Defined Concepts
Concepts are the theoretical foundations — immutable ideas, taxonomies, and definitions that underpin the Agentic SDLC. They describe the "physics" of agent-driven software development.
Start Here
Three landmark concepts that define the domain.
Agentic SDLC
LiveComplete framework for industrializing software development with AI agents. Defines how agents serve as the logistic layer while humans design, govern, and optimize the production flow.
Spec-Driven Development
LiveMethodology that defines specifications before implementation, treating specs as living authorities that code must fulfill.
Levels of Autonomy
LiveSAE-inspired taxonomy classifying AI agent autonomy from L1 assistive to L5 fully autonomous. Learn why L3 conditional autonomy is the current production ceiling and what each level means.
Live Concepts
- Architecture Decision Record A lightweight document that captures a significant architectural decision, its context, and consequences at a specific point in time.
- Behavior-Driven Development A collaborative specification methodology that defines system behavior in natural language scenarios, bridging business intent and machine-verifiable acceptance criteria.
- Context Anchoring The phenomenon where explicit context biases an LLM toward specific concepts or solutions, even when marked as deprecated or irrelevant to the immediate task.
- Context Engineering Context Engineering is the practice of structuring information to optimize LLM comprehension and output quality.
- Extreme Programming A software development methodology emphasizing high-frequency feedback, testing, and continuous refactoring, which maps perfectly to the Agentic SDLC.
- Gherkin A structured, domain-specific language using Given-When-Then syntax to define behavioral specifications that are both human-readable and machine-actionable.
- Mermaid A text-based diagramming language that renders flowcharts, sequences, and architectures from markdown, enabling version-controlled visual specifications.
- Model Context Protocol (MCP) Open standard for connecting AI agents to external tools and data. Covers MCP's role in agentic development, its limits, and why implementation quality matters.
- Model-Driven Development An early 2000s software engineering paradigm that attempted 100% code generation from models, serving as a cautionary tale for modern spec-as-source AI hype.
- OODA Loop The OODA Loop (Observe-Orient-Decide-Act) is a decision-making framework by John Boyd. Learn how it applies to software development and autonomous AI agents.
- Provenance The chain of custody and intent behind software artifacts, distinguishing high-value engineered systems from 'slop'.
- Request for Comments A collaborative proposal document for significant changes that require team consensus before becoming formal decisions.
- The 4D Framework (Anthropic) A cognitive model codifying four essential competencies—Delegation, Description, Discernment, and Diligence—for effective generative AI use.
- The Learning Loop The iterative cycle between exploratory implementation and spec refinement, balancing vibe coding velocity with captured learnings.
- Triple Debt Model A diagnostic framework for software system health built around three interacting debt types that accumulate as AI accelerates development.
- Vibe Coding Natural language code generation without formal specs—powerful for prototyping, problematic for production systems.
- YAML A human-readable data serialization language that serves as the structured specification format for configuration, schemas, and file structures in agentic workflows.
Experimental & Draft
- Agent Skills Open standard for packaging procedural knowledge and workflows, serving as the 'how-to' layer for AI agents.
- AI Amplification The principle that AI tools amplify existing engineering practices—making disciplined teams faster and chaotic teams fail sooner.
- AI Software Factory An industrial-scale approach to software engineering. Explores the dichotomy between Safe ASDLC Factories (L3) and high-risk Dark Factories (L4).
- Coverage Metric Coverage measures task completion reliability: the proportion of required work units an agent successfully completes in a long-horizon task.
- Digital Twins Virtual replicas of complex systems. In software engineering, these are behavioral clones of third-party services used for high-volume scenario testing.
- Event Modeling A system blueprinting method that centers on events as the primary source of truth, serving as a rigorous bridge between visual design and technical implementation.
- Feedback Loop Compression How AI compresses the observe → validate → learn cycle, shifting the bottleneck from code production to code understanding.
- PR Slop PR slop is the flood of AI-generated code that passes automated checks but overwhelms human review capacity. It looks correct, compiles clean, and hides architectural drift.
- Product Requirement Prompt (PRP) A structured methodology combining PRD, codebase context, and agent runbook—the minimum spec for production-ready AI code.
- Product Thinking The practice of engineers thinking about user outcomes, business context, and the 'why' before the 'how'—the core human skill in the AI era.
- Production Readiness Gap The distance between a working generative AI demo and a secure, scalable production system.
- Software Civil Engineering The engineering discipline required for agentic software production, defined by formal specification layers, declarative lifecycles, and adversarial verification.
Deprecated
Next Steps
- Patterns — See how concepts become architectural designs.
- Getting Started — Deploy the methodology from scratch.
- Resources — Field manual, design system, and governance.