Glossary
Common terms in ChooChoo and Context Governance
Key terms and concepts used throughout the ChooChoo documentation. Each term links to the relevant section for a deeper understanding.
| Term | Definition |
|---|---|
| Context Graph | A living, queryable map of decision traces stitched across entities and time. Visualized in The Station and queryable via choochoo lineage. |
| Decision Trace | A structured record linking an Artifact to the Agent/Human that created it, the Policy that authorized it, and the Reasoning behind it. Stored in the Audit Trail. |
| Context Amnesia | The loss of reasoning and institutional memory that occurs when systems capture only the result of a decision but discard the "why". This is the core problem ChooChoo solves — see Why ChooChoo?. |
| Agent Rails | Constraints and boundaries that define what an agent is permitted to do. Enforced by the validation engine and recorded in the Audit Trail. |
| The Engine | The ChooChoo CLI and Validation Layer. Runs locally and in CI/CD pipelines. See the CLI Reference for commands. |
| The Map | The accumulated Decision Traces stored in the Context Graph. Powers impact analysis and risk scoring. |
| The Station | The Governance UI for GRC teams. Provides lineage visualization, risk heatmaps, audit search, and compliance dashboards. |
| Product | A logical unit of data or capability defined using the ODPS standard. The atomic unit of ownership and accountability. |
| Contract | A schema and quality agreement defined using the ODCS standard. Enforces data quality, semantics, and compliance tags. |
| Workflow | A multi-step API orchestration defined using the Arazzo standard. Includes governance gates for approval. |
| System Card | An identity document for an AI system. Documents capabilities, limitations, and compliance alignment. Referenced during risk scoring. |
| Risk Score | A value from 0.0 to 10.0 calculated for every proposed change. Determines whether an agent can auto-proceed or needs human approval. |
| Impact Radius | The count and criticality of downstream artifacts affected by a change. Computed by traversing the Lineage Graph and used in risk scoring. |
| Compliance Tags | Labels applied to contract fields (e.g., pii, gdpr-article-9) that trigger governance workflows and affect risk scoring. |
| Approval Workflow | A policy gate that requires human sign-off before a change can proceed. Triggered by compliance tags and risk scores. |
Architecture Layers
For a visual overview of how these concepts relate to each other, see the Architecture page and the interactive graph at the top of that page.
Architecture
See how The Engine, The Map, and The Station work together in the platform.
Standards Reference
ODPS, ODCS, Arazzo, and Agent Trace — the open standards behind the terminology.
Getting Started
Put these concepts into practice by initializing your first project.
Core Concepts
Deep dive into Products, Contracts, Workflows, Agents, and System Cards.