### Encrypt Payload with AES and RSA using Signer

Source: https://context7.com/xkiian/geekedtest/llms.txt

Encrypts a payload using AES-CBC and RSA. It first generates a random AES key, encrypts the payload with it, and then encrypts the AES key using RSA. This is used for secure payload transmission.

```python
from geeked.sign import Signer

# Encrypt payload with AES-CBC + RSA
random_key = Signer.rand_uid()  # Generate 16-char random key
encrypted_data = Signer.encrypt_symmetrical_1('{"userresponse": 156}', random_key)
encrypted_key = Signer.encrypt_asymmetric_1(random_key)

# Full w parameter generation (called internally by Geeked.submit_captcha)
w_param = Signer.encrypt_w('{"payload": "data"}', pt="1")
```

--------------------------------

### Generate Proof-of-Work Hash with Signer

Source: https://context7.com/xkiian/geekedtest/llms.txt

Generates a proof-of-work hash required by Geetest's anti-bot system. It takes parameters like lot number, captcha ID, hash function, and difficulty bits. The output includes a message and a signature hash.

```python
from geeked.sign import Signer

# Generate proof-of-work hash (required by Geetest anti-bot)
pow_result = Signer.generate_pow(
    lot_number_pow="abc123",
    captcha_id_pow="54088bb07d2df3c46b79f80300b0abbe",
    hash_func="md5",      # Can be: md5, sha1, sha256
    hash_version="1",
    bits=16,              # Difficulty level
    date="1704067200000",
    empty=""
)

print(pow_result)
# Output:
# {
#     'pow_msg': '1|16|md5|1704067200000|54088bb07d2df3c46b79f80300b0abbe|abc123||randomhex',
#     'pow_sign': 'hash_starting_with_0000'
# }
```

--------------------------------

### Solve Icon Matching Captcha with IconSolver (Python)

Source: https://context7.com/xkiian/geekedtest/llms.txt

The `IconSolver` class utilizes ddddocr for AI-based OCR to solve icon matching captchas. It takes the main captcha image URL and a list of question icon URLs, returning the coordinates to click each icon in the correct sequence.

```python
from geeked.icon import IconSolver

# Icon captcha requires matching arrows in the correct order
# imgs: URL path to the main captcha image
# ques: List of question icon URLs to match

solver = IconSolver(
    imgs="captcha_v4/policy/xxx/icon/xxx/image.jpg",
    ques=[
        "nerualpic/original_icon_pic/icon_20201215/315ce8665e781dabcd1eb09d3e604803.png",  # left arrow
        "nerualpic/original_icon_pic/icon_20201215/38bd9dda695098c7dfad74c921923a7d.png",  # left-up arrow
        "nerualpic/original_icon_pic/icon_20201215/cb0eaa639b2117a69a81af3d8c1496a1.png"   # down arrow
    ]
)

positions = solver.find_icon_position()

print(positions)
# Output: [[x1, y1], [x2, y2], [x3, y3]]
# Coordinates for clicking each icon in order

# Supported icon directions:
# 'u' (up), 'd' (down), 'l' (left), 'r' (right)
```

--------------------------------

### Detect and Classify Icons with DdddService OCR

Source: https://context7.com/xkiian/geekedtest/llms.txt

Utilizes the DdddService class, which wraps ddddocr, for icon detection and classification using a custom ONNX model for Geetest icon recognition. It can detect bounding boxes of icons and classify cropped icon images.

```python
from geeked.dddd_server import DdddService
import requests

# Initialize the OCR service (loads ONNX model)
dddd = DdddService()

# Load an image
img_bytes = requests.get("https://example.com/captcha.jpg").content

# Detect bounding boxes of icons in image
bboxes = dddd.detection(img_bytes)
print(bboxes)
# Output: [[x1, y1, x2, y2], [x1, y1, x2, y2], ...]

# Classify a cropped icon image
icon_crop = img_bytes  # Cropped icon region
direction = dddd.classification(icon_crop)
print(direction)
# Output: "arrow_lu" (left-up arrow)
```

--------------------------------

### Solve Gobang (Five-in-a-Row) Captcha with GobangSolver (Python)

Source: https://context7.com/xkiian/geekedtest/llms.txt

The `GobangSolver` class solves the five-in-a-row game captcha by analyzing a 2D board representation. It identifies the optimal move to complete a line of four pieces by finding an empty space adjacent to a line of three.

```python
from geeked.gobang import GobangSolver

# Board is represented as a 2D list where:
# - 0 represents empty space
# - Other numbers represent different colored pieces
board = [
    [1, 2, 1, 2, 1],
    [2, 1, 0, 1, 2],  # Row 1 has a gap (0) that could complete a line
    [1, 2, 1, 2, 1],
    [2, 1, 2, 1, 2],
    [1, 1, 1, 1, 2]   # Row 4 has four 1s - needs one more
]

solver = GobangSolver(board)
result = solver.find_four_in_line()

print(result)
# Output: [[source_row, source_col], [target_row, target_col]]
# Example: [[4, 4], [1, 2]] - Move piece from (4,4) to (1,2)

# The solver:
# 1. Iterates through all rows, columns, and diagonals
# 2. Finds lines with n-1 matching pieces and one empty space
# 3. Identifies which piece to move to complete the line
```

--------------------------------

### Update Solver Constants with deobfuscate.py

Source: https://context7.com/xkiian/geekedtest/llms.txt

A utility script to extract updated constants from Geetest's JavaScript. This is crucial when solver functionality breaks due to Geetest version changes. The script outputs updated values for mapping, abo dictionary, and device ID.

```python
# Run the deobfuscation script to get updated constants
# python deobfuscate.py

# Output:
# [~] Version: v0.4.8
# [+] abo: {"fwjn":"DJkH"}
# [+] mappings (n[11:14])+.+(n[12:14]+n[6:8])'n[21:28]
# [+] device_id: "" (probably empty)

# Update these values in sign.py:
# - self.mapping in LotParser class
# - abo dictionary in Signer.generate_w()
# - device_id in the em payload
```

--------------------------------

### Solve Slide Puzzle Captcha with SlideSolver (Python)

Source: https://context7.com/xkiian/geekedtest/llms.txt

The `SlideSolver` class uses OpenCV for image processing, specifically edge detection and template matching, to determine the correct position for sliding puzzle pieces. It takes the puzzle piece and background images as input.

```python
from geeked.slide import SlideSolver
import requests

# Load puzzle images from Geetest CDN
puzzle_piece = requests.get(
    "https://static.geetest.com/captcha_v4/xxx/slice/image.png"
).content
background = requests.get(
    "https://static.geetest.com/captcha_v4/xxx/bg/image.png"
).content

# Create solver and find puzzle position
solver = SlideSolver(puzzle_piece, background)
position = solver.find_puzzle_piece_position()

print(f"Slide to position: {position}")  # Output: Slide to position: 156

# The solver uses Canny edge detection and template matching:
# 1. Applies edge detection to both images
# 2. Uses cv2.matchTemplate with TM_CCOEFF_NORMED
# 3. Returns the x-coordinate where the piece should be placed
```

--------------------------------

### Solve Geetest v4 CAPTCHA using Python

Source: https://github.com/xkiian/geekedtest/blob/main/readme.md

This Python code snippet demonstrates how to use the 'Geeked' class to solve a Geetest v4 CAPTCHA. It requires the 'captcha_id' and 'risk_type' to be provided. The output includes CAPTCHA solution details such as 'lot_number', 'pass_token', and 'captcha_output'.

```python
from geeked import Geeked

# Replace with your own captcha_id
captcha_id = "54088bb07d2df3c46b79f80300b0abbe"

# Instantiate the solver with captcha_id and risk_type
geeked = Geeked(captcha_id, risk_type="slide")

# Solve the captcha
sec_code = geeked.solve()

# Output the solution
print(sec_code)
# Output will be in the following format:
# {
#    'captcha_id': '...',
#    'lot_number': '...',
#    'pass_token': '...',
#    'gen_time': '...',
#    'captcha_output': '...'
# }
```

--------------------------------

### Solve Geetest v4 Captcha with Geeked Class (Python)

Source: https://context7.com/xkiian/geekedtest/llms.txt

The `Geeked` class is the primary interface for solving various Geetest v4 captcha types. It manages the HTTP session, loads challenges, and orchestrates the solving process. Supports basic usage and proxy configurations.

```python
from geeked import Geeked

# Basic usage - solve a slide captcha
captcha_id = "54088bb07d2df3c46b79f80300b0abbe"
geeked = Geeked(captcha_id, risk_type="slide")
sec_code = geeked.solve()

print(sec_code)
# Output:
# {
#     'captcha_id': '54088bb07d2df3c46b79f80300b0abbe',
#     'lot_number': 'a1b2c3d4e5f6g7h8i9j0',
#     'pass_token': 'token_string_here',
#     'gen_time': '1704067200',
#     'captcha_output': 'encrypted_output_string'
# }

# With proxy support
geeked = Geeked(
    captcha_id="54088bb07d2df3c46b79f80300b0abbe",
    risk_type="slide",
    proxy="http://127.0.0.1:8050",
    verify=False  # Disable SSL verification if needed
)
sec_code = geeked.solve()

# Supported risk_type values:
# - "slide": Slide puzzle captcha
# - "gobang" or "winlinze": Five-in-a-row game captcha
# - "icon": Icon matching captcha
# - "ai" or "invisible": Invisible captcha (no user interaction)
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.