# ATLAS

## Docs

- [Trainers Reference](https://docs.arc.computer/api-reference/trainers.md): Complete reference for ATLAS trainer classes and methods
- [Evaluation Harnesses](https://docs.arc.computer/benchmarks/evaluation-harnesses.md): Measure Atlas runtime, reward judges, and learning progress with the SDK benchmarking scripts.
- [Evaluation Methodology](https://docs.arc.computer/benchmarks/evaluation-methodology.md): Comprehensive testing protocol for verifying ATLAS performance
- [Reproduction Guide](https://docs.arc.computer/benchmarks/reproduction.md): Step-by-step instructions to reproduce ATLAS benchmark results
- [Adaptive Dual-Agent Reasoning](https://docs.arc.computer/concepts/adaptive-dual-agent-reasoning.md): How your agent partners with a verifying teacher to deliver safer, higher-quality results.
- [Hybrid Learning Architecture](https://docs.arc.computer/concepts/hybrid-learning.md): Understanding ATLAS's dual-phase approach to model enhancement
- [The ATLAS Reward System](https://docs.arc.computer/concepts/reward-design.md): How ATLAS measures if teaching actually works
- [Adaptive Tool Use with MCP](https://docs.arc.computer/examples/adaptive-tool-use.md): Production-ready example showing measurable tool efficiency improvements through progressive learning
- [Developer Example: Running GKD](https://docs.arc.computer/examples/gkd-dev-example.md): Step-by-step walkthrough for exporting traces and running AtlasGKDTrainer
- [Installation](https://docs.arc.computer/installation.md): Set up ATLAS environment with validated dependencies
- [Introduction](https://docs.arc.computer/introduction.md): A Continual Learning Framework for Production LLM Agents
- [Telemetry Schema](https://docs.arc.computer/reference/database-schema.md): Understand the core Atlas SDK tables used for discovery, runtime telemetry, and learning persistence.
- [Datasets](https://docs.arc.computer/reference/datasets.md): Official ATLAS training and evaluation datasets
- [Frequently Asked Questions](https://docs.arc.computer/reference/faq.md): Common questions about ATLAS implementation and usage
- [Glossary](https://docs.arc.computer/reference/glossary.md): Key terms and concepts used in ATLAS documentation
- [Models](https://docs.arc.computer/reference/models.md): Pre-trained ATLAS teacher models available on Hugging Face
- [Technical Report](https://docs.arc.computer/reference/technical-report.md): ATLAS research paper and technical specifications
- [Troubleshooting](https://docs.arc.computer/reference/troubleshooting.md): Common issues and solutions for ATLAS deployment
- [Bring Your Own Agent](https://docs.arc.computer/sdk/adapters.md): Connect the Atlas orchestrator to any agent via OpenAI, Python, or HTTP adapters.
- [Atlas CLI Reference](https://docs.arc.computer/sdk/cli-reference.md): Command catalogue for discovery, configuration scaffolding, and runtime execution in the Atlas SDK.
- [SDK Configuration Reference](https://docs.arc.computer/sdk/configuration.md): Understand every block in an Atlas SDK YAML file so you can tailor the orchestrator to your agents.
- [Export Runtime Traces](https://docs.arc.computer/sdk/export-traces.md): Access Atlas SDK session data via direct database queries or JSONL export
- [Learning System Architecture](https://docs.arc.computer/sdk/learning-system.md): Map the decoupled learning synthesis pipeline, telemetry stores, and runtime controls that drive Atlas SDK learning.
- [How Orchestration Works](https://docs.arc.computer/sdk/orchestration.md): Follow the Student, Teacher, and Reward System as they coordinate an Atlas SDK run from plan to final answer.
- [SDK Quickstart: Run Your First Task](https://docs.arc.computer/sdk/quickstart.md): Launch the Atlas SDK runtime, run your first task, and understand how the dual-agent loop (your student agent + verifying teacher) orchestrates work.
- [Runtime Safety & Review](https://docs.arc.computer/sdk/runtime-safety.md): Configure Atlas drift guardrails and session review gating to protect production learning.
- [Training Configuration](https://docs.arc.computer/training/configuration.md): Hydra parameter reference for Atlas Core training
- [Custom Dataset Creation](https://docs.arc.computer/training/custom-datasets.md): How to create custom training datasets from runtime traces
- [GKD Training](https://docs.arc.computer/training/offline/gkd-training.md): On-policy distillation of Atlas runtime traces using Generalized Knowledge Distillation
- [Training Your Own Teacher Model](https://docs.arc.computer/training/offline/grpo-training.md): Complete guide to custom teacher model training with GRPO
- [Training Data Pipeline](https://docs.arc.computer/training/offline/training-data-pipeline.md): Direct database access for training data extraction and preprocessing
- [Reward System Implementation](https://docs.arc.computer/training/reward-system-usage.md): How to use and customize the ATLAS reward system in code

## OpenAPI Specs

- [openapi](https://docs.arc.computer/api-reference/openapi.json)