# aludel v0.7.0 - Table of Contents

> Phoenix-native LLM evaluation, prompt testing, model comparison, red teaming, and observability for Elixir applications

## Pages

- [README](readme.md)

- Guides
  - [Feature Guide](features.md)
  - [Evaluation Guide](evaluations.md)
  - [Curated Red-Team Datasets](red_team.md)
  - [File-Based Evaluation Suites](file_suites.md)
  - [ExUnit Evaluations](ex_unit.md)
  - [Evaluation Reporters](reporters.md)
  - [Embedding and Deployment Guide](embedding.md)

- Project
  - [Changelog](changelog.md)
  - [Contributing to Aludel](contributing.md)
  - [LICENSE](license.md)

## Modules

- [Aludel](Aludel.md): Phoenix-native LLM evaluation workbench.
- [Aludel.Datasets](Aludel.Datasets.md): Context for reusable evaluation datasets and suite population.

- [Aludel.Datasets.Dataset](Aludel.Datasets.Dataset.md): A reusable collection of ordered evaluation examples.

- [Aludel.Datasets.DatasetEntry](Aludel.Datasets.DatasetEntry.md): An ordered single-turn or multi-turn example in a dataset.

- [Aludel.DocumentConverter](Aludel.DocumentConverter.md): Converts documents between formats for LLM consumption.
- [Aludel.Evals](Aludel.Evals.md): Context for managing evaluation suites, test cases, quality policies, and runs.

- [Aludel.Evals.AssertionEvaluator](Aludel.Evals.AssertionEvaluator.md): Backward-compatible facade for behaviour-driven evaluation metrics.

- [Aludel.Evals.AssertionParser](Aludel.Evals.AssertionParser.md): Parses and validates assertion payloads from suite editor forms.

- [Aludel.Evals.DocumentIngestion](Aludel.Evals.DocumentIngestion.md): Validates and persists uploaded documents for suite test cases.

- [Aludel.Evals.FileSuite](Aludel.Evals.FileSuite.md): Loads and executes versioned JSON or YAML suite manifests.
- [Aludel.Evals.JudgeCatalog](Aludel.Evals.JudgeCatalog.md): Versioned rubric templates for common model-backed evaluation tasks.
- [Aludel.Evals.Metric](Aludel.Evals.Metric.md): Contract and shared helpers for evaluation metrics.
- [Aludel.Evals.Metric.Context](Aludel.Evals.Metric.Context.md): Evidence available to an evaluation metric.
- [Aludel.Evals.Metric.Evaluator](Aludel.Evals.Metric.Evaluator.md): Normalized execution details for the evaluator behind a metric result.
- [Aludel.Evals.Metric.Registry](Aludel.Evals.Metric.Registry.md): Registry for the built-in evaluation metrics.
- [Aludel.Evals.Metric.Result](Aludel.Evals.Metric.Result.md): Normalized result returned by every evaluation metric.

- [Aludel.Evals.Metrics.RubricJudge](Aludel.Evals.Metrics.RubricJudge.md): Evaluates generated output against a user-defined rubric with an Aludel provider.
- [Aludel.Evals.QualityPolicy](Aludel.Evals.QualityPolicy.md): Validates and evaluates versioned suite quality policies.
- [Aludel.Evals.Report](Aludel.Evals.Report.md): A normalized, versioned representation of an evaluation suite run.
- [Aludel.Evals.Reporter](Aludel.Evals.Reporter.md): Renders evaluation reports for people and CI systems.
- [Aludel.Evals.Reporters.Console](Aludel.Evals.Reporters.Console.md): Renders a concise, plain-text evaluation summary.
- [Aludel.Evals.Reporters.GitHub](Aludel.Evals.Reporters.GitHub.md): Renders GitHub Actions workflow-command annotations.
- [Aludel.Evals.Reporters.JSON](Aludel.Evals.Reporters.JSON.md): Renders the versioned evaluation report as JSON.
- [Aludel.Evals.Reporters.JUnit](Aludel.Evals.Reporters.JUnit.md): Renders evaluation results as JUnit XML for CI test-report viewers.
- [Aludel.Evals.Sampling](Aludel.Evals.Sampling.md): Configuration and deterministic reduction for repeated test case execution.
- [Aludel.Evals.Suite](Aludel.Evals.Suite.md): Schema for evaluation test suites.
- [Aludel.Evals.SuitePolicy](Aludel.Evals.SuitePolicy.md): Immutable version of a quality policy attached to an evaluation suite.
- [Aludel.Evals.SuiteRun](Aludel.Evals.SuiteRun.md): Schema for tracking suite execution results.
- [Aludel.Evals.SuiteRunner](Aludel.Evals.SuiteRunner.md): Owns suite execution launch under explicit supervision.

- [Aludel.Evals.TestCase](Aludel.Evals.TestCase.md): Schema for individual test cases within a suite.
- [Aludel.Evals.TestCaseDocument](Aludel.Evals.TestCaseDocument.md): Schema for file attachments associated with test cases.
- [Aludel.Evals.TestCaseEditor](Aludel.Evals.TestCaseEditor.md): Coordinates suite test case editing workflows outside the web layer.

- [Aludel.Evals.TestCaseImporter](Aludel.Evals.TestCaseImporter.md): Imports suite test cases from CSV and JSON payloads.
- [Aludel.ExUnit](Aludel.ExUnit.md): Adds evaluation assertions to ExUnit tests.
- [Aludel.Execution](Aludel.Execution.md): Shared execution boundary for native provider calls and host-app callbacks.

- [Aludel.Executor](Aludel.Executor.md): Behaviour and configuration helpers for execution mode dispatch.

- [Aludel.FileValidation](Aludel.FileValidation.md): Validates file content against claimed MIME types using magic bytes.
- [Aludel.Interfaces.Adapters.Http](Aludel.Interfaces.Adapters.Http.md): Generic HTTP client adapter behaviour.

- [Aludel.Interfaces.DocumentConverter.Adapters.Imagemagick](Aludel.Interfaces.DocumentConverter.Adapters.Imagemagick.md): Converts the first page of a PDF to PNG with bounded ImageMagick resources.
- [Aludel.Interfaces.DocumentConverter.Behaviour](Aludel.Interfaces.DocumentConverter.Behaviour.md): Behaviour for document conversion adapters.
- [Aludel.Interfaces.LLM.Adapters.Http.Default](Aludel.Interfaces.LLM.Adapters.Http.Default.md): Default LLM HTTP client using ReqLLM.
- [Aludel.Interfaces.LLM.Behaviour](Aludel.Interfaces.LLM.Behaviour.md): Behaviour for LLM provider implementations.
- [Aludel.Interfaces.LLM.Config](Aludel.Interfaces.LLM.Config.md): Configuration utilities for LLM provider implementations.

- [Aludel.Interfaces.LLM.ErrorParser](Aludel.Interfaces.LLM.ErrorParser.md): Shared error parsing logic for LLM provider adapters.
- [Aludel.Interfaces.LLM.Providers.Anthropic](Aludel.Interfaces.LLM.Providers.Anthropic.md): Anthropic Claude API adapter implementation.
- [Aludel.Interfaces.LLM.Providers.Google](Aludel.Interfaces.LLM.Providers.Google.md): Google Gemini LLM provider implementation.
- [Aludel.Interfaces.LLM.Providers.Groq](Aludel.Interfaces.LLM.Providers.Groq.md): Groq LLM provider implementation through ReqLLM.

- [Aludel.Interfaces.LLM.Providers.Ollama](Aludel.Interfaces.LLM.Providers.Ollama.md): Ollama API adapter implementation.
- [Aludel.Interfaces.LLM.Providers.OpenAI](Aludel.Interfaces.LLM.Providers.OpenAI.md): OpenAI LLM provider implementation.
- [Aludel.Interfaces.LLM.Providers.OpenRouter](Aludel.Interfaces.LLM.Providers.OpenRouter.md): OpenRouter LLM provider implementation through ReqLLM.

- [Aludel.Interfaces.LLM.Providers.XAI](Aludel.Interfaces.LLM.Providers.XAI.md): xAI LLM provider implementation through ReqLLM.

- [Aludel.Interfaces.Storage.Adapters.AWS](Aludel.Interfaces.Storage.Adapters.AWS.md): AWS S3-backed document storage adapter.

- [Aludel.Interfaces.Storage.Adapters.AWS.Client](Aludel.Interfaces.Storage.Adapters.AWS.Client.md): Behaviour for the AWS storage adapter's S3 client boundary.

- [Aludel.Interfaces.Storage.Adapters.AWS.ExAwsClient](Aludel.Interfaces.Storage.Adapters.AWS.ExAwsClient.md): Default AWS client implementation backed by `ExAws.S3`.

- [Aludel.Interfaces.Storage.Adapters.AWS.HTTPClient](Aludel.Interfaces.Storage.Adapters.AWS.HTTPClient.md): ExAws HTTP client implementation backed by Req.
- [Aludel.Interfaces.Storage.Adapters.GCS](Aludel.Interfaces.Storage.Adapters.GCS.md): Google Cloud Storage-backed document storage adapter.

- [Aludel.Interfaces.Storage.Adapters.GCS.Client](Aludel.Interfaces.Storage.Adapters.GCS.Client.md): Behaviour for the GCS storage adapter's Google API client boundary.

- [Aludel.Interfaces.Storage.Adapters.GCS.GoogleApiClient](Aludel.Interfaces.Storage.Adapters.GCS.GoogleApiClient.md): Default GCS client implementation backed by `google_api_storage` and `Goth`.

- [Aludel.Interfaces.Storage.Adapters.Local](Aludel.Interfaces.Storage.Adapters.Local.md): Local filesystem adapter for document storage.

- [Aludel.Interfaces.Storage.Behaviour](Aludel.Interfaces.Storage.Behaviour.md): Behaviour for external document storage adapters.

- [Aludel.LLM](Aludel.LLM.md): LLM client abstraction for multi-provider support.
- [Aludel.Projects](Aludel.Projects.md): Context for managing projects that organize prompts and test suites.

- [Aludel.Projects.Project](Aludel.Projects.Project.md): Schema for organizing prompts and test suites into projects.

- [Aludel.Prompts](Aludel.Prompts.md): Context for managing prompts and their versions.

- [Aludel.Prompts.Evolution](Aludel.Prompts.Evolution.md): Functions for analyzing prompt version evolution and performance metrics.

- [Aludel.Prompts.Evolution.Export](Aludel.Prompts.Evolution.Export.md): Export serialization for prompt evolution metrics.

- [Aludel.Prompts.Optimization](Aludel.Prompts.Optimization.md): Multi-objective prompt analysis and failure-grounded prompt suggestions.
- [Aludel.Prompts.Prompt](Aludel.Prompts.Prompt.md): Schema for managing prompts.
- [Aludel.Prompts.PromptSuggestion](Aludel.Prompts.PromptSuggestion.md): A failure-grounded prompt revision awaiting an explicit human decision.

- [Aludel.Prompts.PromptVersion](Aludel.Prompts.PromptVersion.md): Schema for prompt versions.
- [Aludel.Providers](Aludel.Providers.md): Context for managing AI provider configurations.

- [Aludel.Providers.Pricing](Aludel.Providers.Pricing.md): Resolves pricing for a given provider and model.
- [Aludel.Providers.Provider](Aludel.Providers.Provider.md): Schema for AI provider configurations.

- [Aludel.RedTeam](Aludel.RedTeam.md): Curated and generated adversarial evaluation cases.
- [Aludel.RedTeam.Catalog](Aludel.RedTeam.Catalog.md): Stable, versioned adversarial cases used by `Aludel.RedTeam`.
- [Aludel.RedTeam.GeneratedCase](Aludel.RedTeam.GeneratedCase.md): A validated, reviewable adversarial case proposed by a generator model.
- [Aludel.RedTeam.Generation](Aludel.RedTeam.Generation.md): The review boundary returned by `Aludel.RedTeam.generate/2`.
- [Aludel.Repo](Aludel.Repo.md): Central repository accessor for Aludel.
- [Aludel.Runs](Aludel.Runs.md): Context for managing runs and run results.

- [Aludel.Runs.Execution](Aludel.Runs.Execution.md): Result of executing a run across one or more providers.

- [Aludel.Runs.Executor](Aludel.Runs.Executor.md): Owns run launch and provider execution under explicit supervision.

- [Aludel.Runs.Run](Aludel.Runs.Run.md): Schema for managing runs.
- [Aludel.Runs.RunResult](Aludel.Runs.RunResult.md): Schema for run results.
- [Aludel.Stats.Activity](Aludel.Stats.Activity.md): Recent activity feeds and time-series activity reporting.

- [Aludel.Stats.Costs](Aludel.Stats.Costs.md): Cost reporting and breakdowns for the dashboard.

- [Aludel.Stats.Latency](Aludel.Stats.Latency.md): Latency reporting and provider breakdowns.

- [Aludel.Stats.Overview](Aludel.Stats.Overview.md): Top-line dashboard metrics and bounded period comparisons.

- [Aludel.Storage](Aludel.Storage.md): Domain-facing facade for document storage backends.
- [Aludel.Storage.Config](Aludel.Storage.Config.md): Resolves storage configuration without coupling adapters to the storage facade.
- [Aludel.Web.Assets](Aludel.Web.Assets.md): Serves pre-compiled static assets for Aludel dashboard.
- [Aludel.Web.Authentication](Aludel.Web.Authentication.md): Loads dashboard session context and enforces LiveView access decisions.

- [Aludel.Web.CoreComponents](Aludel.Web.CoreComponents.md): Provides core UI components.
- [Aludel.Web.Hooks](Aludel.Web.Hooks.md): LiveView hooks for common functionality across all LiveViews.

- [Aludel.Web.Resolver](Aludel.Web.Resolver.md): Behavior for customizing dashboard access and user identification.
- [Aludel.Web.Router](Aludel.Web.Router.md): Provides the aludel_dashboard macro for mounting Aludel in host apps.

## Mix Tasks

- [mix aludel.eval](Mix.Tasks.Aludel.Eval.md): Runs one Aludel evaluation suite and emits a report.
- [mix aludel.install](Mix.Tasks.Aludel.Install.md): Installs Aludel by copying migrations to the host application.
- [mix aludel.seed](Mix.Tasks.Aludel.Seed.md): Runs Aludel seed data.

