# SkyDiscover

## Docs

- [AdaEvolve](https://mintlify.wiki/skydiscover-ai/skydiscover/algorithms/adaevolve.md): Multi-island adaptive search with UCB-based island selection and paradigm breakthroughs
- [Beam Search](https://mintlify.wiki/skydiscover-ai/skydiscover/algorithms/beam-search.md): Breadth-first expansion maintaining a beam of top solution candidates
- [Best-of-N](https://mintlify.wiki/skydiscover-ai/skydiscover/algorithms/best-of-n.md): Generate N variants per iteration and keep the best one
- [EvoX](https://mintlify.wiki/skydiscover-ai/skydiscover/algorithms/evox.md): Self-evolving search paradigm that co-adapts solution generation and experience management
- [GEPA Native](https://mintlify.wiki/skydiscover-ai/skydiscover/algorithms/gepa-native.md): Pareto-efficient search with reflective prompting and LLM-mediated merge
- [OpenEvolve Native](https://mintlify.wiki/skydiscover-ai/skydiscover/algorithms/openevolve-native.md): MAP-Elites with island-based evolutionary search
- [Top-K](https://mintlify.wiki/skydiscover-ai/skydiscover/algorithms/topk.md): Simple and effective algorithm that selects the top-K solutions for refinement
- [discover_solution](https://mintlify.wiki/skydiscover-ai/skydiscover/api/discover-solution.md): Convenience wrapper for evolving string solutions with callable evaluators
- [DiscoveryResult](https://mintlify.wiki/skydiscover-ai/skydiscover/api/discovery-result.md): Result object and Program dataclass for discovery runs
- [run_discovery](https://mintlify.wiki/skydiscover-ai/skydiscover/api/run-discovery.md): Run a discovery process and return the best result
- [Runner](https://mintlify.wiki/skydiscover-ai/skydiscover/api/runner.md): Top-level class for managing discovery runs with checkpointing and monitoring
- [CLI Flags Reference](https://mintlify.wiki/skydiscover-ai/skydiscover/cli/flags.md): Complete reference for all SkyDiscover command-line flags and arguments
- [skydiscover-run](https://mintlify.wiki/skydiscover-ai/skydiscover/cli/skydiscover-run.md): Main CLI command to run SkyDiscover evolutionary discovery
- [skydiscover-viewer](https://mintlify.wiki/skydiscover-ai/skydiscover/cli/skydiscover-viewer.md): Replay and visualize completed SkyDiscover runs
- [Search Algorithms](https://mintlify.wiki/skydiscover-ai/skydiscover/concepts/algorithms.md): Understanding SkyDiscover's search algorithms and when to use each one
- [Framework Architecture](https://mintlify.wiki/skydiscover-ai/skydiscover/concepts/architecture.md): Detailed look at SkyDiscover's modular design and component interactions
- [Evaluators](https://mintlify.wiki/skydiscover-ai/skydiscover/concepts/evaluators.md): Writing effective evaluation functions to guide discovery
- [Evolution Blocks](https://mintlify.wiki/skydiscover-ai/skydiscover/concepts/evolution-blocks.md): Control which parts of your code get evolved with EVOLVE-BLOCK markers
- [Core Concepts Overview](https://mintlify.wiki/skydiscover-ai/skydiscover/concepts/overview.md): Understanding SkyDiscover's architecture, algorithms, and evaluation framework
- [LLM Configuration](https://mintlify.wiki/skydiscover-ai/skydiscover/config/llm.md): Configure language models, API settings, and generation parameters for SkyDiscover
- [Monitor Configuration](https://mintlify.wiki/skydiscover-ai/skydiscover/config/monitor.md): Configure the live monitoring dashboard for real-time search progress visualization
- [Configuration Overview](https://mintlify.wiki/skydiscover-ai/skydiscover/config/overview.md): Learn how to configure SkyDiscover using YAML files and understand the configuration hierarchy
- [Prompt Configuration](https://mintlify.wiki/skydiscover-ai/skydiscover/config/prompt.md): Configure system messages, prompt templates, and context building in SkyDiscover
- [Search Configuration](https://mintlify.wiki/skydiscover-ai/skydiscover/config/search.md): Configure evolutionary search algorithms and database settings in SkyDiscover
- [Algorithm Design Examples](https://mintlify.wiki/skydiscover-ai/skydiscover/examples/algorithm-design.md): Competitive programming problems from the Frontier-CS benchmark with 172 algorithmic challenges
- [Creating Custom Problems](https://mintlify.wiki/skydiscover-ai/skydiscover/examples/custom-problems.md): Learn how to create your own optimization benchmarks with custom evaluators and initial programs
- [Math Optimization Examples](https://mintlify.wiki/skydiscover-ai/skydiscover/examples/math-optimization.md): Mathematical optimization problems including circle packing, Heilbronn problems, and more
- [Examples Overview](https://mintlify.wiki/skydiscover-ai/skydiscover/examples/overview.md): Real-world examples of using SkyDiscover across different optimization domains
- [Systems Optimization Examples](https://mintlify.wiki/skydiscover-ai/skydiscover/examples/systems-optimization.md): Real-world systems benchmarks from the ADRS initiative including cloud scheduling, load balancing, and database optimization
- [Custom Context Builders](https://mintlify.wiki/skydiscover-ai/skydiscover/extending/context-builders.md): Customize how program state is turned into LLM prompts
- [Custom Search Algorithms](https://mintlify.wiki/skydiscover-ai/skydiscover/extending/custom-algorithms.md): Implement your own search strategies by extending ProgramDatabase and DiscoveryController
- [Custom Benchmarks](https://mintlify.wiki/skydiscover-ai/skydiscover/extending/custom-benchmarks.md): Add your own optimization tasks by writing an evaluator and optional seed program
- [Benchmarks](https://mintlify.wiki/skydiscover-ai/skydiscover/guides/benchmarks.md): 200+ optimization tasks spanning math, systems, algorithms, and reasoning
- [Configuration](https://mintlify.wiki/skydiscover-ai/skydiscover/guides/configuration.md): Comprehensive guide to SkyDiscover YAML configuration files
- [Model Providers](https://mintlify.wiki/skydiscover-ai/skydiscover/guides/model-providers.md): Configure OpenAI, Gemini, Anthropic, DeepSeek, and local models with SkyDiscover
- [Monitoring](https://mintlify.wiki/skydiscover-ai/skydiscover/guides/monitoring.md): Watch your discovery runs in real-time with the live dashboard
- [Running Discovery](https://mintlify.wiki/skydiscover-ai/skydiscover/guides/running-discovery.md): Learn how to run SkyDiscover to evolve solutions using the CLI and Python API
- [Writing Evaluators](https://mintlify.wiki/skydiscover-ai/skydiscover/guides/writing-evaluators.md): Create scoring functions that guide SkyDiscover toward better solutions
- [Installation](https://mintlify.wiki/skydiscover-ai/skydiscover/installation.md): Install SkyDiscover and configure your development environment
- [Introduction to SkyDiscover](https://mintlify.wiki/skydiscover-ai/skydiscover/introduction.md): A flexible framework for AI-driven scientific and algorithmic discovery
- [Quick Start](https://mintlify.wiki/skydiscover-ai/skydiscover/quickstart.md): Get started with SkyDiscover in under 5 minutes