diff --git a/backend/api/v1/modules/a76/clients_and_providers/imports/__init__.py b/backend/api/v1/modules/a76/clients_and_providers/imports/__init__.py
new file mode 100644
index 00000000..94b7cb55
--- /dev/null
+++ b/backend/api/v1/modules/a76/clients_and_providers/imports/__init__.py
@@ -0,0 +1,2 @@
+# CSV import flow for Clientes y Proveedores (Client Providers).
+# Replicates the same two-phase flow as customs_brokers/imports: upload → scan → waiting_confirmation → commit.
diff --git a/backend/api/v1/modules/a76/clients_and_providers/imports/routes.py b/backend/api/v1/modules/a76/clients_and_providers/imports/routes.py
new file mode 100644
index 00000000..68029417
--- /dev/null
+++ b/backend/api/v1/modules/a76/clients_and_providers/imports/routes.py
@@ -0,0 +1,161 @@
+"""
+Rutas de importación CSV para Clientes y Proveedores.
+Mismo flujo que a76.imports: upload → scan → status (polling) → commit.
+"""
+import base64
+import json
+import logging
+import os
+from uuid import uuid4
+
+from fastapi import APIRouter, File, HTTPException, Query, UploadFile, Depends
+from sqlalchemy.orm import Session
+from typing import Dict, Any
+
+from core.celery_app import celery_app
+from core.database import get_core_db
+from core.security import get_current_user, validate_access_to_resource
+
+from .schemas import ImportJobResponse
+from .tasks import (
+    scan_file,
+    insert_valid_rows,
+    CP_IMPORT_FILE_PREFIX,
+    CP_IMPORT_META_PREFIX,
+    CP_IMPORT_REDIS_TTL,
+)
+
+router = APIRouter()
+logger = logging.getLogger(__name__)
+
+
+def _get_redis():
+    import redis
+    url = os.getenv("VALKEY_URL", os.getenv("REDIS_URL", "redis://valkey:6379/0"))
+    return redis.Redis.from_url(url, decode_responses=False)
+
+
+@router.post("/upload", response_model=ImportJobResponse)
+async def upload_import_file(
+    file: UploadFile = File(...),
+    company_id: int = Query(..., description="Company ID"),
+    db: Session = Depends(get_core_db),
+    current_user: Dict[str, Any] = Depends(get_current_user),
+):
+    """
+    Fase 1: Subir CSV, guardar en Redis, encolar tarea de escaneo.
+    """
+    try:
+        tenant_id = validate_access_to_resource(db, company_id, current_user)
+    except Exception as e:
+        logger.error(f"CP import: access validation failed: {e}")
+        raise HTTPException(status_code=403, detail="Invalid company access")
+
+    if not file.filename or not file.filename.lower().endswith(".csv"):
+        raise HTTPException(status_code=400, detail="Solo se permiten archivos .csv")
+
+    job_id = str(uuid4())
+    contents = await file.read()
+
+    meta_data = {
+        "tenant_id": tenant_id,
+        "company_id": company_id,
+        "user_id": current_user.get("id"),
+        "template_id": "client_providers",
+    }
+
+    try:
+        r = _get_redis()
+        r.set(
+            f"{CP_IMPORT_FILE_PREFIX}{job_id}",
+            base64.b64encode(contents),
+            ex=CP_IMPORT_REDIS_TTL,
+        )
+        r.set(
+            f"{CP_IMPORT_META_PREFIX}{job_id}",
+            json.dumps(meta_data).encode("utf-8"),
+            ex=CP_IMPORT_REDIS_TTL,
+        )
+    except Exception as e:
+        logger.error(f"CP import: Redis store error: {e}")
+        raise HTTPException(status_code=500, detail="No se pudo encolar el archivo.")
+
+    try:
+        upload_dir = os.path.join(os.getcwd(), "uploads", "temp")
+        os.makedirs(upload_dir, exist_ok=True)
+        with open(os.path.join(upload_dir, f"cp_{job_id}.csv"), "wb") as f:
+            f.write(contents)
+        with open(os.path.join(upload_dir, f"cp_{job_id}.meta.json"), "w") as f:
+            json.dump(meta_data, f)
+    except Exception as e:
+        logger.warning(f"CP import: local file save failed: {e}")
+
+    scan_file.apply_async(args=[job_id], task_id=job_id)
+
+    return ImportJobResponse(
+        job_id=job_id,
+        status="queued",
+        message="Archivo subido. Escaneo iniciado.",
+    )
+
+
+@router.get("/{job_id}/status")
+async def get_import_status(job_id: str):
+    """
+    Polling: estado del escaneo o del commit.
+    """
+    task_result = celery_app.AsyncResult(job_id)
+
+    if task_result.state == "PENDING":
+        return {"status": "processing", "progress": 0}
+    if task_result.state == "PROGRESS":
+        info = (task_result.info or {})
+        return {
+            "status": "processing",
+            "progress": info.get("current", 0),
+            "total": info.get("total", 0),
+        }
+    if task_result.state == "SUCCESS":
+        result = task_result.result
+        if isinstance(result, dict) and "status" in result:
+            return result
+        return {"status": "finished", "result": result}
+
+    # A veces Celery tiene el result disponible pero state aún no es SUCCESS; si el result es éxito, devolverlo
+    result = getattr(task_result, "result", None)
+    if isinstance(result, dict) and result.get("status") in ("finished", "warning"):
+        return result
+
+    logger.warning("CP import task %s failed: state=%s", job_id, task_result.state)
+    err_msg = None
+    tb = getattr(task_result, "traceback", None)
+    if tb and isinstance(tb, str):
+        lines = [l.strip() for l in tb.strip().split("\n") if l.strip()]
+        if lines:
+            err_msg = lines[-1]
+    if not err_msg:
+        try:
+            exc = task_result.get(propagate=False)
+            if exc is not None:
+                err_msg = str(exc)
+        except Exception:
+            pass
+    if not err_msg and result is not None:
+        if not isinstance(result, dict):
+            err_msg = str(result)
+        elif result.get("error") or result.get("message"):
+            err_msg = result.get("error") or result.get("message")
+    return {"status": "failed", "error": err_msg or "Task failed"}
+
+
+@router.post("/{job_id}/commit")
+async def commit_import_job(job_id: str):
+    """
+    Fase 2: Usuario confirma; se encola la inserción de filas válidas.
+    """
+    task = insert_valid_rows.delay(job_id)
+    return {
+        "status": "committing",
+        "message": "Inserción iniciada.",
+        "commit_job_id": task.id,
+    }
diff --git a/backend/api/v1/modules/a76/clients_and_providers/imports/schemas.py b/backend/api/v1/modules/a76/clients_and_providers/imports/schemas.py
new file mode 100644
index 00000000..98c39a98
--- /dev/null
+++ b/backend/api/v1/modules/a76/clients_and_providers/imports/schemas.py
@@ -0,0 +1,25 @@
+from pydantic import BaseModel
+from typing import Optional
+
+
+class ImportJobResponse(BaseModel):
+    job_id: str
+    status: str
+    message: str
+
+
+class CommitRequest(BaseModel):
+    pass  # no body needed for single model
+
+
+class ImportJobStatus(BaseModel):
+    status: str
+    job_id: str
+    total_rows: Optional[int] = 0
+    error_count: Optional[int] = 0
+    valid_rows: Optional[int] = 0
+    error: Optional[str] = None
+    inserted: Optional[int] = 0
+    skipped_invalid: Optional[int] = 0
+    skipped_missing_fk: Optional[int] = 0
+    skipped_details: Optional[list] = None
diff --git a/backend/api/v1/modules/a76/clients_and_providers/imports/tasks.py b/backend/api/v1/modules/a76/clients_and_providers/imports/tasks.py
new file mode 100644
index 00000000..34fea87c
--- /dev/null
+++ b/backend/api/v1/modules/a76/clients_and_providers/imports/tasks.py
@@ -0,0 +1,500 @@
+"""
+Tareas Celery para importación CSV de Clientes y Proveedores.
+Flujo en dos fases: scan_file (validación) → insert_valid_rows (commit).
+"""
+import os
+import base64
+import csv
+import json
+import logging
+import re
+import unicodedata
+from typing import Dict, Any, Optional, List
+
+from core.celery_app import celery_app
+from core.database import CoreSessionLocal
+
+from .template_config import row_from_template
+from api.v1.modules.a76.clients_and_providers.models import (
+    ClientProvider,
+    ClientProviderAddress,
+    ClientOrProviderEnum,
+)
+
+logger = logging.getLogger(__name__)
+
+# Redis keys (prefijo propio para no colisionar con cb_ ni a76.imports)
+CP_IMPORT_FILE_PREFIX = "cp_import_file:"
+CP_IMPORT_META_PREFIX = "cp_import_meta:"
+CP_IMPORT_ERROR_LINES_PREFIX = "cp_import_error_lines:"
+CP_IMPORT_REDIS_TTL = 3600  # 1 hour
+
+
+def _get_redis():
+    import redis
+    url = os.getenv("VALKEY_URL", os.getenv("REDIS_URL", "redis://valkey:6379/0"))
+    return redis.Redis.from_url(url, decode_responses=False)
+
+
+def _worker_upload_dir() -> str:
+    return os.path.join(os.getcwd(), "uploads", "temp")
+
+
+def _ensure_worker_has_file_from_redis(job_id: str) -> Optional[str]:
+    r = _get_redis()
+    data = r.get(f"{CP_IMPORT_FILE_PREFIX}{job_id}")
+    if not data:
+        return None
+    try:
+        raw = base64.b64decode(data)
+    except Exception as e:
+        logger.warning(f"CP import: failed to decode file from Redis: {e}")
+        return None
+    upload_dir = _worker_upload_dir()
+    os.makedirs(upload_dir, exist_ok=True)
+    file_path = os.path.join(upload_dir, f"cp_{job_id}.csv")
+    with open(file_path, "wb") as f:
+        f.write(raw)
+    return file_path
+
+
+def _ensure_worker_has_meta_from_redis(job_id: str, file_path: str) -> bool:
+    r = _get_redis()
+    data = r.get(f"{CP_IMPORT_META_PREFIX}{job_id}")
+    if not data:
+        return False
+    try:
+        meta = json.loads(data.decode("utf-8"))
+    except Exception as e:
+        logger.warning(f"CP import: failed to decode meta from Redis: {e}")
+        return False
+    meta_path = file_path.replace(".csv", ".meta.json")
+    with open(meta_path, "w", encoding="utf-8") as f:
+        json.dump(meta, f)
+    return True
+
+
+def _delete_import_from_redis(job_id: str) -> None:
+    try:
+        r = _get_redis()
+        r.delete(
+            f"{CP_IMPORT_FILE_PREFIX}{job_id}",
+            f"{CP_IMPORT_META_PREFIX}{job_id}",
+            f"{CP_IMPORT_ERROR_LINES_PREFIX}{job_id}",
+        )
+    except Exception as e:
+        logger.warning(f"CP import: failed to delete Redis keys: {e}")
+
+
+def normalize_header(name: Optional[str]) -> str:
+    if not name:
+        return ""
+    name = unicodedata.normalize("NFKD", str(name)).upper()
+    name = "".join(ch for ch in name if not unicodedata.combining(ch))
+    name = re.sub(r"[^A-Z0-9]+", " ", name)
+    return re.sub(r"\s+", " ", name).strip()
+
+
+def _parse_client_or_provider(val: Optional[str]) -> Optional[ClientOrProviderEnum]:
+    """Mapea valor CSV a ClientOrProviderEnum. Retorna None si no reconocido."""
+    if not val or not str(val).strip():
+        return None
+    v = str(val).strip().lower()
+    if v in ("client", "cliente", "c"):
+        return ClientOrProviderEnum.CLIENT
+    if v in ("provider", "proveedor", "p"):
+        return ClientOrProviderEnum.PROVIDER
+    if v in ("both", "ambos", "b", "cliente y proveedor"):
+        return ClientOrProviderEnum.BOTH
+    return None
+
+
+def _parse_active(val: Optional[str]) -> bool:
+    """Interpreta ACTIVO: 1/true/si/yes -> True, 0/false/no -> False. Default True."""
+    if not val or not str(val).strip():
+        return True
+    v = str(val).strip().lower()
+    if v in ("1", "true", "si", "sí", "yes", "s", "x"):
+        return True
+    if v in ("0", "false", "no", "n"):
+        return False
+    return True
+
+
+def _validate_row_client_provider(row: Dict[str, Any], line_num: int) -> Optional[Dict[str, Any]]:
+    """Valida una fila para Cliente/Proveedor. Retorna error dict o None."""
+    rfc = (row.get("RFC") or "").strip()
+    if not rfc:
+        return {"line": line_num, "col": "RFC", "msg": "Requerido"}
+    if len(rfc) > 30:
+        return {"line": line_num, "col": "RFC", "msg": "Máximo 30 caracteres"}
+
+    tipo_raw = (row.get("TIPO") or "").strip()
+    if tipo_raw and _parse_client_or_provider(tipo_raw) is None:
+        return {
+            "line": line_num,
+            "col": "TIPO",
+            "msg": "Valor no válido. Use Cliente, Proveedor o Ambos.",
+        }
+
+    name = (row.get("NOMBRE") or "").strip()
+    if len(name) > 256:
+        return {"line": line_num, "col": "NOMBRE", "msg": "Máximo 256 caracteres"}
+
+    short_name = (row.get("SHORT_NAME") or "").strip()
+    if short_name and len(short_name) > 10:
+        return {"line": line_num, "col": "SHORT_NAME", "msg": "Máximo 10 caracteres"}
+
+    curp = (row.get("CURP") or "").strip()
+    if curp and len(curp) > 19:
+        return {"line": line_num, "col": "CURP", "msg": "Máximo 19 caracteres"}
+
+    return None
+
+
+@celery_app.task(bind=True)
+def scan_file(self, job_id: str, config: str = None):
+    """
+    Fase 1: Leer CSV, validar filas, escribir errores en JSONL.
+    Devuelve waiting_confirmation con total_rows, error_count, valid_rows, errors.
+    """
+    logger.info(f"CP import: starting scan for job {job_id}")
+
+    file_path = _ensure_worker_has_file_from_redis(job_id)
+    if not file_path:
+        return {"status": "failed", "error": "Archivo no encontrado (expirado o no subido). Sube de nuevo."}
+    _ensure_worker_has_meta_from_redis(job_id, file_path)
+
+    error_dir = os.path.join(os.path.dirname(file_path).replace("temp", "errors"), "")
+    os.makedirs(error_dir, exist_ok=True)
+    error_path = os.path.join(error_dir, f"cp_{job_id}.jsonl")
+
+    total_rows = 0
+    try:
+        with open(file_path, "r", encoding="utf-8-sig") as f:
+            total_rows = sum(1 for _ in f) - 1
+    except Exception as e:
+        return {"status": "failed", "error": str(e)}
+
+    meta_path = file_path.replace(".csv", ".meta.json")
+    meta = {}
+    if os.path.exists(meta_path):
+        try:
+            with open(meta_path, "r", encoding="utf-8") as f:
+                meta = json.load(f) or {}
+        except Exception as e:
+            logger.warning(f"CP import: failed to read meta: {e}")
+
+    tenant_id = meta.get("tenant_id")
+    company_id = meta.get("company_id")
+    if not tenant_id or not company_id:
+        return {"status": "failed", "error": "Falta contexto (tenant/company)"}
+
+    error_count = 0
+    processed_rows = 0
+    errors_detail: List[Dict[str, Any]] = []
+
+    try:
+        with open(file_path, "r", encoding="utf-8-sig") as f_in, open(
+            error_path, "w", encoding="utf-8"
+        ) as f_err:
+            sample = f_in.read(2048)
+            f_in.seek(0)
+            try:
+                dialect = csv.Sniffer().sniff(sample, delimiters=",;\t")
+            except Exception:
+                dialect = "excel"
+            reader = csv.DictReader(f_in, dialect=dialect)
+
+            for i, row in enumerate(reader, start=1):
+                if i % 500 == 0:
+                    self.update_state(
+                        state="PROGRESS",
+                        meta={"current": i, "total": total_rows, "errors": error_count},
+                    )
+
+                row_norm = row_from_template(row, normalize_header)
+                err = _validate_row_client_provider(row_norm, i)
+                if err:
+                    error_count += 1
+                    f_err.write(json.dumps(err) + "\n")
+                    if len(errors_detail) < 500:
+                        errors_detail.append(
+                            {"line": err["line"], "col": err.get("col", ""), "msg": err.get("msg", "")}
+                        )
+                processed_rows += 1
+
+    except Exception as e:
+        logger.error(f"CP import scan failed: {e}")
+        return {"status": "failed", "error": str(e)}
+
+    error_lines_list = []
+    try:
+        if os.path.exists(error_path):
+            with open(error_path, "r", encoding="utf-8") as f:
+                for line in f:
+                    try:
+                        err = json.loads(line)
+                        if "line" in err:
+                            error_lines_list.append(err["line"])
+                    except Exception:
+                        pass
+        if error_lines_list:
+            r = _get_redis()
+            r.set(
+                f"{CP_IMPORT_ERROR_LINES_PREFIX}{job_id}",
+                json.dumps(error_lines_list).encode("utf-8"),
+                ex=CP_IMPORT_REDIS_TTL,
+            )
+    except Exception as e:
+        logger.warning(f"CP import: failed to store error lines in Redis: {e}")
+
+    return {
+        "status": "waiting_confirmation",
+        "job_id": job_id,
+        "total_rows": processed_rows,
+        "error_count": error_count,
+        "valid_rows": processed_rows - error_count,
+        "errors": errors_detail,
+    }
+
+
+def _str_or_none(val: Any, max_len: Optional[int] = None) -> Optional[str]:
+    if val is None:
+        return None
+    s = str(val).strip()
+    if not s:
+        return None
+    if max_len and len(s) > max_len:
+        return s[:max_len]
+    return s
+
+
+@celery_app.task(bind=True)
+def insert_valid_rows(self, job_id: str):
+    """
+    Fase 2: Re-leer CSV, omitir filas con error, insertar/actualizar ClientProvider.
+    Upsert por (tenant_id, company_id, rfc).
+    """
+    logger.info(f"CP import: starting commit for job {job_id}")
+
+    file_path = _ensure_worker_has_file_from_redis(job_id)
+    if not file_path:
+        alt_path = os.path.join(_worker_upload_dir(), f"cp_{job_id}.csv")
+        if not os.path.exists(alt_path):
+            return {
+                "status": "failed",
+                "error": "Archivo no encontrado (expirado). Sube y confirma de nuevo.",
+            }
+        file_path = alt_path
+    else:
+        _ensure_worker_has_meta_from_redis(job_id, file_path)
+
+    base_dir = os.path.dirname(file_path)
+    error_dir = base_dir.replace("temp", "errors")
+    error_path = os.path.join(error_dir, f"cp_{job_id}.jsonl")
+
+    error_lines = set()
+    try:
+        r = _get_redis()
+        raw = r.get(f"{CP_IMPORT_ERROR_LINES_PREFIX}{job_id}")
+        if raw:
+            error_lines = set(json.loads(raw.decode("utf-8")))
+    except Exception as e:
+        logger.debug(f"CP import: could not load error lines from Redis: {e}")
+    if not error_lines and os.path.exists(error_path):
+        with open(error_path, "r", encoding="utf-8") as f:
+            for line in f:
+                try:
+                    err = json.loads(line)
+                    error_lines.add(err["line"])
+                except Exception:
+                    pass
+
+    meta_path = file_path.replace(".csv", ".meta.json")
+    tenant_id = None
+    company_id = None
+    meta = {}
+    if os.path.exists(meta_path):
+        try:
+            with open(meta_path, "r", encoding="utf-8") as f:
+                meta = json.load(f) or {}
+                tenant_id = meta.get("tenant_id")
+                company_id = meta.get("company_id")
+        except Exception:
+            pass
+
+    if not tenant_id or not company_id:
+        return {"status": "failed", "error": "Falta contexto (tenant/company)"}
+
+    inserted_count = 0
+    skipped_invalid = 0
+    skipped_details: List[Dict[str, Any]] = []
+    response = None
+
+    try:
+        with CoreSessionLocal() as session:
+            # Cargar existentes por (tenant_id, company_id, rfc); rfc puede ser None en BD, usamos '' como key
+            existing_by_rfc: Dict[str, ClientProvider] = {}
+            for cp in (
+                session.query(ClientProvider)
+                .filter(
+                    ClientProvider.tenant_id == tenant_id,
+                    ClientProvider.company_id == company_id,
+                )
+                .all()
+            ):
+                key = (cp.rfc or "").strip()
+                existing_by_rfc[key] = cp
+
+            with open(file_path, "r", encoding="utf-8-sig") as f:
+                sample = f.read(2048)
+                f.seek(0)
+                try:
+                    dialect = csv.Sniffer().sniff(sample, delimiters=",;\t")
+                except Exception:
+                    dialect = "excel"
+                reader = csv.DictReader(f, dialect=dialect)
+
+                for i, row in enumerate(reader, start=1):
+                    if i in error_lines:
+                        continue
+
+                    row_norm = row_from_template(row, normalize_header)
+                    err = _validate_row_client_provider(row_norm, i)
+                    if err:
+                        skipped_invalid += 1
+                        skipped_details.append(
+                            {
+                                "line": i,
+                                "reason": f"{err.get('col', '')}: {err.get('msg', '')}",
+                            }
+                        )
+                        continue
+
+                    rfc = _str_or_none(row_norm.get("RFC"), 30)
+                    if not rfc:
+                        skipped_invalid += 1
+                        skipped_details.append({"line": i, "reason": "RFC requerido"})
+                        continue
+
+                    client_or_provider = _parse_client_or_provider(row_norm.get("TIPO"))
+                    if client_or_provider is None:
+                        client_or_provider = ClientOrProviderEnum.BOTH
+
+                    is_active = _parse_active(row_norm.get("ACTIVO"))
+
+                    existing = existing_by_rfc.get(rfc)
+                    if existing:
+                        existing.name = _str_or_none(row_norm.get("NOMBRE"), 256)
+                        existing.short_name = _str_or_none(row_norm.get("SHORT_NAME"), 10)
+                        existing.curp = _str_or_none(row_norm.get("CURP"), 19)
+                        existing.client_or_provider = client_or_provider
+                        existing.responsible = _str_or_none(row_norm.get("RESPONSABLE"), 80)
+                        existing.position = _str_or_none(row_norm.get("POSICION"), 30)
+                        existing.incoterm = _str_or_none(row_norm.get("INCOTERM"), 19)
+                        existing.is_active = is_active
+                        session.add(existing)
+                        inserted_count += 1
+                    else:
+                        new_cp = ClientProvider(
+                            tenant_id=tenant_id,
+                            company_id=company_id,
+                            rfc=rfc,
+                            name=_str_or_none(row_norm.get("NOMBRE"), 256),
+                            short_name=_str_or_none(row_norm.get("SHORT_NAME"), 10),
+                            curp=_str_or_none(row_norm.get("CURP"), 19),
+                            client_or_provider=client_or_provider,
+                            responsible=_str_or_none(row_norm.get("RESPONSABLE"), 80),
+                            position=_str_or_none(row_norm.get("POSICION"), 30),
+                            incoterm=_str_or_none(row_norm.get("INCOTERM"), 19),
+                            is_active=is_active,
+                        )
+                        session.add(new_cp)
+                        session.flush()
+                        existing_by_rfc[rfc] = new_cp
+                        inserted_count += 1
+
+                        # Opcional: crear dirección si hay email/teléfono/dirección
+                        email = _str_or_none(row_norm.get("EMAIL"), 100)
+                        phone = _str_or_none(row_norm.get("TELEFONO"), 30)
+                        address_str = _str_or_none(row_norm.get("DIRECCION"), 100)
+                        if email or phone or address_str:
+                            addr = ClientProviderAddress(
+                                client_id=new_cp.id,
+                                tenant_id=tenant_id,
+                                company_id=company_id,
+                                streets=address_str,
+                                postal_code=_str_or_none(row_norm.get("CODIGO POSTAL"), 15),
+                                city=_str_or_none(row_norm.get("CIUDAD"), 30),
+                                state=_str_or_none(row_norm.get("ESTADO"), 30),
+                                country=_str_or_none(row_norm.get("PAIS"), 3),
+                                phone=phone,
+                                email=email,
+                                contact=_str_or_none(row_norm.get("CONTACTO"), 50),
+                            )
+                            session.add(addr)
+
+            try:
+                session.commit()
+            except Exception as db_err:
+                session.rollback()
+                logger.error(f"CP import DB error: {db_err}")
+                return {"status": "failed", "error": str(db_err)}
+
+            total_skipped = skipped_invalid
+            if inserted_count == 0 and total_skipped > 0:
+                response = {
+                    "status": "warning",
+                    "inserted": 0,
+                    "skipped_invalid": skipped_invalid,
+                    "skipped_missing_fk": 0,
+                    "skipped_details": skipped_details,
+                    "message": f"No se insertaron registros. {total_skipped} rechazados.",
+                }
+            elif inserted_count == 0:
+                response = {
+                    "status": "failed",
+                    "error": "No hay registros válidos en el archivo CSV",
+                    "inserted": 0,
+                    "skipped_invalid": skipped_invalid,
+                    "skipped_missing_fk": 0,
+                    "skipped_details": skipped_details,
+                }
+            else:
+                response = {
+                    "status": "finished",
+                    "inserted": inserted_count,
+                    "skipped_invalid": skipped_invalid,
+                    "skipped_missing_fk": 0,
+                    "skipped_details": skipped_details,
+                }
+
+    except Exception as e:
+        logger.error(f"CP import task failed: {e}")
+        import traceback
+        logger.error(traceback.format_exc())
+        return {"status": "failed", "error": str(e)}
+
+    try:
+        if file_path and os.path.exists(file_path):
+            os.remove(file_path)
+        if os.path.exists(error_path):
+            os.remove(error_path)
+        meta_path = file_path.replace(".csv", ".meta.json")
+        if os.path.exists(meta_path):
+            os.remove(meta_path)
+        _delete_import_from_redis(job_id)
+    except Exception as cleanup_err:
+        logger.warning(f"CP import cleanup failed: {cleanup_err}")
+
+    if response is None:
+        response = {
+            "status": "failed",
+            "error": "Error inesperado",
+            "inserted": 0,
+            "skipped_invalid": skipped_invalid,
+            "skipped_missing_fk": 0,
+            "skipped_details": skipped_details,
+        }
+    return response
diff --git a/backend/api/v1/modules/a76/clients_and_providers/imports/template_config.py b/backend/api/v1/modules/a76/clients_and_providers/imports/template_config.py
new file mode 100644
index 00000000..f0968d9d
--- /dev/null
+++ b/backend/api/v1/modules/a76/clients_and_providers/imports/template_config.py
@@ -0,0 +1,56 @@
+"""
+Configuración de plantilla CSV para Clientes y Proveedores (EstructuraCatClienteProv.xls).
+Solo se leen columnas definidas aquí; el resto se ignora.
+Definir cabeceras según la primera fila del XLS oficial (frontend/static/csv/EstructuraCatClienteProv.xls).
+"""
+
+from typing import Dict, List, Any, Optional
+
+TEMPLATE_COLUMNS: Dict[str, List[Dict[str, Any]]] = {
+    "client_providers": [
+        {"canonical": "NOMBRE", "aliases": ["RAZON SOCIAL", "NAME", "RAZON SOCIAL O NOMBRE"]},
+        {"canonical": "RFC", "aliases": ["TAX_ID", "TAXID", "IDENTIFICADOR FISCAL", "IDENTIFICACION FISCAL"]},
+        {"canonical": "TIPO", "aliases": ["CLIENT_OR_PROVIDER", "TIPO ENTIDAD", "CLIENTE O PROVEEDOR"]},
+        {"canonical": "EMAIL", "aliases": ["CORREO", "E-MAIL", "CORREO ELECTRONICO"]},
+        {"canonical": "SHORT_NAME", "aliases": ["CLAVE", "CLAVE CORTA", "NOMBRE CORTO", "SIGLAS"]},
+        {"canonical": "CURP", "aliases": []},
+        {"canonical": "TELEFONO", "aliases": ["PHONE", "TEL", "TELEFONO CONTACTO"]},
+        {"canonical": "DIRECCION", "aliases": ["DOMICILIO", "DIRECCION FISCAL", "CALLE"]},
+        {"canonical": "CODIGO POSTAL", "aliases": ["CODIGOPOSTAL", "CP", "C.P."]},
+        {"canonical": "CIUDAD", "aliases": ["MUNICIPIO"]},
+        {"canonical": "ESTADO", "aliases": []},
+        {"canonical": "PAIS", "aliases": ["COUNTRY"]},
+        {"canonical": "CONTACTO", "aliases": ["CONTACT", "PERSONA CONTACTO"]},
+        {"canonical": "RESPONSABLE", "aliases": ["RESPONSABLE AREA"]},
+        {"canonical": "POSICION", "aliases": ["CARGO", "PUESTO"]},
+        {"canonical": "INCOTERM", "aliases": []},
+        {"canonical": "ACTIVO", "aliases": ["IS_ACTIVE", "ACTIVE", "ESTADO ACTIVO"]},
+    ],
+}
+
+
+def build_normalized_lookup(normalize_header_fn) -> Dict[str, str]:
+    """normalized_header -> canonical_name para plantilla client_providers."""
+    cols = TEMPLATE_COLUMNS.get("client_providers")
+    if not cols:
+        return {}
+    lookup: Dict[str, str] = {}
+    for item in cols:
+        canonical = item["canonical"]
+        lookup[normalize_header_fn(canonical)] = canonical
+        for alias in item.get("aliases") or []:
+            lookup[normalize_header_fn(alias)] = canonical
+    return lookup
+
+
+def row_from_template(row: Dict[str, Any], normalize_header_fn) -> Dict[str, Any]:
+    """Fila CSV con solo columnas de la plantilla, en nombres canónicos."""
+    lookup = build_normalized_lookup(normalize_header_fn)
+    if not lookup:
+        return {normalize_header_fn(k): v for k, v in row.items()}
+    out: Dict[str, Any] = {}
+    for csv_header, value in row.items():
+        key_norm = normalize_header_fn(csv_header)
+        if key_norm in lookup:
+            out[lookup[key_norm]] = value
+    return out
diff --git a/backend/api/v1/modules/a76/clients_and_providers/routes.py b/backend/api/v1/modules/a76/clients_and_providers/routes.py
index b1857de0..09f1cd48 100644
--- a/backend/api/v1/modules/a76/clients_and_providers/routes.py
+++ b/backend/api/v1/modules/a76/clients_and_providers/routes.py
@@ -20,10 +20,14 @@ from .dto import (
 )
 from .service import ClientProviderService
 from .models import ClientProvider
+from .imports.routes import router as imports_router
 
 # Create main router to add custom endpoints
 router = APIRouter(prefix="/clients-providers")
 
+# CSV import (mismo flujo que customs_brokers/imports: upload → scan → commit)
+router.include_router(imports_router, prefix="/imports", tags=["clients_and_providers / csv_import"])
+
 
 @router.get("/", response_model=ClientProviderPaginatedResponseDTO)
 async def get_clients_and_providers(
diff --git a/backend/api/v1/modules/a76/customs_brokers/imports/routes.py b/backend/api/v1/modules/a76/customs_brokers/imports/routes.py
index 336f1b73..99f7c609 100644
--- a/backend/api/v1/modules/a76/customs_brokers/imports/routes.py
+++ b/backend/api/v1/modules/a76/customs_brokers/imports/routes.py
@@ -121,6 +121,11 @@ async def get_import_status(job_id: str):
             return result
         return {"status": "finished", "result": result}
 
+    # A veces Celery tiene el result disponible pero state aún no es SUCCESS; si el result es éxito, devolverlo
+    result = getattr(task_result, "result", None)
+    if isinstance(result, dict) and result.get("status") in ("finished", "warning"):
+        return result
+
     logger.warning("CB import task %s failed: state=%s", job_id, task_result.state)
     err_msg = None
     tb = getattr(task_result, "traceback", None)
@@ -135,11 +140,10 @@ async def get_import_status(job_id: str):
                 err_msg = str(exc)
         except Exception:
             pass
-    if not err_msg:
-        result = getattr(task_result, "result", None)
-        if result is not None and not isinstance(result, dict):
+    if not err_msg and result is not None:
+        if not isinstance(result, dict):
             err_msg = str(result)
-        elif isinstance(result, dict) and (result.get("error") or result.get("message")):
+        elif result.get("error") or result.get("message"):
             err_msg = result.get("error") or result.get("message")
     return {"status": "failed", "error": err_msg or "Task failed"}
 
diff --git a/backend/core/celery_app.py b/backend/core/celery_app.py
index 6fc74f9f..6929e5f7 100644
--- a/backend/core/celery_app.py
+++ b/backend/core/celery_app.py
@@ -27,6 +27,7 @@ celery_app = Celery(
         "api.v1.modules.a76.reports.exportacion.descargo.task",
         "api.v1.modules.a76.imports.tasks",
         "api.v1.modules.a76.customs_brokers.imports.tasks",
+        "api.v1.modules.a76.clients_and_providers.imports.tasks",
         "api.v1.modules.a76.reports.exportacion.transmission.MAINX30.task",
         "api.v1.modules.a76.reports.importacion.transmission.temporal.MAINX30.task",
         "api.v1.modules.a76.reports.importacion.transmission.definitive.MAINX30.task"
diff --git a/frontend/src/lib/api.ts b/frontend/src/lib/api.ts
index fd26e485..61ff1f21 100644
--- a/frontend/src/lib/api.ts
+++ b/frontend/src/lib/api.ts
@@ -389,6 +389,21 @@ export const api = {
 			api.post(`/v1/a76/customs-brokers/imports/${jobId}/commit`, {})
 	},
 
+	// CSV import for Clientes y Proveedores (flujo propio en clients_and_providers/imports)
+	clientProviderImports: {
+		upload: (file: File, companyId: number) => {
+			const formData = new FormData();
+			formData.append('file', file);
+			return fetchApi(
+				`/v1/a76/clients-providers/imports/upload?company_id=${companyId}`,
+				{ method: 'POST', body: formData }
+			);
+		},
+		status: (jobId: string) => api.get(`/v1/a76/clients-providers/imports/${jobId}/status`),
+		commit: (jobId: string) =>
+			api.post(`/v1/a76/clients-providers/imports/${jobId}/commit`, {})
+	},
+
 	// Generic request for custom needs (like file uploads)
 	request: <T = any>(endpoint: string, options: RequestInit = {}) => fetchApi<T>(endpoint, options)
 };
diff --git a/frontend/src/routes/dashboard/csv-upload/+page.svelte b/frontend/src/routes/dashboard/csv-upload/+page.svelte
index 3aa127a0..982b080e 100644
--- a/frontend/src/routes/dashboard/csv-upload/+page.svelte
+++ b/frontend/src/routes/dashboard/csv-upload/+page.svelte
@@ -26,6 +26,8 @@
 	let showResultModal = $state(false);
 	// Cuando es true, usamos API de importación de Agentes Aduanales (customs_brokers/imports)
 	let useCustomsBrokerImport = $state(false);
+	// Cuando es true, usamos API de importación de Clientes y Proveedores (clients_and_providers/imports)
+	let useClientProviderImport = $state(false);
 
 	// Initialize settings for all tabs upfront to avoid reactivity loops
 	let allSettings = $state<Record<string, any>>(() => {
@@ -45,6 +47,7 @@
 		activeModelTarget = config.modelTarget || null;
 		scanResults = null;
 		useCustomsBrokerImport = config.id === 'customs_brokers';
+		useClientProviderImport = config.id === 'clients_providers';
 
 		const companyId = companyStore.activeCompany?.id || 1;
 
@@ -66,6 +69,24 @@
 			return;
 		}
 
+		if (useClientProviderImport) {
+			try {
+				const res = await api.clientProviderImports.upload(file, companyId);
+				if (res.data?.job_id) {
+					currentJobId = res.data.job_id;
+					pollStatus();
+				} else {
+					toast.error(res.error || 'Error al subir el archivo');
+					isUploading = false;
+				}
+			} catch (e) {
+				console.error('Upload exception', e);
+				toast.error('Error inesperado al subir el archivo');
+				isUploading = false;
+			}
+			return;
+		}
+
 		const currentSettings = allSettings[activeTab] || {};
 		const footerConfig = { ...currentSettings };
 		if (activeTab === 'importacion') {
@@ -102,7 +123,9 @@
 		try {
 			const res = useCustomsBrokerImport
 				? await api.customsBrokerImports.status(currentJobId)
-				: await api.imports.status(currentJobId);
+				: useClientProviderImport
+					? await api.clientProviderImports.status(currentJobId)
+					: await api.imports.status(currentJobId);
 			console.log('Poll response', res);
 			if (res.error && !res.data) {
 				toast.error(res.error || 'Error al consultar el estado');
@@ -116,7 +139,14 @@
 				toast.success('Escaneo completado. Revisa los resultados.');
 				isUploading = false;
 			} else if (res.data?.status === 'failed' || res.data?.status === 'FAILURE') {
-				toast.error('Error en el procesamiento: ' + (res.data.error || 'Error desconocido'));
+				const errRaw = res.data.error;
+				const errText =
+					typeof errRaw === 'string'
+						? errRaw.includes('finished') && errRaw.includes('inserted')
+							? 'La importación pudo completarse. Revisa el listado de registros.'
+							: errRaw
+						: (errRaw?.message ?? 'Error desconocido');
+				toast.error('Error en el procesamiento: ' + errText);
 				isUploading = false;
 				currentJobId = null;
 				scanResults = null;
@@ -239,7 +269,9 @@
 				isUploading = true;
 				const res = useCustomsBrokerImport
 					? await api.customsBrokerImports.commit(currentJobId)
-					: await api.imports.commit(currentJobId, activeModelTarget || '');
+					: useClientProviderImport
+						? await api.clientProviderImports.commit(currentJobId)
+						: await api.imports.commit(currentJobId, activeModelTarget || '');
 				if (res.data?.commit_job_id) {
 					currentJobId = res.data.commit_job_id;
 					pollStatus();