dialectus-engine 0.1.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -0,0 +1,9 @@
1
+ MIT License
2
+
3
+ Copyright (c) 2025 Dialectus AI
4
+
5
+ Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
6
+
7
+ The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
8
+
9
+ THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
@@ -0,0 +1,216 @@
1
+ Metadata-Version: 2.4
2
+ Name: dialectus-engine
3
+ Version: 0.1.0
4
+ Summary: AI debate engine framework - importable Python library
5
+ Author: Dialectus AI
6
+ License-Expression: MIT
7
+ Project-URL: Homepage, https://github.com/dialectus-ai/dialectus-engine
8
+ Project-URL: Repository, https://github.com/dialectus-ai/dialectus-engine
9
+ Project-URL: Issues, https://github.com/dialectus-ai/dialectus-engine/issues
10
+ Keywords: ai,debate,llm,argumentation,ollama,openrouter,artificial-intelligence
11
+ Classifier: Development Status :: 4 - Beta
12
+ Classifier: Intended Audience :: Developers
13
+ Classifier: Intended Audience :: Education
14
+ Classifier: Programming Language :: Python :: 3
15
+ Classifier: Programming Language :: Python :: 3.13
16
+ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
17
+ Classifier: Topic :: Software Development :: Libraries :: Python Modules
18
+ Requires-Python: >=3.13
19
+ Description-Content-Type: text/markdown
20
+ License-File: LICENSE
21
+ Requires-Dist: httpx>=0.28.1
22
+ Requires-Dist: openai>=1.105.0
23
+ Requires-Dist: pydantic>=2.11.7
24
+ Provides-Extra: dev
25
+ Requires-Dist: pip-tools; extra == "dev"
26
+ Dynamic: license-file
27
+
28
+ <img src="./assets/logo.png" alt="Dialectus Engine" width="500">
29
+
30
+ <br />
31
+
32
+ # Dialectus Engine
33
+
34
+ A Python library for orchestrating AI-powered debates with multi-provider model support.
35
+
36
+ ![Python](https://img.shields.io/badge/python-3.13+-blue.svg)
37
+ ![License](https://img.shields.io/badge/license-MIT-green.svg)
38
+ ![Status](https://img.shields.io/badge/status-Production%20Ready-brightgreen.svg)
39
+
40
+ ## Overview
41
+
42
+ The Dialectus Engine is a standalone Python library that provides core debate orchestration logic, including participant coordination, turn management, AI judge integration, and multi-provider model support. It's designed to be imported and used by other applications to build debate systems.
43
+
44
+ ## Components
45
+
46
+ - **Core Engine** (`debate_engine/`) - Main debate orchestration logic
47
+ - **Models** (`models/`) - AI model provider integrations (Ollama, OpenRouter)
48
+ - **Configuration** (`config/`) - System configuration management
49
+ - **Judges** (`judges/`) - AI judge implementations with ensemble support
50
+ - **Formats** (`formats/`) - Debate format definitions (Oxford, Parliamentary, Socratic, Public Forum)
51
+
52
+ ## Installation
53
+
54
+ ### From PyPI
55
+
56
+ ```bash
57
+ pip install dialectus-engine
58
+ ```
59
+
60
+ ### From Source
61
+
62
+ ```bash
63
+ # Clone the repository
64
+ git clone https://github.com/dialectus-ai/dialectus-engine.git
65
+ cd dialectus-engine
66
+
67
+ # Install in development mode
68
+ pip install -e .
69
+
70
+ # Or install with dev dependencies
71
+ pip install -e ".[dev]"
72
+ ```
73
+
74
+ ### As a Dependency
75
+
76
+ Add to your `requirements.txt` or `pyproject.toml`:
77
+
78
+ ```
79
+ dialectus-engine>=0.1.0
80
+ ```
81
+
82
+ Or install directly from git:
83
+
84
+ ```bash
85
+ pip install git+https://github.com/dialectus-ai/dialectus-engine.git@main
86
+ ```
87
+
88
+ ## Quick Start
89
+
90
+ ```python
91
+ import asyncio
92
+ from debate_engine.core import DebateEngine
93
+ from models.manager import ModelManager
94
+ from config.settings import AppConfig, ModelConfig
95
+
96
+ async def run_debate():
97
+ # Load configuration
98
+ config = AppConfig.from_json_file("debate_config.json")
99
+
100
+ # Set up model manager
101
+ model_manager = ModelManager()
102
+
103
+ # Register models
104
+ for model_id, model_config in config.models.items():
105
+ model_manager.register_model(model_id, model_config)
106
+
107
+ # Create debate engine
108
+ engine = DebateEngine(
109
+ config=config,
110
+ model_manager=model_manager
111
+ )
112
+
113
+ # Run debate
114
+ transcript = await engine.run_debate()
115
+ print(transcript)
116
+
117
+ asyncio.run(run_debate())
118
+ ```
119
+
120
+ ## Configuration
121
+
122
+ The engine uses `debate_config.json` for system configuration. To get started:
123
+
124
+ ```bash
125
+ # Copy the example configuration
126
+ cp debate_config.example.json debate_config.json
127
+
128
+ # Edit with your settings and API keys
129
+ nano debate_config.json
130
+ ```
131
+
132
+ Key configuration sections:
133
+ - **Models**: Define debate participants with provider, personality, and parameters
134
+ - **Providers**: Configure Ollama (local) and OpenRouter (cloud) settings
135
+ - **Judging**: Set evaluation criteria and judge models
136
+ - **Debate**: Default topic, format, and word limits
137
+
138
+ For detailed configuration documentation, see [CONFIG_GUIDE.md](CONFIG_GUIDE.md).
139
+
140
+ ## Features
141
+
142
+ ### Multi-Provider Model Support
143
+ - **Ollama**: Local model management with hardware optimization
144
+ - **OpenRouter**: Cloud model access to 100+ models
145
+ - **Async streaming**: Chunk-by-chunk response generation
146
+ - **Auto-discovery**: Dynamic model listing from all configured providers
147
+ - **Caching**: In-memory cache with TTL for model metadata
148
+
149
+ ### Debate Formats
150
+ - **Oxford**: Classic opening/rebuttal/closing structure
151
+ - **Parliamentary**: British-style government vs. opposition
152
+ - **Socratic**: Question-driven dialogue format
153
+ - **Public Forum**: American high school debate style
154
+
155
+ ### AI Judge System
156
+ - **LLM-based evaluation**: Detailed criterion scoring
157
+ - **Ensemble judging**: Aggregate decisions from multiple judges
158
+ - **Structured decisions**: JSON-serializable judge results
159
+ - **Configurable criteria**: Logic, evidence, persuasiveness, etc.
160
+
161
+ ## Architecture
162
+
163
+ Key architectural principles:
164
+ - **Library-first**: Designed to be imported by other applications
165
+ - **Provider agnostic**: Support for multiple AI model sources
166
+ - **Async by default**: All model interactions are async
167
+ - **Type-safe**: Strict Pyright configuration with modern type hints
168
+ - **Pydantic everywhere**: All config and data models use Pydantic v2
169
+ - **Configurable**: JSON-based configuration with validation
170
+
171
+ ### Technology Stack
172
+ - **Python 3.13+** with modern type hints (`X | None`, `list[T]`, `dict[K, V]`)
173
+ - **Pydantic v2** for data validation and settings management
174
+ - **OpenAI SDK** for OpenRouter API integration (streaming support)
175
+ - **httpx** for async HTTP requests (Ollama provider)
176
+ - **asyncio** for concurrent debate operations
177
+
178
+ ## Usage Examples
179
+
180
+ ### Listing Available Models
181
+
182
+ ```python
183
+ from models.manager import ModelManager
184
+
185
+ async def list_models():
186
+ manager = ModelManager()
187
+ models = await manager.get_all_models()
188
+ for model_id, model_info in models.items():
189
+ print(f"{model_id}: {model_info.description}")
190
+ ```
191
+
192
+ ### Running a Custom Format
193
+
194
+ ```python
195
+ from formats.registry import format_registry
196
+
197
+ # Get available formats
198
+ formats = format_registry.list_formats()
199
+
200
+ # Load a specific format
201
+ oxford = format_registry.get_format("oxford")
202
+ phases = oxford.phases()
203
+ ```
204
+
205
+ ### Ensemble Judging
206
+
207
+ ```python
208
+ from judges.factory import JudgeFactory
209
+
210
+ # Create judge with multiple models
211
+ config.judging.judge_models = ["openthinker:7b", "llama3.2:3b", "qwen2.5:3b"]
212
+ judge = JudgeFactory.create_judge(config.judging, model_manager)
213
+
214
+ # Get aggregated decision
215
+ decision = await judge.judge_debate(context)
216
+ ```
@@ -0,0 +1,189 @@
1
+ <img src="./assets/logo.png" alt="Dialectus Engine" width="500">
2
+
3
+ <br />
4
+
5
+ # Dialectus Engine
6
+
7
+ A Python library for orchestrating AI-powered debates with multi-provider model support.
8
+
9
+ ![Python](https://img.shields.io/badge/python-3.13+-blue.svg)
10
+ ![License](https://img.shields.io/badge/license-MIT-green.svg)
11
+ ![Status](https://img.shields.io/badge/status-Production%20Ready-brightgreen.svg)
12
+
13
+ ## Overview
14
+
15
+ The Dialectus Engine is a standalone Python library that provides core debate orchestration logic, including participant coordination, turn management, AI judge integration, and multi-provider model support. It's designed to be imported and used by other applications to build debate systems.
16
+
17
+ ## Components
18
+
19
+ - **Core Engine** (`debate_engine/`) - Main debate orchestration logic
20
+ - **Models** (`models/`) - AI model provider integrations (Ollama, OpenRouter)
21
+ - **Configuration** (`config/`) - System configuration management
22
+ - **Judges** (`judges/`) - AI judge implementations with ensemble support
23
+ - **Formats** (`formats/`) - Debate format definitions (Oxford, Parliamentary, Socratic, Public Forum)
24
+
25
+ ## Installation
26
+
27
+ ### From PyPI
28
+
29
+ ```bash
30
+ pip install dialectus-engine
31
+ ```
32
+
33
+ ### From Source
34
+
35
+ ```bash
36
+ # Clone the repository
37
+ git clone https://github.com/dialectus-ai/dialectus-engine.git
38
+ cd dialectus-engine
39
+
40
+ # Install in development mode
41
+ pip install -e .
42
+
43
+ # Or install with dev dependencies
44
+ pip install -e ".[dev]"
45
+ ```
46
+
47
+ ### As a Dependency
48
+
49
+ Add to your `requirements.txt` or `pyproject.toml`:
50
+
51
+ ```
52
+ dialectus-engine>=0.1.0
53
+ ```
54
+
55
+ Or install directly from git:
56
+
57
+ ```bash
58
+ pip install git+https://github.com/dialectus-ai/dialectus-engine.git@main
59
+ ```
60
+
61
+ ## Quick Start
62
+
63
+ ```python
64
+ import asyncio
65
+ from debate_engine.core import DebateEngine
66
+ from models.manager import ModelManager
67
+ from config.settings import AppConfig, ModelConfig
68
+
69
+ async def run_debate():
70
+ # Load configuration
71
+ config = AppConfig.from_json_file("debate_config.json")
72
+
73
+ # Set up model manager
74
+ model_manager = ModelManager()
75
+
76
+ # Register models
77
+ for model_id, model_config in config.models.items():
78
+ model_manager.register_model(model_id, model_config)
79
+
80
+ # Create debate engine
81
+ engine = DebateEngine(
82
+ config=config,
83
+ model_manager=model_manager
84
+ )
85
+
86
+ # Run debate
87
+ transcript = await engine.run_debate()
88
+ print(transcript)
89
+
90
+ asyncio.run(run_debate())
91
+ ```
92
+
93
+ ## Configuration
94
+
95
+ The engine uses `debate_config.json` for system configuration. To get started:
96
+
97
+ ```bash
98
+ # Copy the example configuration
99
+ cp debate_config.example.json debate_config.json
100
+
101
+ # Edit with your settings and API keys
102
+ nano debate_config.json
103
+ ```
104
+
105
+ Key configuration sections:
106
+ - **Models**: Define debate participants with provider, personality, and parameters
107
+ - **Providers**: Configure Ollama (local) and OpenRouter (cloud) settings
108
+ - **Judging**: Set evaluation criteria and judge models
109
+ - **Debate**: Default topic, format, and word limits
110
+
111
+ For detailed configuration documentation, see [CONFIG_GUIDE.md](CONFIG_GUIDE.md).
112
+
113
+ ## Features
114
+
115
+ ### Multi-Provider Model Support
116
+ - **Ollama**: Local model management with hardware optimization
117
+ - **OpenRouter**: Cloud model access to 100+ models
118
+ - **Async streaming**: Chunk-by-chunk response generation
119
+ - **Auto-discovery**: Dynamic model listing from all configured providers
120
+ - **Caching**: In-memory cache with TTL for model metadata
121
+
122
+ ### Debate Formats
123
+ - **Oxford**: Classic opening/rebuttal/closing structure
124
+ - **Parliamentary**: British-style government vs. opposition
125
+ - **Socratic**: Question-driven dialogue format
126
+ - **Public Forum**: American high school debate style
127
+
128
+ ### AI Judge System
129
+ - **LLM-based evaluation**: Detailed criterion scoring
130
+ - **Ensemble judging**: Aggregate decisions from multiple judges
131
+ - **Structured decisions**: JSON-serializable judge results
132
+ - **Configurable criteria**: Logic, evidence, persuasiveness, etc.
133
+
134
+ ## Architecture
135
+
136
+ Key architectural principles:
137
+ - **Library-first**: Designed to be imported by other applications
138
+ - **Provider agnostic**: Support for multiple AI model sources
139
+ - **Async by default**: All model interactions are async
140
+ - **Type-safe**: Strict Pyright configuration with modern type hints
141
+ - **Pydantic everywhere**: All config and data models use Pydantic v2
142
+ - **Configurable**: JSON-based configuration with validation
143
+
144
+ ### Technology Stack
145
+ - **Python 3.13+** with modern type hints (`X | None`, `list[T]`, `dict[K, V]`)
146
+ - **Pydantic v2** for data validation and settings management
147
+ - **OpenAI SDK** for OpenRouter API integration (streaming support)
148
+ - **httpx** for async HTTP requests (Ollama provider)
149
+ - **asyncio** for concurrent debate operations
150
+
151
+ ## Usage Examples
152
+
153
+ ### Listing Available Models
154
+
155
+ ```python
156
+ from models.manager import ModelManager
157
+
158
+ async def list_models():
159
+ manager = ModelManager()
160
+ models = await manager.get_all_models()
161
+ for model_id, model_info in models.items():
162
+ print(f"{model_id}: {model_info.description}")
163
+ ```
164
+
165
+ ### Running a Custom Format
166
+
167
+ ```python
168
+ from formats.registry import format_registry
169
+
170
+ # Get available formats
171
+ formats = format_registry.list_formats()
172
+
173
+ # Load a specific format
174
+ oxford = format_registry.get_format("oxford")
175
+ phases = oxford.phases()
176
+ ```
177
+
178
+ ### Ensemble Judging
179
+
180
+ ```python
181
+ from judges.factory import JudgeFactory
182
+
183
+ # Create judge with multiple models
184
+ config.judging.judge_models = ["openthinker:7b", "llama3.2:3b", "qwen2.5:3b"]
185
+ judge = JudgeFactory.create_judge(config.judging, model_manager)
186
+
187
+ # Get aggregated decision
188
+ decision = await judge.judge_debate(context)
189
+ ```
@@ -0,0 +1,3 @@
1
+ """Dialectus - AI Debate System"""
2
+
3
+ __version__ = "0.1.0"
@@ -0,0 +1,6 @@
1
+ """Dialectus Engine - Core debate orchestration library"""
2
+
3
+ from dialectus.engine.debate_engine.engine import DebateEngine
4
+ from dialectus.engine.debate_engine.models import DebateContext, DebateMessage
5
+
6
+ __all__ = ["DebateEngine", "DebateContext", "DebateMessage"]
@@ -0,0 +1,216 @@
1
+ Metadata-Version: 2.4
2
+ Name: dialectus-engine
3
+ Version: 0.1.0
4
+ Summary: AI debate engine framework - importable Python library
5
+ Author: Dialectus AI
6
+ License-Expression: MIT
7
+ Project-URL: Homepage, https://github.com/dialectus-ai/dialectus-engine
8
+ Project-URL: Repository, https://github.com/dialectus-ai/dialectus-engine
9
+ Project-URL: Issues, https://github.com/dialectus-ai/dialectus-engine/issues
10
+ Keywords: ai,debate,llm,argumentation,ollama,openrouter,artificial-intelligence
11
+ Classifier: Development Status :: 4 - Beta
12
+ Classifier: Intended Audience :: Developers
13
+ Classifier: Intended Audience :: Education
14
+ Classifier: Programming Language :: Python :: 3
15
+ Classifier: Programming Language :: Python :: 3.13
16
+ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
17
+ Classifier: Topic :: Software Development :: Libraries :: Python Modules
18
+ Requires-Python: >=3.13
19
+ Description-Content-Type: text/markdown
20
+ License-File: LICENSE
21
+ Requires-Dist: httpx>=0.28.1
22
+ Requires-Dist: openai>=1.105.0
23
+ Requires-Dist: pydantic>=2.11.7
24
+ Provides-Extra: dev
25
+ Requires-Dist: pip-tools; extra == "dev"
26
+ Dynamic: license-file
27
+
28
+ <img src="./assets/logo.png" alt="Dialectus Engine" width="500">
29
+
30
+ <br />
31
+
32
+ # Dialectus Engine
33
+
34
+ A Python library for orchestrating AI-powered debates with multi-provider model support.
35
+
36
+ ![Python](https://img.shields.io/badge/python-3.13+-blue.svg)
37
+ ![License](https://img.shields.io/badge/license-MIT-green.svg)
38
+ ![Status](https://img.shields.io/badge/status-Production%20Ready-brightgreen.svg)
39
+
40
+ ## Overview
41
+
42
+ The Dialectus Engine is a standalone Python library that provides core debate orchestration logic, including participant coordination, turn management, AI judge integration, and multi-provider model support. It's designed to be imported and used by other applications to build debate systems.
43
+
44
+ ## Components
45
+
46
+ - **Core Engine** (`debate_engine/`) - Main debate orchestration logic
47
+ - **Models** (`models/`) - AI model provider integrations (Ollama, OpenRouter)
48
+ - **Configuration** (`config/`) - System configuration management
49
+ - **Judges** (`judges/`) - AI judge implementations with ensemble support
50
+ - **Formats** (`formats/`) - Debate format definitions (Oxford, Parliamentary, Socratic, Public Forum)
51
+
52
+ ## Installation
53
+
54
+ ### From PyPI
55
+
56
+ ```bash
57
+ pip install dialectus-engine
58
+ ```
59
+
60
+ ### From Source
61
+
62
+ ```bash
63
+ # Clone the repository
64
+ git clone https://github.com/dialectus-ai/dialectus-engine.git
65
+ cd dialectus-engine
66
+
67
+ # Install in development mode
68
+ pip install -e .
69
+
70
+ # Or install with dev dependencies
71
+ pip install -e ".[dev]"
72
+ ```
73
+
74
+ ### As a Dependency
75
+
76
+ Add to your `requirements.txt` or `pyproject.toml`:
77
+
78
+ ```
79
+ dialectus-engine>=0.1.0
80
+ ```
81
+
82
+ Or install directly from git:
83
+
84
+ ```bash
85
+ pip install git+https://github.com/dialectus-ai/dialectus-engine.git@main
86
+ ```
87
+
88
+ ## Quick Start
89
+
90
+ ```python
91
+ import asyncio
92
+ from debate_engine.core import DebateEngine
93
+ from models.manager import ModelManager
94
+ from config.settings import AppConfig, ModelConfig
95
+
96
+ async def run_debate():
97
+ # Load configuration
98
+ config = AppConfig.from_json_file("debate_config.json")
99
+
100
+ # Set up model manager
101
+ model_manager = ModelManager()
102
+
103
+ # Register models
104
+ for model_id, model_config in config.models.items():
105
+ model_manager.register_model(model_id, model_config)
106
+
107
+ # Create debate engine
108
+ engine = DebateEngine(
109
+ config=config,
110
+ model_manager=model_manager
111
+ )
112
+
113
+ # Run debate
114
+ transcript = await engine.run_debate()
115
+ print(transcript)
116
+
117
+ asyncio.run(run_debate())
118
+ ```
119
+
120
+ ## Configuration
121
+
122
+ The engine uses `debate_config.json` for system configuration. To get started:
123
+
124
+ ```bash
125
+ # Copy the example configuration
126
+ cp debate_config.example.json debate_config.json
127
+
128
+ # Edit with your settings and API keys
129
+ nano debate_config.json
130
+ ```
131
+
132
+ Key configuration sections:
133
+ - **Models**: Define debate participants with provider, personality, and parameters
134
+ - **Providers**: Configure Ollama (local) and OpenRouter (cloud) settings
135
+ - **Judging**: Set evaluation criteria and judge models
136
+ - **Debate**: Default topic, format, and word limits
137
+
138
+ For detailed configuration documentation, see [CONFIG_GUIDE.md](CONFIG_GUIDE.md).
139
+
140
+ ## Features
141
+
142
+ ### Multi-Provider Model Support
143
+ - **Ollama**: Local model management with hardware optimization
144
+ - **OpenRouter**: Cloud model access to 100+ models
145
+ - **Async streaming**: Chunk-by-chunk response generation
146
+ - **Auto-discovery**: Dynamic model listing from all configured providers
147
+ - **Caching**: In-memory cache with TTL for model metadata
148
+
149
+ ### Debate Formats
150
+ - **Oxford**: Classic opening/rebuttal/closing structure
151
+ - **Parliamentary**: British-style government vs. opposition
152
+ - **Socratic**: Question-driven dialogue format
153
+ - **Public Forum**: American high school debate style
154
+
155
+ ### AI Judge System
156
+ - **LLM-based evaluation**: Detailed criterion scoring
157
+ - **Ensemble judging**: Aggregate decisions from multiple judges
158
+ - **Structured decisions**: JSON-serializable judge results
159
+ - **Configurable criteria**: Logic, evidence, persuasiveness, etc.
160
+
161
+ ## Architecture
162
+
163
+ Key architectural principles:
164
+ - **Library-first**: Designed to be imported by other applications
165
+ - **Provider agnostic**: Support for multiple AI model sources
166
+ - **Async by default**: All model interactions are async
167
+ - **Type-safe**: Strict Pyright configuration with modern type hints
168
+ - **Pydantic everywhere**: All config and data models use Pydantic v2
169
+ - **Configurable**: JSON-based configuration with validation
170
+
171
+ ### Technology Stack
172
+ - **Python 3.13+** with modern type hints (`X | None`, `list[T]`, `dict[K, V]`)
173
+ - **Pydantic v2** for data validation and settings management
174
+ - **OpenAI SDK** for OpenRouter API integration (streaming support)
175
+ - **httpx** for async HTTP requests (Ollama provider)
176
+ - **asyncio** for concurrent debate operations
177
+
178
+ ## Usage Examples
179
+
180
+ ### Listing Available Models
181
+
182
+ ```python
183
+ from models.manager import ModelManager
184
+
185
+ async def list_models():
186
+ manager = ModelManager()
187
+ models = await manager.get_all_models()
188
+ for model_id, model_info in models.items():
189
+ print(f"{model_id}: {model_info.description}")
190
+ ```
191
+
192
+ ### Running a Custom Format
193
+
194
+ ```python
195
+ from formats.registry import format_registry
196
+
197
+ # Get available formats
198
+ formats = format_registry.list_formats()
199
+
200
+ # Load a specific format
201
+ oxford = format_registry.get_format("oxford")
202
+ phases = oxford.phases()
203
+ ```
204
+
205
+ ### Ensemble Judging
206
+
207
+ ```python
208
+ from judges.factory import JudgeFactory
209
+
210
+ # Create judge with multiple models
211
+ config.judging.judge_models = ["openthinker:7b", "llama3.2:3b", "qwen2.5:3b"]
212
+ judge = JudgeFactory.create_judge(config.judging, model_manager)
213
+
214
+ # Get aggregated decision
215
+ decision = await judge.judge_debate(context)
216
+ ```
@@ -0,0 +1,10 @@
1
+ LICENSE
2
+ README.md
3
+ pyproject.toml
4
+ dialectus/__init__.py
5
+ dialectus/engine/__init__.py
6
+ dialectus_engine.egg-info/PKG-INFO
7
+ dialectus_engine.egg-info/SOURCES.txt
8
+ dialectus_engine.egg-info/dependency_links.txt
9
+ dialectus_engine.egg-info/requires.txt
10
+ dialectus_engine.egg-info/top_level.txt
@@ -0,0 +1,6 @@
1
+ httpx>=0.28.1
2
+ openai>=1.105.0
3
+ pydantic>=2.11.7
4
+
5
+ [dev]
6
+ pip-tools
@@ -0,0 +1,42 @@
1
+ [build-system]
2
+ requires = ["setuptools>=61.0", "wheel"]
3
+ build-backend = "setuptools.build_meta"
4
+
5
+ [project]
6
+ name = "dialectus-engine"
7
+ version = "0.1.0"
8
+ description = "AI debate engine framework - importable Python library"
9
+ readme = "README.md"
10
+ license = "MIT"
11
+ requires-python = ">=3.13"
12
+ authors = [
13
+ {name = "Dialectus AI"}
14
+ ]
15
+ keywords = ["ai", "debate", "llm", "argumentation", "ollama", "openrouter", "artificial-intelligence"]
16
+ classifiers = [
17
+ "Development Status :: 4 - Beta",
18
+ "Intended Audience :: Developers",
19
+ "Intended Audience :: Education",
20
+ "Programming Language :: Python :: 3",
21
+ "Programming Language :: Python :: 3.13",
22
+ "Topic :: Scientific/Engineering :: Artificial Intelligence",
23
+ "Topic :: Software Development :: Libraries :: Python Modules",
24
+ ]
25
+ dependencies = [
26
+ "httpx>=0.28.1",
27
+ "openai>=1.105.0",
28
+ "pydantic>=2.11.7",
29
+ ]
30
+
31
+ [project.urls]
32
+ Homepage = "https://github.com/dialectus-ai/dialectus-engine"
33
+ Repository = "https://github.com/dialectus-ai/dialectus-engine"
34
+ Issues = "https://github.com/dialectus-ai/dialectus-engine/issues"
35
+
36
+ [project.optional-dependencies]
37
+ dev = [
38
+ "pip-tools",
39
+ ]
40
+
41
+ [tool.setuptools]
42
+ packages = ["dialectus", "dialectus.engine"]
@@ -0,0 +1,4 @@
1
+ [egg_info]
2
+ tag_build =
3
+ tag_date = 0
4
+