Skip to content

API Reference

Complete API documentation for COSMIC.

Core Classes

Class Description
COSMICChunker Main chunking orchestrator
Document Input document representation
COSMICChunk Output chunk with metadata
COSMICConfig Configuration system

Quick Reference

Basic Usage

from cosmic import COSMICChunker, Document

# Create chunker
chunker = COSMICChunker()

# Create document
doc = Document.from_text("Your text here...")

# Chunk document
chunks = chunker.chunk_document(doc, strategy="auto")

With Configuration

from cosmic import COSMICChunker, COSMICConfig
from cosmic.core.config import DCSConfig

config = COSMICConfig(
    dcs=DCSConfig(alpha=0.5, beta=0.3, gamma=0.2)
)
chunker = COSMICChunker(config=config)

Modules

cosmic

Main package exports:

from cosmic import (
    COSMICChunker,
    COSMICChunk,
    COSMICConfig,
    Document,
    ProcessingMode,
    Intent,
    DomainType,
)

cosmic.core

Core data structures:

  • cosmic.core.chunk - COSMICChunk dataclass
  • cosmic.core.config - Configuration classes
  • cosmic.core.document - Document class
  • cosmic.core.enums - Enumerations

cosmic.pipeline

Pipeline stages:

  • cosmic.pipeline.structure - Stage 1: Structure Analysis
  • cosmic.pipeline.semantic - Stage 2: Semantic Boundaries
  • cosmic.pipeline.domain - Stage 3: Domain Classification
  • cosmic.pipeline.fusion - Stage 4: Boundary Fusion
  • cosmic.pipeline.verification - Stage 5: LLM Verification
  • cosmic.pipeline.reference - Stage 6: Reference Linking

cosmic.scoring

Scoring algorithms:

  • cosmic.scoring.dcs - Discourse Coherence Scorer
  • cosmic.scoring.clustering - MST Clustering

cosmic.models

ML model wrappers:

  • cosmic.models.embeddings - Sentence-transformers
  • cosmic.models.llm - LLM client
  • cosmic.models.ollama - Ollama integration
  • cosmic.models.coreference - spaCy coreference

cosmic.fallback

Fallback strategies:

  • cosmic.fallback.semantic_only - Semantic-only chunking
  • cosmic.fallback.sliding_window - Sliding window
  • cosmic.fallback.fixed_length - Fixed-length splitting

Enumerations

ProcessingMode

from cosmic import ProcessingMode

ProcessingMode.FULL_COSMIC      # Full 6-stage pipeline
ProcessingMode.SEMANTIC_ONLY    # DCS-based boundaries
ProcessingMode.SLIDING_WINDOW   # Similarity-based
ProcessingMode.FIXED_LENGTH     # Token-based

Intent

from cosmic import Intent

Intent.DEFINE      # Defines a concept
Intent.EXPLAIN     # Explains how something works
Intent.LIST        # Enumerates items
Intent.ARGUE       # Makes an argument
Intent.DESCRIBE    # Describes characteristics
Intent.INSTRUCT    # Provides instructions
Intent.SUMMARIZE   # Summarizes information
Intent.COMPARE     # Compares alternatives
Intent.NARRATE     # Tells a story

DomainType

from cosmic import DomainType

DomainType.TECHNICAL
DomainType.MEDICAL
DomainType.LEGAL
DomainType.SCIENTIFIC
DomainType.GENERAL

Exceptions

from cosmic.exceptions import (
    StructureAnalysisError,    # E001
    SemanticBoundaryError,     # E002
    LLMVerificationError,      # E003
    DomainClassificationError, # E004
    ReferenceResolutionError,  # E005
    ConfigurationError,        # E006
)