diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
new file mode 100644
index 000000000..8a7f725b0
--- /dev/null
+++ b/.github/workflows/ci.yml
@@ -0,0 +1,42 @@
+name: CI
+
+on:
+  pull_request:
+  push:
+    branches: [ master ]
+
+concurrency:
+  group: ci-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  lint:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: '3.12'
+          cache: 'pip'
+      - run: pip install ruff==0.6.9
+      - run: ruff check .
+
+  tests:
+    strategy:
+      matrix:
+        python-version: ['3.9','3.10','3.11','3.12']
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+          cache: 'pip'
+      - run: pip install -e .[dev]
+      - run: pytest -q -k "not online"
+
+
diff --git a/.github/workflows/docker-build.yml b/.github/workflows/docker-build.yml
new file mode 100644
index 000000000..fe8a478ed
--- /dev/null
+++ b/.github/workflows/docker-build.yml
@@ -0,0 +1,16 @@
+name: Docker Build
+
+on:
+  push:
+    branches: [ master, rc ]
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+    steps:
+      - uses: actions/checkout@v4
+      - run: docker build -t sherlock:test .
+
+
diff --git a/.github/workflows/nightly-online.yml b/.github/workflows/nightly-online.yml
new file mode 100644
index 000000000..b5f8ef2f4
--- /dev/null
+++ b/.github/workflows/nightly-online.yml
@@ -0,0 +1,22 @@
+name: Nightly Online Tests
+
+on:
+  schedule:
+    - cron: '0 3 * * *'
+  workflow_dispatch:
+
+jobs:
+  online:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: '3.12'
+          cache: 'pip'
+      - run: pip install -e .[dev]
+      - run: pytest -q -m online --maxfail=1
+
+
diff --git a/docs/README.md b/docs/README.md
index af9011092..bc6e3413d 100644
--- a/docs/README.md
+++ b/docs/README.md
@@ -48,6 +48,15 @@ sherlock user1 user2 user3
 ```
 
 Accounts found will be stored in an individual text file with the corresponding username (e.g ```user123.txt```).
+Use `--no-txt` to skip creating the `.txt` file.
+
+### Output options
+
+- **Disable txt output**: `--no-txt`
+- **Write to a specific file (single user)**: `--output FILE`
+- **Write per-user files into a folder (multiple users)**: `--folderoutput DIR`
+- **Export CSV**: `--csv` (creates `username.csv`)
+- **Export Excel**: `--xlsx` (creates `username.xlsx`)
 
 ```console
 $ sherlock --help
@@ -55,7 +64,7 @@ usage: sherlock [-h] [--version] [--verbose] [--folderoutput FOLDEROUTPUT]
                 [--output OUTPUT] [--tor] [--unique-tor] [--csv] [--xlsx]
                 [--site SITE_NAME] [--proxy PROXY_URL] [--json JSON_FILE]
                 [--timeout TIMEOUT] [--print-all] [--print-found] [--no-color]
-                [--browse] [--local] [--nsfw]
+                [--browse] [--local] [--nsfw] [--no-txt] [--ignore-exclusions]
                 USERNAMES [USERNAMES ...]
 
 Sherlock: Find Usernames Across Social Networks (Version 0.14.3)
@@ -96,6 +105,8 @@ optional arguments:
   --browse, -b          Browse to all results on default browser.
   --local, -l           Force the use of the local data.json file.
   --nsfw                Include checking of NSFW sites from default list.
+  --no-txt             Disable creation of a txt file
+  --ignore-exclusions  Ignore upstream exclusions (may return more false positives)
 ```
 ## Apify Actor Usage [![Sherlock Actor](https://apify.com/actor-badge?actor=netmilk/sherlock)](https://apify.com/netmilk/sherlock?fpr=sherlock)
 
diff --git a/sherlock_project/sherlock.py b/sherlock_project/sherlock.py
index 250175a57..c86bbbbb0 100644
--- a/sherlock_project/sherlock.py
+++ b/sherlock_project/sherlock.py
@@ -18,7 +18,6 @@
 
 import csv
 import signal
-import pandas as pd
 import os
 import re
 from argparse import ArgumentParser, RawDescriptionHelpFormatter
@@ -27,7 +26,9 @@
 from typing import Optional
 
 import requests
+from requests.adapters import HTTPAdapter
 from requests_futures.sessions import FuturesSession
+from urllib3.util.retry import Retry
 
 from sherlock_project.__init__ import (
     __longname__,
@@ -110,6 +111,35 @@ def response_time(resp, *args, **kwargs):
         )
 
 
+def _mount_session_with_retries(session: requests.Session) -> requests.Session:
+    """Configure retries, backoff, and connection pooling on a requests session."""
+    retry = Retry(
+        total=3,
+        backoff_factor=0.5,
+        status_forcelist=[429, 500, 502, 503, 504],
+        allowed_methods=["HEAD", "GET", "OPTIONS", "POST", "PUT"],
+        raise_on_status=False,
+    )
+    adapter = HTTPAdapter(pool_connections=100, pool_maxsize=100, max_retries=retry)
+    session.mount("http://", adapter)
+    session.mount("https://", adapter)
+    return session
+
+
+def _is_valid_proxy_url(proxy_url: str) -> bool:
+    """Basic validation for proxy URLs.
+
+    Accept common schemes and require a netloc.
+    """
+    from urllib.parse import urlparse
+
+    parsed = urlparse(proxy_url)
+    if parsed.scheme not in {"http", "https", "socks5", "socks5h", "socks4"}:
+        return False
+    if not parsed.netloc:
+        return False
+    return True
+
 def get_response(request_future, error_type, social_network):
     # Default for Response object if some failure occurs.
     response = None
@@ -237,6 +267,9 @@ def sherlock(
         underlying_session = requests.session()
         underlying_request = requests.Request()
 
+    # Mount retries/connection pooling for robustness and performance
+    underlying_session = _mount_session_with_retries(underlying_session)
+
     # Limit number of workers to 20.
     # This is probably vastly overkill.
     if len(site_data) >= 20:
@@ -756,12 +789,15 @@ def main():
         print(f"A problem occurred while checking for an update: {error}")
 
     # Argument check
-    # TODO regex check on args.proxy
     if args.tor and (args.proxy is not None):
         raise Exception("Tor and Proxy cannot be set at the same time.")
 
     # Make prompts
     if args.proxy is not None:
+        if not _is_valid_proxy_url(args.proxy):
+            raise ArgumentTypeError(
+                f"Invalid proxy URL: {args.proxy}. Expected scheme://host:port with scheme in http, https, socks4, socks5, socks5h."
+            )
         print("Using the proxy: " + args.proxy)
 
     if args.tor or args.unique_tor:
@@ -942,6 +978,8 @@ def main():
                         ]
                     )
         if args.xlsx:
+            # Lazy import to reduce startup time and optional dependency overhead
+            import pandas as pd
             usernames = []
             names = []
             url_main = []
diff --git a/tests/test_ux.py b/tests/test_ux.py
index 3c62463b5..600afa2af 100644
--- a/tests/test_ux.py
+++ b/tests/test_ux.py
@@ -33,6 +33,25 @@ def test_wildcard_username_expansion():
     assert sherlock.multiple_usernames('test{?}test') == ["test_test" , "test-test" , "test.test"]
 
 
+def test_no_txt_flag(tmp_path):
+    """Ensure that --no-txt prevents creating the txt output file."""
+    username = "totallyfakeuserforunittest"
+    # Run in a temp working directory to avoid polluting local repo
+    import os
+    cwd = os.getcwd()
+    try:
+        os.chdir(tmp_path)
+        # Run with --no-txt; expect no username.txt file created
+        try:
+            Interactives.run_cli(f"--no-txt {username}")
+        except Exception:
+            # Ignore network errors; only assert file is not created
+            pass
+        assert not os.path.exists(f"{username}.txt")
+    finally:
+        os.chdir(cwd)
+
+
 @pytest.mark.parametrize('cliargs', [
     '',
     '--site urghrtuight --egiotr',