Skip to content

Add ReadTheDocs Documentation Site #8

Description

@SameerSamji

Background

SciKGExtract currently has no hosted documentation. There is no structured reference site covering architecture, configuration, API reference, or tutorials. A ReadTheDocs site is needed so that external users can understand and adopt the framework without reading source code directly.

Proposed Work

4.1 Documentation Sections to Write

Section Source Material
Getting Started Expand root README.md → install, .env setup, first extraction
Architecture Overview Orchestrator/agents diagram of the LangGraph workflow
Configuration Guide OrchestratorConfig, WorkflowConfig, ExtractionConfig, etc.
LLM Provider Setup OpenAI / SAIA / Ollama / HuggingFace setup, API key config
Custom Schemas How to define JSON schemas, configure ProcessConfig
Normalization PubChem LMDB setup, NormalizationConfig JSONPaths
Evaluation Guide LLM-as-Judge, quantitative metrics, PubChem normalized evaluation
API Reference Auto-generated via sphinx-autoapi from scikg_extract/
Tutorials Notebooks rendered as docs via nbsphinx
Contributing Expand CONTRIBUTING.md

4.2 Docstring Pass

Add docstrings to all public functions in:

  • scikg_extract/evaluation/
  • scikg_extract/tools/
  • scikg_extract/agents/
    ...

These are the user-facing modules that sphinx-autoapi will render into the API Reference section.

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Fields

No fields configured for issues without a type.

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions