artifact-pyramids
Artifact Pyramids for Agentic AI Research
Progressive disclosure governs how we feed agents context: metadata at startup, instructions on activation, resources on demand. The Artifact Pyramid applies the same principle to what agents produce. Three layers of increasing depth, each independently consumable, each linking down to the next.
Loading Guidance
When producing an artifact pyramid, load these references as a required set (not piecemeal):
| Reference | File |
|---|---|
| Pipeline Stages — layer definitions, navigation format, production flow | references/pipeline-stages.md |
| Output Classification Framework — role-agnostic content contracts, 00-index/L1 boundary | references/output-classification-framework.md |
| Quality Gates — verification checklists per layer | references/quality-gates.md |
| Delegation Context Template — exact text for subagent output mandates | references/delegation-context-template.md |
These four define complementary aspects of the spec that the others assume. Loading only a subset risks violating content contracts (e.g., putting findings in 00-index) or skipping required navigation affordances. Load the full set before writing any pyramid files.
The reference table below describes when to load each file; the four above are always required for pyramid production. The remaining references (framework, worked example, canonical article, intellectual lineage, provenance, composite synthesis) are supplementary — load when the task calls for conceptual depth or a worked pattern.