+    result = []
+    for x in xs:
+        result.extend(fn(x))
+    return result
+
+
+def flatten(xss: list) -> list:
+    """Flatten a list of lists one level."""
+    result = []
+    for xs in xss:
+        result.extend(xs)
+    return result
+
+
+def chunk(xs: list, size: int) -> list:
+    """Split list into chunks of given size. Last chunk may be smaller."""
+    if size <= 0:
+        return []
+    return [xs[i : i + size] for i in range(0, len(xs), size)]
+
+
+def take(xs: list, n: int) -> list:
+    """Take first n elements from list."""
+    return xs[:n]
+
+
+def drop(xs: list, n: int) -> list:
+    """Drop first n elements from list."""
+    return xs[n:]
+
+
+def unique(xs: list) -> list:
+    """Remove duplicates preserving order. Uses identity for hashable elements."""
+    seen = set()
+    result = []
+    for x in xs:
+        if x not in seen:
+            seen.add(x)
+            result.append(x)
+    return result
+
+
+def group_by(xs: list, key_fn: Callable) -> Dict:
+    """Group elements by key function. Returns dict of key -> list."""
+    groups: Dict = {}
+    for x in xs:
+        k = key_fn(x)
+        if k not in groups:
+            groups[k] = []
+        groups[k].append(x)
+    return groups
+
+
+def partition(xs: list, pred: Callable) -> Tuple[list, list]:
+    """Split list into (matches, non_matches) based on predicate."""
+    matches = []
+    non_matches = []
+    for x in xs:
+        if pred(x):
+            matches.append(x)
+        else:
+            non_matches.append(x)
+    return (matches, non_matches)
+
+
+def find(xs: list, pred: Callable) -> Any:
+    """Find first element matching predicate. Returns None if not found."""
+    for x in xs:
+        if pred(x):
+            return x
+    return None
+
+
+def find_index(xs: list, pred: Callable) -> int:
+    """Find index of first element matching predicate. Returns -1 if not found."""
+    for i, x in enumerate(xs):
+        if pred(x):
+            return i
+    return -1
+
+
+def zip_with(xs: list, ys: list, fn: Callable) -> list:
+    """Zip two lists with a combining function. Stops at shorter list."""
+    return [fn(x, y) for x, y in zip(xs, ys)]
+
+
+def all_of(xs: list, pred: Callable) -> bool:
+    """Return True if all elements match predicate."""
+    return all(pred(x) for x in xs)
+
+
+def any_of(xs: list, pred: Callable) -> bool:
+    """Return True if any element matches predicate."""
+    return any(pred(x) for x in xs)
+
+
+def pipe(value: Any, *fns: Callable) -> Any:
+    """Pipe a value through a sequence of functions left-to-right."""
+    result = value
+    for fn in fns:
+        result = fn(result)
+    return result
+
+
+def compose(*fns: Callable) -> Callable:
+    """Compose functions right-to-left. compose(f, g)(x) == f(g(x))."""
+    def composed(x: Any) -> Any:
+        result = x
+        for fn in reversed(fns):
+            result = fn(result)
+        return result
+    return composed

python/functions/core/drop.md

@@ -0,0 +1,32 @@
+---
+name: drop
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def drop(xs: list, n: int) -> list"
+description: "Descarta los primeros n elementos de una lista."
+tags: [list, functional, drop, slice, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = drop([1, 2, 3, 4, 5], 2)
+# [3, 4, 5]
+```
+
+## Notas
+
+Funcion pura. Si n > len(xs), retorna lista vacia. No muta la original.

python/functions/core/filter_list.md

@@ -0,0 +1,32 @@
+---
+name: filter_list
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def filter_list(xs: list, pred: callable) -> list"
+description: "Filtra una lista aplicando un predicado sin mutar la original."
+tags: [list, functional, filter, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+evens = filter_list([1, 2, 3, 4], lambda n: n % 2 == 0)
+# [2, 4]
+```
+
+## Notas
+
+Funcion pura. No muta la lista original. Equivalente a `[x for x in xs if pred(x)]`.

python/functions/core/find.md

@@ -0,0 +1,32 @@
+---
+name: find
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def find(xs: list, pred: callable)"
+description: "Encuentra el primer elemento que cumple el predicado. Retorna None si no hay coincidencia."
+tags: [list, functional, find, search, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = find([1, 2, 3, 4], lambda n: n > 2)
+# 3
+```
+
+## Notas
+
+Funcion pura. Retorna None si ningun elemento cumple el predicado. Cortocircuita al primer match.

python/functions/core/find_index.md

@@ -0,0 +1,32 @@
+---
+name: find_index
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def find_index(xs: list, pred: callable) -> int"
+description: "Encuentra el indice del primer elemento que cumple el predicado. Retorna -1 si no hay coincidencia."
+tags: [list, functional, find, index, search, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+idx = find_index([10, 20, 30, 40], lambda n: n > 25)
+# 2
+```
+
+## Notas
+
+Funcion pura. Retorna -1 si ningun elemento cumple el predicado.

python/functions/core/flat_map.md

@@ -0,0 +1,32 @@
+---
+name: flat_map
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def flat_map(xs: list, fn: callable) -> list"
+description: "Aplica una funcion que retorna listas a cada elemento y aplana el resultado un nivel."
+tags: [list, functional, flatmap, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = flat_map([1, 2, 3], lambda n: [n, n * 10])
+# [1, 10, 2, 20, 3, 30]
+```
+
+## Notas
+
+Funcion pura. Equivalente a flatten(map_list(xs, fn)). Solo aplana un nivel.

python/functions/core/flatten.md

@@ -0,0 +1,32 @@
+---
+name: flatten
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def flatten(xss: list) -> list"
+description: "Aplana una lista de listas un nivel, concatenando las sublistas."
+tags: [list, functional, flatten, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = flatten([[1, 2], [3], [4, 5]])
+# [1, 2, 3, 4, 5]
+```
+
+## Notas
+
+Funcion pura. Solo aplana un nivel de anidamiento.

python/functions/core/group_by.md

@@ -0,0 +1,32 @@
+---
+name: group_by
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def group_by(xs: list, key_fn: callable) -> dict"
+description: "Agrupa elementos de una lista por una funcion clave. Retorna dict de clave a lista."
+tags: [list, functional, group, classify, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = group_by(["hi", "hey", "bye"], lambda s: s[0])
+# {"h": ["hi", "hey"], "b": ["bye"]}
+```
+
+## Notas
+
+Funcion pura. El orden de los elementos dentro de cada grupo se preserva.

python/functions/core/map_list.md

@@ -0,0 +1,32 @@
+---
+name: map_list
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def map_list(xs: list, fn: callable) -> list"
+description: "Aplica una funcion a cada elemento de una lista, retornando una nueva lista."
+tags: [list, functional, map, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+doubled = map_list([1, 2, 3], lambda n: n * 2)
+# [2, 4, 6]
+```
+
+## Notas
+
+Funcion pura. No muta la lista original. Equivalente a `[fn(x) for x in xs]`.

python/functions/core/partition.md

@@ -0,0 +1,32 @@
+---
+name: partition
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def partition(xs: list, pred: callable) -> tuple"
+description: "Divide una lista en dos: (elementos que cumplen el predicado, elementos que no)."
+tags: [list, functional, partition, split, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+evens, odds = partition([1, 2, 3, 4, 5], lambda n: n % 2 == 0)
+# evens = [2, 4], odds = [1, 3, 5]
+```
+
+## Notas
+
+Funcion pura. Retorna tupla de dos listas: (matches, non_matches).

python/functions/core/pipe.md

@@ -0,0 +1,36 @@
+---
+name: pipe
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def pipe(value, *fns)"
+description: "Pasa un valor a traves de una secuencia de funciones de izquierda a derecha."
+tags: [functional, pipe, composition, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = pipe(
+    [1, 2, 3, 4, 5],
+    lambda xs: filter_list(xs, lambda n: n > 2),
+    lambda xs: map_list(xs, lambda n: n * 10),
+)
+# [30, 40, 50]
+```
+
+## Notas
+
+Funcion pura. Ejecuta las funciones en orden de izquierda a derecha: f1(value), luego f2(result), etc.

python/functions/core/reduce_list.md

@@ -0,0 +1,32 @@
+---
+name: reduce_list
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def reduce_list(xs: list, initial, fn: callable)"
+description: "Reduce una lista con un acumulador y una funcion binaria fn(acc, x)."
+tags: [list, functional, reduce, fold, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+total = reduce_list([1, 2, 3, 4], 0, lambda acc, x: acc + x)
+# 10
+```
+
+## Notas
+
+Funcion pura. Usa functools.reduce internamente. El valor inicial es obligatorio para evitar errores con listas vacias.

python/functions/core/take.md

@@ -0,0 +1,32 @@
+---
+name: take
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def take(xs: list, n: int) -> list"
+description: "Toma los primeros n elementos de una lista."
+tags: [list, functional, take, slice, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = take([1, 2, 3, 4, 5], 3)
+# [1, 2, 3]
+```
+
+## Notas
+
+Funcion pura. Si n > len(xs), retorna toda la lista. No muta la original.

python/functions/core/unique.md

@@ -0,0 +1,32 @@
+---
+name: unique
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def unique(xs: list) -> list"
+description: "Elimina duplicados de una lista preservando el orden de aparicion."
+tags: [list, functional, unique, deduplicate, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = unique([1, 2, 2, 3, 1, 4])
+# [1, 2, 3, 4]
+```
+
+## Notas
+
+Funcion pura. Preserva el orden de la primera aparicion. Requiere elementos hashables.

python/functions/core/zip_with.md

@@ -0,0 +1,32 @@
+---
+name: zip_with
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def zip_with(xs: list, ys: list, fn: callable) -> list"
+description: "Combina dos listas elemento a elemento con una funcion. Se detiene en la mas corta."
+tags: [list, functional, zip, combine, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/core/core.py"
+---
+
+## Ejemplo
+
+```python
+result = zip_with([1, 2, 3], [10, 20, 30], lambda a, b: a + b)
+# [11, 22, 33]
+```
+
+## Notas
+
+Funcion pura. Se detiene en la lista mas corta, como zip() de Python.

python/functions/cybersecurity/__init__.py

@@ -0,0 +1,25 @@
+from .cybersecurity import (
+    hash_sha256,
+    hash_md5,
+    entropy_shannon,
+    detect_sql_injection,
+    extract_urls,
+    is_base64,
+    is_hex,
+    levenshtein_distance,
+    jaccard_similarity,
+    normalize_url,
+)
+
+__all__ = [
+    "hash_sha256",
+    "hash_md5",
+    "entropy_shannon",
+    "detect_sql_injection",
+    "extract_urls",
+    "is_base64",
+    "is_hex",
+    "levenshtein_distance",
+    "jaccard_similarity",
+    "normalize_url",
+]

python/functions/cybersecurity/cybersecurity.py

@@ -0,0 +1,167 @@
+"""Cybersecurity pure functions: hashing, parsing, and security utilities."""
+
+import hashlib
+import math
+import re
+import base64
+from collections import Counter
+from urllib.parse import urlparse, urlunparse, parse_qs, urlencode
+
+
+def hash_sha256(data: bytes) -> str:
+    """Calcula el hash SHA-256 de datos binarios. Retorna hex digest."""
+    return hashlib.sha256(data).hexdigest()
+
+
+def hash_md5(data: bytes) -> str:
+    """Calcula el hash MD5 de datos binarios. Retorna hex digest."""
+    return hashlib.md5(data).hexdigest()
+
+
+def entropy_shannon(data: bytes) -> float:
+    """Calcula la entropia de Shannon de datos binarios (0-8 bits por byte).
+
+    Entropia alta (>7.5) sugiere datos cifrados o comprimidos.
+    Entropia baja (<3) sugiere datos estructurados o repetitivos.
+    """
+    if not data:
+        return 0.0
+    length = len(data)
+    counts = Counter(data)
+    entropy = 0.0
+    for count in counts.values():
+        p = count / length
+        if p > 0:
+            entropy -= p * math.log2(p)
+    return entropy
+
+
+_SQL_INJECTION_PATTERNS = [
+    (r"('\s*OR\s+'[^']*'\s*=\s*'[^']*'?)", "string_tautology"),
+    (r"('\s*(OR|AND)\s+'?\d+\s*=\s*\d+)", "tautology"),
+    (r"(;\s*(DROP|DELETE|UPDATE|INSERT)\b)", "stacked_query"),
+    (r"(UNION\s+(ALL\s+)?SELECT)", "union_select"),
+    (r"(\b(SELECT|INSERT|UPDATE|DELETE|DROP|ALTER|CREATE|EXEC)\b\s)", "sql_keyword"),
+    (r"(--\s*$|/\*|\*/)", "comment_injection"),
+    (r"(BENCHMARK\s*\(|SLEEP\s*\(|WAITFOR\s+DELAY)", "time_based"),
+    (r"(CHAR\s*\(\s*\d+)", "char_function"),
+    (r"(CONCAT\s*\()", "concat_function"),
+    (r"(0x[0-9a-fA-F]{4,})", "hex_literal"),
+]
+
+
+def detect_sql_injection(input_str: str) -> tuple:
+    """Detecta patrones de SQL injection en un string.
+
+    Retorna (is_threat, pattern) donde pattern es el nombre del patron
+    detectado o cadena vacia si no hay amenaza.
+    """
+    for pattern, name in _SQL_INJECTION_PATTERNS:
+        if re.search(pattern, input_str, re.IGNORECASE):
+            return (True, name)
+    return (False, "")
+
+
+_URL_REGEX = re.compile(
+    r"https?://[^\s<>\"'\)\]]+",
+    re.IGNORECASE,
+)
+
+
+def extract_urls(text: str) -> list:
+    """Extrae todas las URLs (http/https) de un texto."""
+    return _URL_REGEX.findall(text)
+
+
+def is_base64(s: str) -> bool:
+    """Verifica si un string es base64 valido.
+
+    Acepta base64 estandar y URL-safe. Requiere al menos 4 caracteres.
+    """
+    if len(s) < 4:
+        return False
+    b64_pattern = re.compile(r"^[A-Za-z0-9+/\-_]*={0,2}$")
+    if not b64_pattern.match(s):
+        return False
+    try:
+        decoded = base64.b64decode(s, validate=True)
+        return len(decoded) > 0
+    except Exception:
+        try:
+            decoded = base64.urlsafe_b64decode(s)
+            return len(decoded) > 0
+        except Exception:
+            return False
+
+
+def is_hex(s: str) -> bool:
+    """Verifica si un string es hexadecimal valido.
+
+    Acepta con o sin prefijo 0x. Requiere al menos 2 caracteres (sin prefijo).
+    """
+    clean = s.strip()
+    if clean.startswith(("0x", "0X")):
+        clean = clean[2:]
+    if len(clean) < 2:
+        return False
+    return bool(re.fullmatch(r"[0-9a-fA-F]+", clean))
+
+
+def levenshtein_distance(a: str, b: str) -> int:
+    """Calcula la distancia de Levenshtein (edit distance) entre dos strings.
+
+    Util para deteccion de typosquatting en dominios y fuzzy matching.
+    """
+    if len(a) < len(b):
+        return levenshtein_distance(b, a)
+    if len(b) == 0:
+        return len(a)
+
+    prev_row = list(range(len(b) + 1))
+    for i, ca in enumerate(a):
+        curr_row = [i + 1]
+        for j, cb in enumerate(b):
+            cost = 0 if ca == cb else 1
+            curr_row.append(
+                min(
+                    curr_row[j] + 1,       # insert
+                    prev_row[j + 1] + 1,   # delete
+                    prev_row[j] + cost,     # substitute
+                )
+            )
+        prev_row = curr_row
+    return prev_row[-1]
+
+
+def jaccard_similarity(a: list, b: list) -> float:
+    """Calcula el coeficiente de similitud de Jaccard entre dos listas.
+
+    J(A,B) = |A interseccion B| / |A union B|. Retorna 0.0 si ambas vacias.
+    Util para comparar conjuntos de tokens, features, o IoCs.
+    """
+    set_a = set(a)
+    set_b = set(b)
+    if not set_a and not set_b:
+        return 0.0
+    intersection = set_a & set_b
+    union = set_a | set_b
+    return len(intersection) / len(union)
+
+
+def normalize_url(raw_url: str) -> str:
+    """Normaliza una URL: lowercase del host, elimina fragmentos, ordena parametros.
+
+    Util para deduplicacion de URLs y comparacion de IoCs.
+    """
+    parsed = urlparse(raw_url)
+    scheme = parsed.scheme.lower() or "http"
+    netloc = parsed.netloc.lower()
+    path = parsed.path or "/"
+    # Remove trailing slash except for root
+    if path != "/" and path.endswith("/"):
+        path = path.rstrip("/")
+    # Sort query parameters
+    params = parse_qs(parsed.query, keep_blank_values=True)
+    sorted_query = urlencode(sorted(params.items()), doseq=True)
+    # Drop fragment
+    return urlunparse((scheme, netloc, path, parsed.params, sorted_query, ""))

python/functions/cybersecurity/detect_sql_injection.md

@@ -0,0 +1,38 @@
+---
+name: detect_sql_injection
+kind: function
+lang: py
+domain: cybersecurity
+version: "1.0.0"
+purity: pure
+signature: "def detect_sql_injection(input_str: str) -> tuple"
+description: "Detecta patrones de SQL injection en un string. Retorna (is_threat, pattern) con el nombre del patron detectado."
+tags: [sqli, injection, detection, security, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [re]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/cybersecurity/cybersecurity.py"
+---
+
+## Ejemplo
+
+```python
+detect_sql_injection("' OR '1'='1")
+# (True, "string_tautology")
+
+detect_sql_injection("; DROP TABLE users")
+# (True, "stacked_query")
+
+detect_sql_injection("hello world")
+# (False, "")
+```
+
+## Notas
+
+Detecta 10 patrones: sql_keyword, tautology, stacked_query, comment_injection, string_tautology, union_select, hex_literal, char_function, concat_function, time_based. No reemplaza un WAF pero es util para logging y alertas tempranas.

python/functions/cybersecurity/entropy_shannon.md

@@ -0,0 +1,41 @@
+---
+name: entropy_shannon
+kind: function
+lang: py
+domain: cybersecurity
+version: "1.0.0"
+purity: pure
+signature: "def entropy_shannon(data: bytes) -> float"
+description: "Calcula la entropia de Shannon de datos binarios (0-8 bits por byte). Util para detectar datos cifrados o comprimidos."
+tags: [entropy, shannon, analysis, crypto, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [math, collections]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/cybersecurity/cybersecurity.py"
+---
+
+## Ejemplo
+
+```python
+# Datos aleatorios (alta entropia)
+entropy_shannon(bytes(range(256)))
+# ~8.0
+
+# Datos repetitivos (baja entropia)
+entropy_shannon(b"aaaaaaaaaa")
+# 0.0
+
+# Texto normal
+entropy_shannon(b"hello world")
+# ~2.84
+```
+
+## Notas
+
+Entropia alta (>7.5) sugiere datos cifrados o comprimidos. Entropia baja (<3) sugiere datos estructurados o repetitivos. Retorna 0.0 para datos vacios.

python/functions/cybersecurity/extract_urls.md

@@ -0,0 +1,35 @@
+---
+name: extract_urls
+kind: function
+lang: py
+domain: cybersecurity
+version: "1.0.0"
+purity: pure
+signature: "def extract_urls(text: str) -> list"
+description: "Extrae todas las URLs (http/https) de un texto. Util para analisis de IoCs y threat intelligence."
+tags: [url, extract, parsing, ioc, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [re]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/cybersecurity/cybersecurity.py"
+---
+
+## Ejemplo
+
+```python
+extract_urls("Visit https://example.com and http://test.org/path?q=1")
+# ["https://example.com", "http://test.org/path?q=1"]
+
+extract_urls("no urls here")
+# []
+```
+
+## Notas
+
+Usa regex para extraer URLs con esquema http/https. No valida que las URLs sean alcanzables. Util para extraer indicadores de compromiso (IoCs) de logs, emails o reportes de threat intelligence.

python/functions/cybersecurity/hash_md5.md

@@ -0,0 +1,32 @@
+---
+name: hash_md5
+kind: function
+lang: py
+domain: cybersecurity
+version: "1.0.0"
+purity: pure
+signature: "def hash_md5(data: bytes) -> str"
+description: "Calcula el hash MD5 de datos binarios. Retorna hex digest."
+tags: [hash, md5, crypto, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [hashlib]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/cybersecurity/cybersecurity.py"
+---
+
+## Ejemplo
+
+```python
+h = hash_md5(b"hello")
+# "5d41402abc4b2a76b9719d911017c592"
+```
+
+## Notas
+
+Usa hashlib de stdlib. MD5 no es seguro para propositos criptograficos pero es util para checksums, fingerprinting de archivos e identificacion rapida de IoCs.

python/functions/cybersecurity/hash_sha256.md

@@ -0,0 +1,32 @@
+---
+name: hash_sha256
+kind: function
+lang: py
+domain: cybersecurity
+version: "1.0.0"
+purity: pure
+signature: "def hash_sha256(data: bytes) -> str"
+description: "Calcula el hash SHA-256 de datos binarios. Retorna hex digest."
+tags: [hash, sha256, crypto, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [hashlib]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/cybersecurity/cybersecurity.py"
+---
+
+## Ejemplo
+
+```python
+h = hash_sha256(b"hello")
+# "2cf24dba5fb0a30e26e83b2ac5b9e29e1b161e5c1fa7425e73043362938b9824"
+```
+
+## Notas
+
+Usa hashlib de stdlib. Funcion pura sin side effects. SHA-256 produce un digest de 64 caracteres hexadecimales (256 bits).

python/functions/cybersecurity/is_base64.md

@@ -0,0 +1,38 @@
+---
+name: is_base64
+kind: function
+lang: py
+domain: cybersecurity
+version: "1.0.0"
+purity: pure
+signature: "def is_base64(s: str) -> bool"
+description: "Verifica si un string es base64 valido. Acepta base64 estandar y URL-safe. Requiere minimo 4 caracteres."
+tags: [base64, validation, encoding, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [re, base64]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/cybersecurity/cybersecurity.py"
+---
+
+## Ejemplo
+
+```python
+is_base64("aGVsbG8=")
+# True
+
+is_base64("not!valid")
+# False
+
+is_base64("ab")
+# False (menos de 4 caracteres)
+```
+
+## Notas
+
+Verifica tanto el formato (regex) como que el decode sea exitoso. Util para detectar datos codificados en payloads sospechosos, headers HTTP o parametros de URL.

python/functions/cybersecurity/is_hex.md

@@ -0,0 +1,41 @@
+---
+name: is_hex
+kind: function
+lang: py
+domain: cybersecurity
+version: "1.0.0"
+purity: pure
+signature: "def is_hex(s: str) -> bool"
+description: "Verifica si un string es hexadecimal valido. Acepta con o sin prefijo 0x. Requiere minimo 2 caracteres."
+tags: [hex, validation, encoding, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [re]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/cybersecurity/cybersecurity.py"
+---
+
+## Ejemplo
+
+```python
+is_hex("4a6f686e")
+# True
+
+is_hex("0x4a6f686e")
+# True
+
+is_hex("xyz")
+# False
+
+is_hex("a")
+# False (menos de 2 caracteres)
+```
+
+## Notas
+
+Util para validar hashes, direcciones de memoria, shellcode y otros datos hexadecimales en contexto de seguridad.

python/functions/cybersecurity/jaccard_similarity.md

@@ -0,0 +1,38 @@
+---
+name: jaccard_similarity
+kind: function
+lang: py
+domain: cybersecurity
+version: "1.0.0"
+purity: pure
+signature: "def jaccard_similarity(a: list, b: list) -> float"
+description: "Calcula el coeficiente de similitud de Jaccard entre dos listas. J(A,B) = |A interseccion B| / |A union B|."
+tags: [jaccard, similarity, comparison, sets, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/cybersecurity/cybersecurity.py"
+---
+
+## Ejemplo
+
+```python
+jaccard_similarity(["a", "b", "c"], ["b", "c", "d"])
+# 0.5
+
+jaccard_similarity(["a", "b"], ["a", "b"])
+# 1.0
+
+jaccard_similarity([], [])
+# 0.0
+```
+
+## Notas
+
+Convierte las listas a sets internamente. Retorna 0.0 si ambas listas son vacias. Util para comparar conjuntos de tokens, features de malware, IoCs compartidos entre muestras, o tags de vulnerabilidades.

python/functions/cybersecurity/levenshtein_distance.md

@@ -0,0 +1,38 @@
+---
+name: levenshtein_distance
+kind: function
+lang: py
+domain: cybersecurity
+version: "1.0.0"
+purity: pure
+signature: "def levenshtein_distance(a: str, b: str) -> int"
+description: "Calcula la distancia de Levenshtein (edit distance) entre dos strings. Util para deteccion de typosquatting en dominios."
+tags: [levenshtein, distance, fuzzy, typosquatting, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/cybersecurity/cybersecurity.py"
+---
+
+## Ejemplo
+
+```python
+levenshtein_distance("google.com", "gooogle.com")
+# 1
+
+levenshtein_distance("paypal.com", "paypa1.com")
+# 1
+
+levenshtein_distance("abc", "abc")
+# 0
+```
+
+## Notas
+
+Implementacion O(n*m) con optimizacion de espacio (dos filas). Sin dependencias externas. Util para detectar dominios de typosquatting comparando contra dominios legitimos conocidos.

python/functions/cybersecurity/normalize_url.md

@@ -0,0 +1,35 @@
+---
+name: normalize_url
+kind: function
+lang: py
+domain: cybersecurity
+version: "1.0.0"
+purity: pure
+signature: "def normalize_url(raw_url: str) -> str"
+description: "Normaliza una URL: lowercase del host, elimina fragmentos, ordena parametros. Util para deduplicacion de IoCs."
+tags: [url, normalize, ioc, dedup, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [urllib.parse]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/cybersecurity/cybersecurity.py"
+---
+
+## Ejemplo
+
+```python
+normalize_url("HTTPS://Example.COM/path?b=2&a=1#frag")
+# "https://example.com/path?a=1&b=2"
+
+normalize_url("http://test.org/path/")
+# "http://test.org/path"
+```
+
+## Notas
+
+Operaciones de normalizacion: lowercase de scheme y host, eliminacion de trailing slash (excepto root), ordenamiento alfabetico de query parameters, eliminacion de fragmentos. Usa urllib.parse de stdlib.

python/functions/datascience/__init__.py

@@ -0,0 +1,25 @@
+from .datascience import (
+    pearson,
+    standardize,
+    min_max_scale,
+    clip,
+    detect_outliers,
+    impute,
+    histogram,
+    rolling_window,
+    autocorrelation,
+    linspace,
+)
+
+__all__ = [
+    "pearson",
+    "standardize",
+    "min_max_scale",
+    "clip",
+    "detect_outliers",
+    "impute",
+    "histogram",
+    "rolling_window",
+    "autocorrelation",
+    "linspace",
+]

python/functions/datascience/autocorrelation.md

@@ -0,0 +1,32 @@
+---
+name: autocorrelation
+kind: function
+lang: py
+domain: datascience
+version: "1.0.0"
+purity: pure
+signature: "def autocorrelation(data: list, lag: int) -> float"
+description: "Calcula la autocorrelacion de una serie temporal para un lag dado."
+tags: [statistics, timeseries, correlation, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/datascience/datascience.py"
+---
+
+## Ejemplo
+
+```python
+autocorrelation([1, 2, 3, 4, 5, 4, 3, 2, 1], 1)
+# ~0.489
+```
+
+## Notas
+
+Autocorrelacion normalizada por la varianza. Retorna 0.0 si lag es invalido o la varianza es cero.

python/functions/datascience/clip.md

@@ -0,0 +1,32 @@
+---
+name: clip
+kind: function
+lang: py
+domain: datascience
+version: "1.0.0"
+purity: pure
+signature: "def clip(data: list, lo: float, hi: float) -> list"
+description: "Recorta los valores de la lista al rango [lo, hi]."
+tags: [clipping, bounds, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/datascience/datascience.py"
+---
+
+## Ejemplo
+
+```python
+clip([1, 5, 10, -3], 0, 7)
+# [1, 5, 7, 0]
+```
+
+## Notas
+
+Funcion pura sin dependencias. Equivalente a numpy.clip pero sin numpy.

python/functions/datascience/datascience.py

@@ -0,0 +1,123 @@
+"""Pure datascience utilities — statistics and numerical functions.
+
+Uses only math stdlib. No external dependencies.
+"""
+
+import math
+
+
+def pearson(xs: list, ys: list) -> float:
+    """Pearson correlation coefficient between two lists of floats."""
+    n = len(xs)
+    if n != len(ys) or n == 0:
+        return 0.0
+    mean_x = sum(xs) / n
+    mean_y = sum(ys) / n
+    num = sum((x - mean_x) * (y - mean_y) for x, y in zip(xs, ys))
+    den_x = math.sqrt(sum((x - mean_x) ** 2 for x in xs))
+    den_y = math.sqrt(sum((y - mean_y) ** 2 for y in ys))
+    if den_x == 0.0 or den_y == 0.0:
+        return 0.0
+    return num / (den_x * den_y)
+
+
+def standardize(data: list) -> list:
+    """Z-score standardization (mean=0, std=1)."""
+    n = len(data)
+    if n == 0:
+        return []
+    mean = sum(data) / n
+    std = math.sqrt(sum((x - mean) ** 2 for x in data) / n)
+    if std == 0.0:
+        return [0.0] * n
+    return [(x - mean) / std for x in data]
+
+
+def min_max_scale(data: list) -> list:
+    """Scale values to [0, 1] range."""
+    if not data:
+        return []
+    lo = min(data)
+    hi = max(data)
+    if hi == lo:
+        return [0.0] * len(data)
+    return [(x - lo) / (hi - lo) for x in data]
+
+
+def clip(data: list, lo: float, hi: float) -> list:
+    """Clip values to [lo, hi]."""
+    return [max(lo, min(hi, x)) for x in data]
+
+
+def detect_outliers(data: list, threshold: float) -> list:
+    """Returns list of bools, True where |z-score| > threshold."""
+    n = len(data)
+    if n == 0:
+        return []
+    mean = sum(data) / n
+    std = math.sqrt(sum((x - mean) ** 2 for x in data) / n)
+    if std == 0.0:
+        return [False] * n
+    return [abs((x - mean) / std) > threshold for x in data]
+
+
+def impute(data: list) -> list:
+    """Replace None/NaN with mean of non-null values."""
+    valid = [x for x in data if x is not None and not (isinstance(x, float) and math.isnan(x))]
+    if not valid:
+        return [0.0] * len(data)
+    mean = sum(valid) / len(valid)
+    return [
+        mean if (x is None or (isinstance(x, float) and math.isnan(x))) else x
+        for x in data
+    ]
+
+
+def histogram(data: list, buckets: int) -> list:
+    """Returns list of counts per bucket."""
+    if not data or buckets <= 0:
+        return []
+    lo = min(data)
+    hi = max(data)
+    if hi == lo:
+        counts = [0] * buckets
+        counts[0] = len(data)
+        return counts
+    width = (hi - lo) / buckets
+    counts = [0] * buckets
+    for x in data:
+        idx = int((x - lo) / width)
+        if idx >= buckets:
+            idx = buckets - 1
+        counts[idx] += 1
+    return counts
+
+
+def rolling_window(xs: list, size: int) -> list:
+    """Returns list of sublists (sliding windows of given size)."""
+    if size <= 0 or size > len(xs):
+        return []
+    return [xs[i : i + size] for i in range(len(xs) - size + 1)]
+
+
+def autocorrelation(data: list, lag: int) -> float:
+    """Autocorrelation at given lag."""
+    n = len(data)
+    if lag < 0 or lag >= n or n == 0:
+        return 0.0
+    mean = sum(data) / n
+    var = sum((x - mean) ** 2 for x in data) / n
+    if var == 0.0:
+        return 0.0
+    cov = sum((data[i] - mean) * (data[i + lag] - mean) for i in range(n - lag)) / n
+    return cov / var
+
+
+def linspace(start: float, stop: float, num: int) -> list:
+    """Generate evenly spaced values from start to stop (inclusive)."""
+    if num <= 0:
+        return []
+    if num == 1:
+        return [start]
+    step = (stop - start) / (num - 1)
+    return [start + i * step for i in range(num)]

python/functions/datascience/detect_outliers.md

@@ -0,0 +1,32 @@
+---
+name: detect_outliers
+kind: function
+lang: py
+domain: datascience
+version: "1.0.0"
+purity: pure
+signature: "def detect_outliers(data: list, threshold: float) -> list"
+description: "Detecta outliers por z-score. Retorna lista de bools, True donde |z-score| > threshold."
+tags: [statistics, outliers, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [math]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/datascience/datascience.py"
+---
+
+## Ejemplo
+
+```python
+detect_outliers([1, 2, 3, 100, 2, 3], 2.0)
+# [False, False, False, True, False, False]
+```
+
+## Notas
+
+Usa z-score poblacional. Threshold tipico: 2.0 o 3.0. Si la desviacion es cero, no hay outliers.

python/functions/datascience/histogram.md

@@ -0,0 +1,32 @@
+---
+name: histogram
+kind: function
+lang: py
+domain: datascience
+version: "1.0.0"
+purity: pure
+signature: "def histogram(data: list, buckets: int) -> list"
+description: "Calcula histograma con N buckets. Retorna lista de conteos por bucket."
+tags: [statistics, histogram, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/datascience/datascience.py"
+---
+
+## Ejemplo
+
+```python
+histogram([1, 2, 3, 4, 5, 6, 7, 8, 9, 10], 5)
+# [2, 2, 2, 2, 2]
+```
+
+## Notas
+
+Los buckets cubren el rango [min, max] uniformemente. El ultimo bucket incluye el valor maximo. Si todos los valores son iguales, todos caen en el primer bucket.

python/functions/datascience/impute.md

@@ -0,0 +1,32 @@
+---
+name: impute
+kind: function
+lang: py
+domain: datascience
+version: "1.0.0"
+purity: pure
+signature: "def impute(data: list) -> list"
+description: "Reemplaza None y NaN con la media de los valores validos."
+tags: [imputation, missing, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [math]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/datascience/datascience.py"
+---
+
+## Ejemplo
+
+```python
+impute([1.0, None, 3.0, float('nan'), 5.0])
+# [1.0, 3.0, 3.0, 3.0, 5.0]
+```
+
+## Notas
+
+Detecta tanto None como float('nan'). Si no hay valores validos, rellena con 0.0.

python/functions/datascience/linspace.md

@@ -0,0 +1,32 @@
+---
+name: linspace
+kind: function
+lang: py
+domain: datascience
+version: "1.0.0"
+purity: pure
+signature: "def linspace(start: float, stop: float, num: int) -> list"
+description: "Genera una lista de valores equiespaciados entre start y stop (inclusivos)."
+tags: [numeric, range, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/datascience/datascience.py"
+---
+
+## Ejemplo
+
+```python
+linspace(0, 1, 5)
+# [0.0, 0.25, 0.5, 0.75, 1.0]
+```
+
+## Notas
+
+Equivalente a numpy.linspace pero sin numpy. Si num=1, retorna [start]. Si num<=0, retorna lista vacia.

python/functions/datascience/min_max_scale.md

@@ -0,0 +1,32 @@
+---
+name: min_max_scale
+kind: function
+lang: py
+domain: datascience
+version: "1.0.0"
+purity: pure
+signature: "def min_max_scale(data: list) -> list"
+description: "Escala los valores al rango [0, 1] usando min-max normalization."
+tags: [normalization, scaling, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/datascience/datascience.py"
+---
+
+## Ejemplo
+
+```python
+min_max_scale([2, 4, 6, 8, 10])
+# [0.0, 0.25, 0.5, 0.75, 1.0]
+```
+
+## Notas
+
+Si todos los valores son iguales, retorna lista de ceros. No requiere imports externos.

python/functions/datascience/pearson.md

@@ -0,0 +1,32 @@
+---
+name: pearson
+kind: function
+lang: py
+domain: datascience
+version: "1.0.0"
+purity: pure
+signature: "def pearson(xs: list, ys: list) -> float"
+description: "Calcula el coeficiente de correlacion de Pearson entre dos listas de floats."
+tags: [statistics, correlation, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [math]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/datascience/datascience.py"
+---
+
+## Ejemplo
+
+```python
+r = pearson([1, 2, 3], [2, 4, 6])
+# r = 1.0
+```
+
+## Notas
+
+Usa solo math stdlib. No requiere numpy. Retorna 0.0 si las listas tienen longitud diferente, estan vacias, o la desviacion es cero.

python/functions/datascience/rolling_window.md

@@ -0,0 +1,32 @@
+---
+name: rolling_window
+kind: function
+lang: py
+domain: datascience
+version: "1.0.0"
+purity: pure
+signature: "def rolling_window(xs: list, size: int) -> list"
+description: "Genera ventanas deslizantes de tamanio fijo sobre una lista."
+tags: [windowing, timeseries, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/datascience/datascience.py"
+---
+
+## Ejemplo
+
+```python
+rolling_window([1, 2, 3, 4, 5], 3)
+# [[1, 2, 3], [2, 3, 4], [3, 4, 5]]
+```
+
+## Notas
+
+Retorna lista vacia si size <= 0 o size > len(xs). Util para calcular medias moviles u otras metricas sobre ventanas.

python/functions/datascience/standardize.md

@@ -0,0 +1,32 @@
+---
+name: standardize
+kind: function
+lang: py
+domain: datascience
+version: "1.0.0"
+purity: pure
+signature: "def standardize(data: list) -> list"
+description: "Estandarizacion Z-score: transforma los datos a media=0 y desviacion=1."
+tags: [statistics, normalization, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [math]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/datascience/datascience.py"
+---
+
+## Ejemplo
+
+```python
+standardize([10, 20, 30])
+# [-1.2247..., 0.0, 1.2247...]
+```
+
+## Notas
+
+Si la desviacion estandar es cero, retorna lista de ceros. Usa desviacion poblacional (N, no N-1).

python/functions/finance/annualized_volatility.md

@@ -0,0 +1,35 @@
+---
+name: annualized_volatility
+kind: function
+lang: py
+domain: finance
+version: "1.0.0"
+purity: pure
+signature: "def annualized_volatility(returns: list, periods_per_year: float) -> float"
+description: "Calcula la volatilidad anualizada de una serie de retornos."
+tags: [finance, volatility, risk, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [math]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/finance/finance.py"
+---
+
+## Ejemplo
+
+```python
+daily_returns = [0.01, -0.005, 0.008, 0.003, -0.002, 0.006, 0.004]
+vol = annualized_volatility(daily_returns, 252.0)
+# Volatilidad anualizada (std * sqrt(252))
+```
+
+## Notas
+
+Formula: std_muestral(returns) * sqrt(periods_per_year).
+Usa desviacion estandar muestral (n-1) para ser consistente con la practica financiera.
+Retorna 0.0 si hay menos de 2 retornos o periods_per_year es menor o igual a cero.

python/functions/finance/bollinger_bands.md

@@ -0,0 +1,35 @@
+---
+name: bollinger_bands
+kind: function
+lang: py
+domain: finance
+version: "1.0.0"
+purity: pure
+signature: "def bollinger_bands(data: list, period: int, num_std: float) -> tuple"
+description: "Calcula las Bandas de Bollinger (upper, middle, lower) de una serie de precios."
+tags: [finance, bollinger, volatility, indicator, python]
+uses_functions: [sma_py_finance]
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/finance/finance.py"
+---
+
+## Ejemplo
+
+```python
+prices = [10, 11, 12, 13, 14, 15, 14, 13, 12, 11]
+upper, middle, lower = bollinger_bands(prices, 5, 2.0)
+# middle es la SMA(5), upper/lower son middle +/- 2*std
+```
+
+## Notas
+
+Retorna tupla de tres listas (upper, middle, lower). Cada lista tiene len(data) - period + 1 elementos.
+La desviacion estandar se calcula sobre la ventana de tamanio period (poblacional, no muestral).
+Usa internamente la funcion sma para la banda media.

python/functions/finance/ema.md

@@ -0,0 +1,34 @@
+---
+name: ema
+kind: function
+lang: py
+domain: finance
+version: "1.0.0"
+purity: pure
+signature: "def ema(data: list, period: int) -> list"
+description: "Calcula la media movil exponencial (EMA) de una serie de precios."
+tags: [finance, moving-average, exponential, indicator, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/finance/finance.py"
+---
+
+## Ejemplo
+
+```python
+prices = [10, 11, 12, 13, 14, 15]
+result = ema(prices, 3)
+# [11.0, 11.5, 12.25, 13.125]
+```
+
+## Notas
+
+El primer valor de la EMA es el SMA del primer periodo. El multiplicador es 2 / (period + 1).
+Retorna len(data) - period + 1 elementos. Lista vacia si period invalido.

python/functions/finance/finance.py

@@ -0,0 +1,137 @@
+"""Finance domain — pure functions for financial indicators and calculations."""
+
+import math
+
+
+def sma(data: list, period: int) -> list:
+    """Calcula la media movil simple (SMA) de una serie de precios."""
+    if period <= 0 or period > len(data):
+        return []
+    result = []
+    for i in range(period - 1, len(data)):
+        window = data[i - period + 1 : i + 1]
+        result.append(sum(window) / period)
+    return result
+
+
+def ema(data: list, period: int) -> list:
+    """Calcula la media movil exponencial (EMA) de una serie de precios."""
+    if period <= 0 or period > len(data):
+        return []
+    multiplier = 2.0 / (period + 1)
+    # Primer valor es SMA del primer periodo
+    first_sma = sum(data[:period]) / period
+    result = [first_sma]
+    for i in range(period, len(data)):
+        val = (data[i] - result[-1]) * multiplier + result[-1]
+        result.append(val)
+    return result
+
+
+def rsi(data: list, period: int) -> list:
+    """Calcula el Relative Strength Index (RSI) de una serie de precios."""
+    if period <= 0 or len(data) < period + 1:
+        return []
+    deltas = [data[i] - data[i - 1] for i in range(1, len(data))]
+    gains = [d if d > 0 else 0.0 for d in deltas]
+    losses = [-d if d < 0 else 0.0 for d in deltas]
+
+    avg_gain = sum(gains[:period]) / period
+    avg_loss = sum(losses[:period]) / period
+
+    result = []
+    if avg_loss == 0:
+        result.append(100.0)
+    else:
+        rs = avg_gain / avg_loss
+        result.append(100.0 - 100.0 / (1.0 + rs))
+
+    for i in range(period, len(deltas)):
+        avg_gain = (avg_gain * (period - 1) + gains[i]) / period
+        avg_loss = (avg_loss * (period - 1) + losses[i]) / period
+        if avg_loss == 0:
+            result.append(100.0)
+        else:
+            rs = avg_gain / avg_loss
+            result.append(100.0 - 100.0 / (1.0 + rs))
+
+    return result
+
+
+def bollinger_bands(data: list, period: int, num_std: float) -> tuple:
+    """Calcula las Bandas de Bollinger (upper, middle, lower)."""
+    if period <= 0 or period > len(data):
+        return ([], [], [])
+    middle = sma(data, period)
+    upper = []
+    lower = []
+    for i in range(len(middle)):
+        window = data[i : i + period]
+        mean = middle[i]
+        variance = sum((x - mean) ** 2 for x in window) / period
+        std = math.sqrt(variance)
+        upper.append(mean + num_std * std)
+        lower.append(mean - num_std * std)
+    return (upper, middle, lower)
+
+
+def sharpe_ratio(returns: list, risk_free_rate: float, periods_per_year: float) -> float:
+    """Calcula el Sharpe Ratio anualizado."""
+    if len(returns) == 0 or periods_per_year <= 0:
+        return 0.0
+    n = len(returns)
+    mean_return = sum(returns) / n
+    excess = mean_return - risk_free_rate / periods_per_year
+    variance = sum((r - mean_return) ** 2 for r in returns) / n
+    std = math.sqrt(variance)
+    if std == 0:
+        return 0.0
+    return (excess / std) * math.sqrt(periods_per_year)
+
+
+def max_drawdown(values: list) -> tuple:
+    """Calcula el max drawdown y los indices de inicio y fin."""
+    if len(values) < 2:
+        return (0.0, 0, 0)
+    peak = values[0]
+    peak_idx = 0
+    max_dd = 0.0
+    dd_start = 0
+    dd_end = 0
+    for i in range(1, len(values)):
+        if values[i] > peak:
+            peak = values[i]
+            peak_idx = i
+        dd = (peak - values[i]) / peak if peak != 0 else 0.0
+        if dd > max_dd:
+            max_dd = dd
+            dd_start = peak_idx
+            dd_end = i
+    return (max_dd, dd_start, dd_end)
+
+
+def vwap(prices: list, volumes: list) -> float:
+    """Calcula el Volume-Weighted Average Price (VWAP)."""
+    if len(prices) == 0 or len(prices) != len(volumes):
+        return 0.0
+    total_volume = sum(volumes)
+    if total_volume == 0:
+        return 0.0
+    return sum(p * v for p, v in zip(prices, volumes)) / total_volume
+
+
+def log_return(price_start: float, price_end: float) -> float:
+    """Calcula el retorno logaritmico entre dos precios."""
+    if price_start <= 0 or price_end <= 0:
+        return 0.0
+    return math.log(price_end / price_start)
+
+
+def annualized_volatility(returns: list, periods_per_year: float) -> float:
+    """Calcula la volatilidad anualizada de una serie de retornos."""
+    if len(returns) < 2 or periods_per_year <= 0:
+        return 0.0
+    n = len(returns)
+    mean = sum(returns) / n
+    variance = sum((r - mean) ** 2 for r in returns) / (n - 1)
+    return math.sqrt(variance) * math.sqrt(periods_per_year)

python/functions/finance/log_return.md

@@ -0,0 +1,34 @@
+---
+name: log_return
+kind: function
+lang: py
+domain: finance
+version: "1.0.0"
+purity: pure
+signature: "def log_return(price_start: float, price_end: float) -> float"
+description: "Calcula el retorno logaritmico entre dos precios."
+tags: [finance, return, logarithmic, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [math]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/finance/finance.py"
+---
+
+## Ejemplo
+
+```python
+r = log_return(100.0, 110.0)
+# 0.09531... (aprox 9.53%)
+```
+
+## Notas
+
+Formula: ln(price_end / price_start).
+Retorna 0.0 si alguno de los precios es menor o igual a cero.
+Los retornos logaritmicos son aditivos en el tiempo, a diferencia de los retornos simples.

python/functions/finance/max_drawdown.md

@@ -0,0 +1,35 @@
+---
+name: max_drawdown
+kind: function
+lang: py
+domain: finance
+version: "1.0.0"
+purity: pure
+signature: "def max_drawdown(values: list) -> tuple"
+description: "Calcula el maximo drawdown y los indices de inicio y fin del peor periodo."
+tags: [finance, drawdown, risk, performance, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/finance/finance.py"
+---
+
+## Ejemplo
+
+```python
+portfolio = [100, 110, 105, 95, 102, 108, 90, 95]
+dd, start, end = max_drawdown(portfolio)
+# dd = 0.1818..., start = 1, end = 6  (de 110 a 90)
+```
+
+## Notas
+
+Retorna tupla (max_dd, start_idx, end_idx) donde max_dd es fraccion (0.0 a 1.0).
+start_idx es el indice del pico previo, end_idx es el indice del valle.
+Retorna (0.0, 0, 0) si la lista tiene menos de 2 elementos.

python/functions/finance/rsi.md

@@ -0,0 +1,36 @@
+---
+name: rsi
+kind: function
+lang: py
+domain: finance
+version: "1.0.0"
+purity: pure
+signature: "def rsi(data: list, period: int) -> list"
+description: "Calcula el Relative Strength Index (RSI) de una serie de precios."
+tags: [finance, rsi, momentum, indicator, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/finance/finance.py"
+---
+
+## Ejemplo
+
+```python
+prices = [44, 44.34, 44.09, 43.61, 44.33, 44.83, 45.10, 45.42, 45.84, 46.08,
+          45.89, 46.03, 45.61, 46.28, 46.28, 46.00, 46.03, 46.41, 46.22, 45.64]
+result = rsi(prices, 14)
+# Lista de valores RSI entre 0 y 100
+```
+
+## Notas
+
+Usa el metodo de suavizado de Wilder (media exponencial modificada).
+Requiere al menos period + 1 datos de entrada. Retorna len(data) - period valores.
+RSI = 100 si no hay perdidas en el periodo (avg_loss == 0).

python/functions/finance/sharpe_ratio.md

@@ -0,0 +1,35 @@
+---
+name: sharpe_ratio
+kind: function
+lang: py
+domain: finance
+version: "1.0.0"
+purity: pure
+signature: "def sharpe_ratio(returns: list, risk_free_rate: float, periods_per_year: float) -> float"
+description: "Calcula el Sharpe Ratio anualizado de una serie de retornos."
+tags: [finance, sharpe, risk, performance, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: [math]
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/finance/finance.py"
+---
+
+## Ejemplo
+
+```python
+daily_returns = [0.01, -0.005, 0.008, 0.003, -0.002, 0.006, 0.004]
+sr = sharpe_ratio(daily_returns, 0.02, 252.0)
+# Sharpe ratio anualizado
+```
+
+## Notas
+
+risk_free_rate es la tasa anual (ej: 0.02 para 2%). Se convierte a tasa por periodo internamente.
+periods_per_year indica la frecuencia de los retornos (252 para diarios, 12 para mensuales).
+Retorna 0.0 si la desviacion estandar es cero o la lista esta vacia.

python/functions/finance/sma.md

@@ -0,0 +1,34 @@
+---
+name: sma
+kind: function
+lang: py
+domain: finance
+version: "1.0.0"
+purity: pure
+signature: "def sma(data: list, period: int) -> list"
+description: "Calcula la media movil simple (SMA) de una serie de precios."
+tags: [finance, moving-average, indicator, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/finance/finance.py"
+---
+
+## Ejemplo
+
+```python
+prices = [10, 11, 12, 13, 14, 15]
+result = sma(prices, 3)
+# [11.0, 12.0, 13.0, 14.0]
+```
+
+## Notas
+
+Retorna lista mas corta que la entrada (len - period + 1 elementos).
+Si period es mayor que len(data) o menor/igual a 0, retorna lista vacia.

python/functions/finance/vwap.md

@@ -0,0 +1,35 @@
+---
+name: vwap
+kind: function
+lang: py
+domain: finance
+version: "1.0.0"
+purity: pure
+signature: "def vwap(prices: list, volumes: list) -> float"
+description: "Calcula el Volume-Weighted Average Price (VWAP)."
+tags: [finance, vwap, volume, indicator, python]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: false
+tests: []
+test_file_path: ""
+file_path: "python/functions/finance/finance.py"
+---
+
+## Ejemplo
+
+```python
+prices = [100.0, 101.0, 102.0, 101.5]
+volumes = [1000, 1500, 1200, 800]
+result = vwap(prices, volumes)
+# 101.0888...
+```
+
+## Notas
+
+Formula: sum(price_i * volume_i) / sum(volume_i).
+Retorna 0.0 si las listas estan vacias, tienen distinto tamanio, o el volumen total es cero.