mal-toolbox 0.0.40__tar.gz → 0.1.1__tar.gz

{mal_toolbox-0.0.40/mal_toolbox.egg-info → mal_toolbox-0.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mal-toolbox
-Version: 0.0.40
+Version: 0.1.1
 Summary: A collection of tools used to create MAL models and attack graphs.
 Author-email: Andrei Buhaiu <buhaiu@kth.se>, Giuseppe Nebbione <nebbione@kth.se>, Nikolaos Kakouros <nkak@kth.se>, Jakob Nyberg <jaknyb@kth.se>, Joakim Loxdal <loxdal@kth.se>
 License: Apache Software License
@@ -25,16 +25,19 @@ Requires-Dist: antlr4-python3-runtime
 Requires-Dist: docopt
 Requires-Dist: PyYAML
 
-# Overview
+# MAL Toolbox overview
 
-A collection of python modules to help developers create and work with MAL
-models and attack graphs.
+MAL Toolbox is a collection of python modules to help developers create and work with
+MAL ([Meta Attack Language](https://mal-lang.org/)) models and attack graphs.
 
-# The Language Module
+Attack graphs can be used to run simulations (see MAL Simulator) or analysis.
+MAL Toolbox also gives the ability to view the AttackGraph/Model graphically in neo4j.
+
+## The Language Module
 
 The language module provides various tools to process MAL languages.
 
-## The Language Specification Submodule
+### The Language Specification Submodule
 
 The language specification submodule provides functions to load the
 specification from a .mar archive(`load_language_specification_from_mar`) or a
@@ -43,7 +46,7 @@ then be used to generate python classes representing the assets and
 associations of the language and to determine the attack steps for each asset
 when generating the attack graph.
 
-## The Language Classes Factory Submodule
+### The Language Classes Factory Submodule
 
 The language classes factory submodule is used to generate python classes
 using the `python_jsonschema_objects` package from a language specification.
@@ -54,19 +57,23 @@ using JSON Schema validators they will enforce their restrictions when using
 the python objects created. These classes are typically used in conjunction
 with model module to create instance models.
 
-# The Model Module
+## The Model Module
+
+With a MAL language a Model (a MAL instance model) can be created either
+from a model file or empty.
+
+The model class will store all of the relevant information to the MAL
+instance model, most importantly the assets and associations that make it up.
+
+Assets and associations are objects of classes created using the language
+classes factory submodule in runtime. It also allows for `Attacker` objects
+to be created and associated with attack steps on assets in the model.
+The most relevant methods of the Model are the ones used to add different
+elements to the model, `add_asset`, `add_association`, and `add_attacker`.
 
-The model module is used to create MAL instance models. The model class will
-store all of the relevant information to the MAL instance model, most
-importantly the assets and associations that make it up. These assets and
-associations should be objects created using the language classes factory
-submodule. It also allows for `Attacker` objects to be created and associated
-with attack steps on assets in the model. The most relevant functions here are
-the ones used to add different elements to the model, `add_asset`,
-`add_association`, and `add_attacker`. Model objects can be used to generate
-attack graphs using the attack graph module.
+Model objects can be used to generate attack graphs with the AttackGraph module.
 
-# The Attack Graph Module
+## The Attack Graph Module
 
 The attack graph module contains tools used to generate attack graphs from
 existing MAL instance models and analyse MAL attack graphs. The function used
@@ -84,14 +91,21 @@ resulting attack graph with the instance model given as a parameter in order
 to create attack step nodes that represent the entry points of the attackers
 and attach them to the attack steps specified in the instance model.
 
-# Ingestors Module
+## Ingestors Module
 
 The ingestors module contains various tools that can make use of the instance
 model or attack graph. Currently the Neo4J ingestor is the only one available
 and it can be used to visualise the instance model and the attack graph.
 
-# Command Line Client
 
+# Usage
+
+## Configuration
+A default configuration file `default.conf` can be found in the package
+directory. This contains the default values to use for logging and can also be
+used to store the information needed to access the local Neo4J instance.
+
+## Command Line Client
 In addition to the modules that make up the MAL-Toolbox package it also
 provides a simple command line client that can be used to easily generate
 attack graphs from a .mar language specification file and a JSON instance
@@ -103,10 +117,10 @@ The usage is: `maltoolbox gen_ag [--neo4j] <model_json_file>
 If the `--neo4j` flag is specified the model and attack graph will be loaded
 into a local Neo4J instance.
 
-# Configuration
-A default configuration file `default.conf` can be found in the package
-directory. This contains the default values to use for logging and can also be
-used to store the information needed to access the local Neo4J instance.
+## Code examples / Tutorial
+
+To find code examples and tutorials, visit the
+[MAL Toolbox Tutorial](https://github.com/mal-lang/mal-toolbox-tutorial/tree/main) repository.
 
 # Tests
 There are unit tests inside of ./tests.

{mal_toolbox-0.0.40 → mal_toolbox-0.1.1}/README.md

@@ -1,13 +1,16 @@
-# Overview
+# MAL Toolbox overview
 
-A collection of python modules to help developers create and work with MAL
-models and attack graphs.
+MAL Toolbox is a collection of python modules to help developers create and work with
+MAL ([Meta Attack Language](https://mal-lang.org/)) models and attack graphs.
 
-# The Language Module
+Attack graphs can be used to run simulations (see MAL Simulator) or analysis.
+MAL Toolbox also gives the ability to view the AttackGraph/Model graphically in neo4j.
+
+## The Language Module
 
 The language module provides various tools to process MAL languages.
 
-## The Language Specification Submodule
+### The Language Specification Submodule
 
 The language specification submodule provides functions to load the
 specification from a .mar archive(`load_language_specification_from_mar`) or a
@@ -16,7 +19,7 @@ then be used to generate python classes representing the assets and
 associations of the language and to determine the attack steps for each asset
 when generating the attack graph.
 
-## The Language Classes Factory Submodule
+### The Language Classes Factory Submodule
 
 The language classes factory submodule is used to generate python classes
 using the `python_jsonschema_objects` package from a language specification.
@@ -27,19 +30,23 @@ using JSON Schema validators they will enforce their restrictions when using
 the python objects created. These classes are typically used in conjunction
 with model module to create instance models.
 
-# The Model Module
+## The Model Module
+
+With a MAL language a Model (a MAL instance model) can be created either
+from a model file or empty.
+
+The model class will store all of the relevant information to the MAL
+instance model, most importantly the assets and associations that make it up.
+
+Assets and associations are objects of classes created using the language
+classes factory submodule in runtime. It also allows for `Attacker` objects
+to be created and associated with attack steps on assets in the model.
+The most relevant methods of the Model are the ones used to add different
+elements to the model, `add_asset`, `add_association`, and `add_attacker`.
 
-The model module is used to create MAL instance models. The model class will
-store all of the relevant information to the MAL instance model, most
-importantly the assets and associations that make it up. These assets and
-associations should be objects created using the language classes factory
-submodule. It also allows for `Attacker` objects to be created and associated
-with attack steps on assets in the model. The most relevant functions here are
-the ones used to add different elements to the model, `add_asset`,
-`add_association`, and `add_attacker`. Model objects can be used to generate
-attack graphs using the attack graph module.
+Model objects can be used to generate attack graphs with the AttackGraph module.
 
-# The Attack Graph Module
+## The Attack Graph Module
 
 The attack graph module contains tools used to generate attack graphs from
 existing MAL instance models and analyse MAL attack graphs. The function used
@@ -57,14 +64,21 @@ resulting attack graph with the instance model given as a parameter in order
 to create attack step nodes that represent the entry points of the attackers
 and attach them to the attack steps specified in the instance model.
 
-# Ingestors Module
+## Ingestors Module
 
 The ingestors module contains various tools that can make use of the instance
 model or attack graph. Currently the Neo4J ingestor is the only one available
 and it can be used to visualise the instance model and the attack graph.
 
-# Command Line Client
 
+# Usage
+
+## Configuration
+A default configuration file `default.conf` can be found in the package
+directory. This contains the default values to use for logging and can also be
+used to store the information needed to access the local Neo4J instance.
+
+## Command Line Client
 In addition to the modules that make up the MAL-Toolbox package it also
 provides a simple command line client that can be used to easily generate
 attack graphs from a .mar language specification file and a JSON instance
@@ -76,10 +90,10 @@ The usage is: `maltoolbox gen_ag [--neo4j] <model_json_file>
 If the `--neo4j` flag is specified the model and attack graph will be loaded
 into a local Neo4J instance.
 
-# Configuration
-A default configuration file `default.conf` can be found in the package
-directory. This contains the default values to use for logging and can also be
-used to store the information needed to access the local Neo4J instance.
+## Code examples / Tutorial
+
+To find code examples and tutorials, visit the
+[MAL Toolbox Tutorial](https://github.com/mal-lang/mal-toolbox-tutorial/tree/main) repository.
 
 # Tests
 There are unit tests inside of ./tests.

{mal_toolbox-0.0.40 → mal_toolbox-0.1.1/mal_toolbox.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mal-toolbox
-Version: 0.0.40
+Version: 0.1.1
 Summary: A collection of tools used to create MAL models and attack graphs.
 Author-email: Andrei Buhaiu <buhaiu@kth.se>, Giuseppe Nebbione <nebbione@kth.se>, Nikolaos Kakouros <nkak@kth.se>, Jakob Nyberg <jaknyb@kth.se>, Joakim Loxdal <loxdal@kth.se>
 License: Apache Software License
@@ -25,16 +25,19 @@ Requires-Dist: antlr4-python3-runtime
 Requires-Dist: docopt
 Requires-Dist: PyYAML
 
-# Overview
+# MAL Toolbox overview
 
-A collection of python modules to help developers create and work with MAL
-models and attack graphs.
+MAL Toolbox is a collection of python modules to help developers create and work with
+MAL ([Meta Attack Language](https://mal-lang.org/)) models and attack graphs.
 
-# The Language Module
+Attack graphs can be used to run simulations (see MAL Simulator) or analysis.
+MAL Toolbox also gives the ability to view the AttackGraph/Model graphically in neo4j.
+
+## The Language Module
 
 The language module provides various tools to process MAL languages.
 
-## The Language Specification Submodule
+### The Language Specification Submodule
 
 The language specification submodule provides functions to load the
 specification from a .mar archive(`load_language_specification_from_mar`) or a
@@ -43,7 +46,7 @@ then be used to generate python classes representing the assets and
 associations of the language and to determine the attack steps for each asset
 when generating the attack graph.
 
-## The Language Classes Factory Submodule
+### The Language Classes Factory Submodule
 
 The language classes factory submodule is used to generate python classes
 using the `python_jsonschema_objects` package from a language specification.
@@ -54,19 +57,23 @@ using JSON Schema validators they will enforce their restrictions when using
 the python objects created. These classes are typically used in conjunction
 with model module to create instance models.
 
-# The Model Module
+## The Model Module
+
+With a MAL language a Model (a MAL instance model) can be created either
+from a model file or empty.
+
+The model class will store all of the relevant information to the MAL
+instance model, most importantly the assets and associations that make it up.
+
+Assets and associations are objects of classes created using the language
+classes factory submodule in runtime. It also allows for `Attacker` objects
+to be created and associated with attack steps on assets in the model.
+The most relevant methods of the Model are the ones used to add different
+elements to the model, `add_asset`, `add_association`, and `add_attacker`.
 
-The model module is used to create MAL instance models. The model class will
-store all of the relevant information to the MAL instance model, most
-importantly the assets and associations that make it up. These assets and
-associations should be objects created using the language classes factory
-submodule. It also allows for `Attacker` objects to be created and associated
-with attack steps on assets in the model. The most relevant functions here are
-the ones used to add different elements to the model, `add_asset`,
-`add_association`, and `add_attacker`. Model objects can be used to generate
-attack graphs using the attack graph module.
+Model objects can be used to generate attack graphs with the AttackGraph module.
 
-# The Attack Graph Module
+## The Attack Graph Module
 
 The attack graph module contains tools used to generate attack graphs from
 existing MAL instance models and analyse MAL attack graphs. The function used
@@ -84,14 +91,21 @@ resulting attack graph with the instance model given as a parameter in order
 to create attack step nodes that represent the entry points of the attackers
 and attach them to the attack steps specified in the instance model.
 
-# Ingestors Module
+## Ingestors Module
 
 The ingestors module contains various tools that can make use of the instance
 model or attack graph. Currently the Neo4J ingestor is the only one available
 and it can be used to visualise the instance model and the attack graph.
 
-# Command Line Client
 
+# Usage
+
+## Configuration
+A default configuration file `default.conf` can be found in the package
+directory. This contains the default values to use for logging and can also be
+used to store the information needed to access the local Neo4J instance.
+
+## Command Line Client
 In addition to the modules that make up the MAL-Toolbox package it also
 provides a simple command line client that can be used to easily generate
 attack graphs from a .mar language specification file and a JSON instance
@@ -103,10 +117,10 @@ The usage is: `maltoolbox gen_ag [--neo4j] <model_json_file>
 If the `--neo4j` flag is specified the model and attack graph will be loaded
 into a local Neo4J instance.
 
-# Configuration
-A default configuration file `default.conf` can be found in the package
-directory. This contains the default values to use for logging and can also be
-used to store the information needed to access the local Neo4J instance.
+## Code examples / Tutorial
+
+To find code examples and tutorials, visit the
+[MAL Toolbox Tutorial](https://github.com/mal-lang/mal-toolbox-tutorial/tree/main) repository.
 
 # Tests
 There are unit tests inside of ./tests.

{mal_toolbox-0.0.40 → mal_toolbox-0.1.1}/mal_toolbox.egg-info/SOURCES.txt

@@ -19,7 +19,9 @@ maltoolbox/attackgraph/attacker.py
 maltoolbox/attackgraph/attackgraph.py
 maltoolbox/attackgraph/node.py
 maltoolbox/attackgraph/query.py
+maltoolbox/attackgraph/analyzers/__init__.py
 maltoolbox/attackgraph/analyzers/apriori.py
+maltoolbox/ingestors/__init__.py
 maltoolbox/ingestors/neo4j.py
 maltoolbox/language/__init__.py
 maltoolbox/language/classes_factory.py
@@ -29,5 +31,8 @@ maltoolbox/language/compiler/__init__.py
 maltoolbox/language/compiler/mal_lexer.py
 maltoolbox/language/compiler/mal_parser.py
 maltoolbox/language/compiler/mal_visitor.py
+maltoolbox/translators/__init__.py
 maltoolbox/translators/securicad.py
-tests/test_model.py
+maltoolbox/translators/updater.py
+tests/test_model.py
+tests/test_wrappers.py

{mal_toolbox-0.0.40 → mal_toolbox-0.1.1}/maltoolbox/__init__.py

@@ -1,5 +1,5 @@
 # -*- encoding: utf-8 -*-
-# MAL Toolbox v0.0.40
+# MAL Toolbox v0.1.1
 # Copyright 2024, Andrei Buhaiu.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
@@ -21,7 +21,7 @@ MAL-Toolbox Framework
 """
 
 __title__ = 'maltoolbox'
-__version__ = '0.0.40'
+__version__ = '0.1.1'
 __authors__ = ['Andrei Buhaiu',
     'Giuseppe Nebbione',
     'Nikolaos Kakouros',
@@ -33,7 +33,6 @@ __docformat__ = 'restructuredtext en'
 __all__ = ()
 
 import os
-import sys
 import configparser
 import logging
 
@@ -74,15 +73,18 @@ log_level = log_configs['log_level']
 if log_level != '':
     level = logging.getLevelName(log_level)
     logger.setLevel(level)
-    logger.info(f'Set loggin level of {__name__} to {log_level}.')
+    logger.info('Set loggin level of %s to %s.', __name__, log_level)
 
 if 'neo4j' in config:
     for term in ['uri', 'username', 'password', 'dbname']:
         if term not in config['neo4j']:
-            logger.critical('Config file is missing essential '\
-                f'Neo4J information: {term}, cannot proceed.')
-            raise ValueError('Config file is missing essential '\
-                f'Neo4J information: {term}, cannot proceed.')
+
+            msg = (
+                'Config file is missing essential Neo4J '
+                f'information: {term}, cannot proceed.'
+            )
+            logger.critical(msg)
+            raise ValueError(msg)
 
     neo4j_configs = {
         'uri': config['neo4j']['uri'],
@@ -90,4 +92,3 @@ if 'neo4j' in config:
         'password': config['neo4j']['password'],
         'dbname': config['neo4j']['dbname'],
     }
-

{mal_toolbox-0.0.40 → mal_toolbox-0.1.1}/maltoolbox/__main__.py

@@ -68,10 +68,9 @@ def generate_attack_graph(
             delete=False)
 
 
-def compile(lang_file, output_file):
+def compile(lang_file: str, output_file: str) -> None:
     """Compile language and dump into output file"""
     compiler = MalCompiler()
-
     with open(output_file, "w") as f:
         json.dump(compiler.compile(lang_file), f, indent=2)
 

{mal_toolbox-0.0.40 → mal_toolbox-0.1.1}/maltoolbox/attackgraph/analyzers/apriori.py

@@ -1,5 +1,3 @@
-from __future__ import annotations
-
 """
 MAL-Toolbox Attack Graph Apriori Analyzer Submodule
 
@@ -13,6 +11,7 @@ Currently these are:
   compromised) to compromise children attack steps.
 """
 
+from __future__ import annotations
 import logging
 
 from ..attackgraph import AttackGraph
@@ -20,14 +19,16 @@ from ..node import AttackGraphNode
 
 logger = logging.getLogger(__name__)
 
-def propagate_viability_from_node(node: AttackGraphNode):
+def propagate_viability_from_node(node: AttackGraphNode) -> None:
     """
     Arguments:
     node        - the attack graph node from which to propagate the viable
                   status
     """
-    logger.debug(f'Propagate viability from {node.id} with viability'
-        f' status {node.is_viable}.')
+    logger.debug(
+        'Propagate viability from "%s"(%d) with viability status %s.',
+        node.full_name, node.id, node.is_viable
+    )
     for child in node.children:
         original_value = child.is_viable
         if child.type == 'or':
@@ -40,14 +41,16 @@ def propagate_viability_from_node(node: AttackGraphNode):
         if child.is_viable != original_value:
             propagate_viability_from_node(child)
 
-def propagate_necessity_from_node(node: AttackGraphNode):
+def propagate_necessity_from_node(node: AttackGraphNode) -> None:
     """
     Arguments:
     node        - the attack graph node from which to propagate the necessary
                   status
     """
-    logger.debug(f'Propagate necessity from {node.id} with necessity'
-        f' status {node.is_necessary}.')
+    logger.debug(
+        'Propagate necessity from "%s"(%d) with necessity status %s.',
+        node.full_name, node.id, node.is_necessary
+    )
     for child in node.children:
         original_value = child.is_necessary
         if child.type == 'or':
@@ -63,17 +66,24 @@ def propagate_necessity_from_node(node: AttackGraphNode):
             propagate_necessity_from_node(child)
 
 
-def evaluate_viability(node):
+def evaluate_viability(node: AttackGraphNode) -> None:
     """
     Arguments:
     graph       - the node to evaluate viability for.
     """
     match (node.type):
         case 'exist':
+            assert isinstance(node.existence_status, bool), \
+                f'Existence status not defined for {node.full_name}.'
             node.is_viable = node.existence_status
         case 'notExist':
+            assert isinstance(node.existence_status, bool), \
+                f'Existence status not defined for {node.full_name}.'
             node.is_viable = not node.existence_status
         case 'defense':
+            assert node.defense_status is not None and \
+                   0.0 <= node.defense_status <= 1.0, \
+                f'{node.full_name} defense status invalid: {node.defense_status}.'
             node.is_viable = node.defense_status != 1.0
         case 'or':
             node.is_viable = False
@@ -84,22 +94,29 @@ def evaluate_viability(node):
             for parent in node.parents:
                 node.is_viable = node.is_viable and parent.is_viable
         case _:
-            msg = f'Evaluate viability was provided node \"node.id\" which '\
-                f'is of unknown type \"{node.type}\"'
-            logger.error(msg)
-            raise ValueError(msg)
+            msg = ('Evaluate viability was provided node "%s"(%d) which '
+                   'is of unknown type "%s"')
+            logger.error(msg, node.full_name, node.id, node.type)
+            raise ValueError(msg % (node.full_name, node.id, node.type))
 
-def evaluate_necessity(node):
+def evaluate_necessity(node: AttackGraphNode) -> None:
     """
     Arguments:
     graph       - the node to evaluate necessity for.
     """
     match (node.type):
         case 'exist':
+            assert isinstance(node.existence_status, bool), \
+                f'Existence status not defined for {node.full_name}.'
             node.is_necessary = not node.existence_status
         case 'notExist':
-            node.is_necessary = node.existence_status
+            assert isinstance(node.existence_status, bool), \
+                f'Existence status not defined for {node.full_name}.'
+            node.is_necessary = bool(node.existence_status)
         case 'defense':
+            assert node.defense_status is not None and \
+                   0.0 <= node.defense_status <= 1.0, \
+                f'{node.full_name} defense status invalid: {node.defense_status}.'
             node.is_necessary = node.defense_status != 0.0
         case 'or':
             node.is_necessary = True
@@ -110,12 +127,12 @@ def evaluate_necessity(node):
             for parent in node.parents:
                 node.is_necessary = node.is_necessary or parent.is_necessary
         case _:
-            msg = f'Evaluate necessity was provided node \"node.id\" which '\
-                f'is of unknown type \"{node.type}\"'
-            logger.error(msg)
-            raise ValueError(msg)
+            msg = ('Evaluate necessity was provided node "%s"(%d) which '
+                   'is of unknown type "%s"')
+            logger.error(msg, node.full_name, node.id, node.type)
+            raise ValueError(msg % (node.full_name, node.id, node.type))
 
-def evaluate_viability_and_necessity(node):
+def evaluate_viability_and_necessity(node: AttackGraphNode) -> None:
     """
     Arguments:
     graph       - the node to evaluate viability and necessity for.
@@ -123,7 +140,7 @@ def evaluate_viability_and_necessity(node):
     evaluate_viability(node)
     evaluate_necessity(node)
 
-def calculate_viability_and_necessity(graph: AttackGraph):
+def calculate_viability_and_necessity(graph: AttackGraph) -> None:
     """
     Arguments:
     graph       - the attack graph for which we wish to determine the
@@ -137,7 +154,7 @@ def calculate_viability_and_necessity(graph: AttackGraph):
             if not node.is_necessary:
                 propagate_necessity_from_node(node)
 
-def prune_unviable_and_unnecessary_nodes(graph: AttackGraph):
+def prune_unviable_and_unnecessary_nodes(graph: AttackGraph) -> None:
     """
     Arguments:
     graph       - the attack graph for which we wish to remove the

mal_toolbox-0.1.1/maltoolbox/attackgraph/attacker.py

@@ -0,0 +1,102 @@
+"""
+MAL-Toolbox Attack Graph Attacker Class
+"""
+
+from __future__ import annotations
+from dataclasses import dataclass, field
+import logging
+
+from typing import Optional
+from typing import TYPE_CHECKING
+if TYPE_CHECKING:
+    from .attackgraph import AttackGraphNode
+
+logger = logging.getLogger(__name__)
+
+@dataclass
+class Attacker:
+    name: str
+    entry_points: list[AttackGraphNode] = field(default_factory=list)
+    reached_attack_steps: list[AttackGraphNode] = \
+        field(default_factory=list)
+    id: Optional[int] = None
+
+    def to_dict(self) -> dict:
+        attacker_dict: dict = {
+            'id': self.id,
+            'name': self.name,
+            'entry_points': {},
+            'reached_attack_steps': {}
+        }
+
+        for entry_point in self.entry_points:
+            attacker_dict['entry_points'][entry_point.id] = \
+                entry_point.full_name
+        for attack_step in self.reached_attack_steps:
+            attacker_dict['reached_attack_steps'][attack_step.id] = \
+                attack_step.full_name
+
+        return attacker_dict
+
+    def __repr__(self) -> str:
+        return str(self.to_dict())
+
+    def compromise(self, node: AttackGraphNode) -> None:
+        """
+        Have the attacke compromise the node given as a parameter.
+
+        Arguments:
+        node    - the node that the attacker will compromise
+        """
+
+        logger.debug(
+            'Attacker "%s"(%d) is compromising node "%s"(%d).',
+            self.name,
+            self.id,
+            node.full_name,
+            node.id
+        )
+        if node.is_compromised_by(self):
+            logger.info(
+                'Attacker "%s"(%d) already compromised node "%s"(%d). '
+                'Do nothing.',
+                self.name,
+                self.id,
+                node.full_name,
+                node.id
+            )
+            return
+
+        node.compromised_by.append(self)
+        self.reached_attack_steps.append(node)
+
+    def undo_compromise(self, node: AttackGraphNode) -> None:
+        """
+        Remove the attacker from the list of attackers that have compromised
+        the node given as a parameter.
+
+        Arguments:
+        node    - the node that we wish to remove this attacker from.
+        """
+
+        logger.debug(
+            'Removing attacker "%s"(%d) from compromised_by '
+            'list of node "%s"(%d).',
+            self.name,
+            self.id,
+            node.full_name,
+            node.id
+        )
+        if not node.is_compromised_by(self):
+            logger.info(
+                'Attacker "%s"(%d) had not compromised node "%s"(%d).'
+                ' Do nothing.',
+                self.name,
+                self.id,
+                node.full_name,
+                node.id
+            )
+            return
+
+        node.compromised_by.remove(self)
+        self.reached_attack_steps.remove(node)