Cognition · Data artifact

Decomposition Method Library

Data artifactCognitionCognition & Memoryarc:DecompositionMethodLibrary

A curated set of decomposition methods, each naming an abstract task, its applicability preconditions, the replacing subtask network, and ordering, resource and causal constraints.

Responsibility. Encodes domain knowledge on how abstract tasks decompose into subtask networks.

Also known as: HTN method library, Methods, Task decomposition knowledge base

Direct neighbourhood (hover for relationship types)

Relationships

configures structural

HTN Planner Ch5.4

is read by dependency

Design guidance

SHOULD encode meaningfully different approaches as separate methods and parameter variations as method parameters.
MUST model subtask interdependencies (ordering, shared resources, causal links, temporal constraints) explicitly.
SHOULD use the weakest ordering constraints sufficient for correctness to maximise execution flexibility.
SHOULD avoid overlapping subtasks so each primitive appears once unless duplication is intentional.
SHOULD organise methods hierarchically (strategy, tactical, operational) mirroring the task hierarchy.

Classification

Patterns: Method multiplicityParameterized methodsHierarchically organised method librariesPartial ordering (weakest sufficient constraints)
Quality attributes: Interaction capability (ISO/IEC 25010)Maintainability (ISO/IEC 25010)
Risks mitigated: Method proliferationIgnored decomposition interdependenciesRedundant subtasks across methods

Sources

Ch5.4: T. Nguyen, "Hierarchical Planning Fundamentals," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.4. ISBN: 9798244538229.