Version: 2.0.0
Last Updated: 2026-02-10
Purpose: Centralized definitions of common terminology used across agent profiles, directives, and documentation.
This glossary provides standardized definitions for terms used throughout the SDD Agent Framework. These definitions ensure consistent understanding and communication across all agents, directives, and collaborative workflows.
For context on how to use this framework, see:
- AGENTS.md - Core agent specification document
- directives/ - Extended directive set
- Agent profiles (e.g., curator.agent.md, synthesizer.agent.md)
Reasoning mode focusing on systematic decomposition, structural analysis, and diagnostic evaluation; default mode for most agents
Context: Agent Modes - Reasoning
Source: All agent profiles
Related: /creative-mode, /meta-mode, Mode Protocol
Reasoning mode focusing on option generation, pattern shaping, alternative exploration, and narrative construction
Context: Agent Modes - Reasoning
Source: All agent profiles
Related: /analysis-mode, /meta-mode, Mode Protocol
Reasoning mode focusing on process reflection, alignment validation, governance review, and methodology evaluation
Context: Agent Modes - Reasoning
Source: All agent profiles
Related: /analysis-mode, /creative-mode, Mode Protocol
Complete workflow sequence through all six phases of Spec-Driven Development from analysis through review with explicit phase authority and handoff protocols
Context: Workflows - Spec-Driven Development
Source: analyst-annie.agent.md, architect.agent.md, project-planner.agent.md
Related: Spec-Driven Development, Phase Authority, Hand-off Protocol, Phase Checkpoint Protocol
Explicit specification of when a system refuses, constrains, or safely degrades rather than accepting input
Context: System behavior specification
Source: ATDD_adversarial-acceptance.tactic.md
Related: Adversarial Acceptance Testing, Contract Decision
Explicit, testable conditions that must be satisfied for a feature or requirement to be considered complete and acceptable
Context: Testing - Requirements
Source: analyst-annie.agent.md, python-pedro.agent.md
Related: ATDD, Phase 1 (Analysis), Executable Test, Specification
Quantitative metric (percentage) measuring how accurately tests document system across three dimensions: Behavioral (WHAT), Architectural (WHY), Operational (HOW runs)
Context: Quality Validation / Metrics
Source: reverse-speccing.md
Related: Reverse Speccing, Behavioral Accuracy, Architectural Accuracy
Percentage of domain terms in code/docs that use glossary terminology; target >80% for effective ubiquitous language
Context: Architecture - DDD / Metrics
Source: living-glossary-practice.md
Related: Living Glossary, Ubiquitous Language
Systematic process for creating Architecture Decision Records with trade-off analysis and risk assessment
Context: Decision documentation
Source: adr-drafting-workflow.tactic.md
Related: Trade-off Analysis, Option Impact Matrix
ATDD practice of deliberately exploring failure scenarios to define acceptance boundaries for misuse, edge cases, and unsafe defaults
Context: Testing methodology
Source: ATDD_adversarial-acceptance.tactic.md
Related: Acceptance Boundary, Adversarial Testing, Contract Decision
Stress-testing technique that deliberately attempts to make a proposal, design, or decision fail to surface weaknesses and blind spots
Context: Decision validation
Source: adversarial-testing.tactic.md
Related: Failure Scenario, Premortem Risk Identification
Autonomous operation protocol where agents work independently when human is away from keyboard, with clear decision boundaries and commit frequency expectations
Context: Agent autonomy
Source: autonomous-operation-protocol.tactic.md
Related: Decision Boundary, Commit Checkpoint, Escalation Protocol
Process of selecting and routing specific tasks to the most appropriate specialized agent based on task type, required capabilities, and phase authority
Context: Orchestration - Task Routing
Source: manager.agent.md
Related: Manager Mike, Hand-off Protocol, Phase Authority, AGENT_STATUS
A specialized configuration file (
.agent.md) that defines an agent's purpose, specialization, collaboration contract, mode defaults, and directive usage. Profiles extend the base AGENTS.md specification with role-specific competencies.
Location: agent profile files
Reference: Directive 005
Related: Specialization, Collaboration Contract
Documentation in agent profiles of common handoff patterns observed in practice (outgoing, incoming, special cases) providing guidance without prescriptive rules
Context: Orchestration / Agent Coordination
Source: agent-profile-handoff-patterns.md
Related: Handoff Pattern, Organic Emergence
Coordination artifact maintained by Manager Mike documenting which agent did what, when, and current workflow state
Context: Artifacts - Orchestration Source: manager.agent.md Related: Manager Mike, WORKFLOW_LOG, HANDOFFS, Agent Assignment
Parent-child relationship where specialized agents refine their parent's scope to narrower contexts. Orchestrator prefers specialists when context matches, falls back to parent when specialist unavailable or overloaded.
Context: Agent Collaboration - Orchestration Source: DDR-011, agent-specialization-hierarchy.md Related: Specialization Context, Routing Priority, SELECT_APPROPRIATE_AGENT, Parent Agent, Child Agent
Capability shift where AI agents make previously labor-intensive practices tractable (continuous capture, multi-source pattern detection, incremental maintenance)
Context: Meta-Pattern / Technology Impact
Source: language-first-architecture.md, living-glossary-practice.md
Related: Feasibility Shift, Continuous Capture
Decision-making framework evaluating practices across seven dimensions: Agile, Minimal, Maintainable, Environmental, Reachable, Solvable, Extensible
Context: Trade-off analysis
Source: ammerse-analysis.tactic.md
Related: Trade-off Analysis, Contextual Fit
Seven evaluation criteria for practices: Agile, Minimal, Maintainable, Environmental, Reachable, Solvable, Extensible
Context: Decision framework
Source: ammerse-analysis.tactic.md
Related: AMMERSE Analysis
Endless validation or research without delivery due to perfectionism or risk aversion
Context: Process anti-pattern
Source: requirements-validation-workflow.tactic.md
Related: Time-boxing, Evidence-Based Requirements
Requirements and validation specialist agent focused on producing testable, data-backed specifications with validated acceptance criteria
Context: Agent Roles - Requirements Analysis
Source: analyst-annie.agent.md
Related: Requirements Specialist, Specification-Driven Development, Phase 1 (Analysis), Validation Script
Systematic extraction of behaviors to actively discourage, common traps, or misapplications of guidance from operational logs
Context: Framework Improvement / Quality Assurance
Source: meta-analysis.md
Related: Meta-Analysis, Pattern Detection
Explicit specification of API endpoints, request/response formats, error handling, versioning, and behavioral guarantees
Context: Backend Architecture - Interface Definition
Source: backend-dev.agent.md
Related: Backend Benny, Service Design, Integration Surface
Architecture specialist agent who clarifies complex systems with contextual trade-offs, creates ADRs, and provides system decomposition with explicit decision rationale
Context: Agent Roles - Architecture
Source: architect.agent.md
Related: Architecture Specialist, ADR, System Decomposition, Phase 2 (Architecture), Trade-off Analysis
Aspect of system design (rationale, deployment, operations, security) that tests naturally miss and must be documented elsewhere (typically ADRs)
Context: Quality Validation / Documentation Gap
Source: reverse-speccing.md, test-readability-clarity-check.md
Related: Reverse Speccing, Test-as-Documentation
Agent specialization focused on system decomposition, design interfaces, explicit decision records (ADRs), and trade-off analysis for complex systems
Context: Agent Specializations
Source: architect.agent.md
Related: Architect Alphonso, ADR, System Decomposition, Trade-off Analysis
Structured document produced by Framework Guardian comparing installed framework state against canonical manifest, classifying files as MISSING, UNCHANGED, DIVERGED, or CUSTOM
Context: Artifacts - Framework Maintenance
Source: framework-guardian.agent.md
Related: Framework Guardian, Manifest, Drift Detection, Framework Integrity
Distinctive writing style, tone, rhythm, and personality of the original author that must be preserved during editing, translation, or curation
Context: Content Quality - Voice Preservation
Source: lexical.agent.md, translator.agent.md, writer-editor.agent.md
Related: Voice Fidelity, Tonal Integrity, Tone Preservation
Backend developer specialist agent focused on resilient service backends, integration surfaces, API design, and persistence strategy with traceable decisions
Context: Agent Roles - Backend Development
Source: backend-dev.agent.md
Related: Backend Developer Specialist, Service Design, API Contract, Persistence Strategy, Integration Surface
Agent specialization focused on API/service design, persistence strategy, performance budgets, failure-mode mapping, and backend architecture
Context: Agent Specializations
Source: backend-dev.agent.md
Related: Backend Benny, Service Design, Persistence Strategy, Performance Budget
"Do nothing" alternative that must be evaluated in trade-off analysis; current state may be acceptable, or organic emergence may solve issue naturally
Context: Development Methodology / Decision Framework
Source: locality-of-change.md
Related: Problem Assessment Protocol, Organic Emergence
Planning approach that breaks work into small, time-boxed batches (typically 1-2 weeks) rather than fixed-date commitments, enabling adaptive execution
Context: Planning - Execution Strategy
Source: project-planner.agent.md
Related: Planning Petra, NEXT_BATCH, Milestone Definition
Behavior specification in Given/When/Then format describing observable outcomes in business-relevant language
Context: Behavior-driven development
Source: development-bdd.tactic.md
Related: Given-When-Then, Executable Specification
Practice of maintaining links both forward (requirements → tests → code) and backward (code → tests → requirements) to enable navigation in both directions
Context: Development Methodology / Documentation
Source: traceability-chain-pattern.md
Related: Traceability Chain, Forward Link, Backward Link
Opinionated Python code formatter ensuring consistent code style with PEP 8 compliance
Context: Python Development - Code Formatting
Source: python-pedro.agent.md
Related: Python Pedro, ruff, Code Quality
Scope of potential impact from experiment failure, deliberately contained through boundaries
Context: Risk containment
Source: safe-to-fail-experiment-design.tactic.md
Related: Experiment Boundary, Safe-to-Fail Experiment
Repository scaffolding specialist agent who maps repository topology, generates structural artifacts (REPO_MAP, SURFACES, WORKFLOWS), and creates doctrine configuration for efficient multi-agent collaboration
Context: Agent Roles - Repository Initialization
Source: bootstrap-bill.agent.md
Related: Repository Scaffolding, Doctrine Configuration, REPO_MAP, Topology Mapping, Structural Artifact
Indicator of context boundaries: event clusters that separate naturally, different actors, vocabulary divergence, or team responsibility shifts
Context: Context boundary detection
Source: event-storming-discovery.tactic.md
Related: Event Cluster, Context Boundary
Technique for identifying hidden context boundaries by analyzing terminology patterns, communication structures, and semantic conflicts in existing systems
Context: Architecture - DDD
Source: bounded-context-linguistic-discovery.md
Related: Bounded Context, Semantic Conflict, Conway's Law Applied
Automated alert when feature branch exceeds age threshold (8h warning, 24h maximum) to enforce short-lived branch discipline
Context: Version Control / Automation
Source: trunk-based-development.md
Related: Short-Lived Branch, Trunk-Based Development
Agent specialization focused on build graph modeling, CI/CD flow design, caching strategy, dependency integrity, and reproducible pipelines
Context: Agent Specializations
Source: build-automation.agent.md
Related: DevOps Danny, CI/CD Pipeline, Reproducible Build, Build Graph
Directed acyclic graph representing build dependencies, execution order, caching boundaries, and artifact relationships in CI/CD pipeline
Context: Build Automation - Dependency Management
Source: build-automation.agent.md
Related: DevOps Danny, CI/CD Pipeline, Reproducible Build
Pre-execution verification that checks for LOCAL_ENV.md file containing environment-specific constraints or instructions before proceeding with standard directives
Context: Framework - Environment Configuration
Source: Directive 001 (CLI and Shell Tooling)
Related: LOCAL_ENV.md, Remediation Technique
Lightweight hierarchical technique for structuring software architecture diagrams: Context (system+users), Container (deployable units), Component (internal modules), Code (implementation)
Context: Architecture / Diagramming
Source: design_diagramming-incremental_detail.md
Related: Incremental Detail Design, Progressive Disclosure
Explicit definition of agent expertise boundaries - what they can and cannot do
Context: Agent specialization
Source: agent-profile-creation.tactic.md
Related: Agent Profile, Specialization
Ignoring claims that contradict preferences, undermining intellectual honesty
Context: Research anti-pattern
Source: requirements-validation-workflow.tactic.md
Related: Claim Inventory, Confirmation Bias
System for categorizing requirement claims by evidence type: Empirical (quantitative data), Observational (qualitative patterns), Theoretical (conceptual models), Prescriptive (best practices)
Context: Requirements Engineering
Source: evidence-based-requirements.md
Related: Evidence-Based Requirements, Testability Assessment
Systematic catalog of verifiable assertions from research sources with evidence classification enabling testable requirements
Context: Evidence-based requirements
Source: claim-inventory-development.tactic.md
Related: Evidence Type, Testability Assessment, Traceability Chain
Connections between claims: Supporting, Contradicting, Prerequisite, or Alternative explanations
Context: Knowledge modeling
Source: claim-inventory-development.tactic.md
Related: Claim Inventory, Claim Map
Technique for clustering terminology by analyzing which terms appear together in the same files or contexts
Context: Linguistic analysis Source: code-documentation-analysis.tactic.md Related: Semantic Clustering, Terminology Extraction
Agent that inherits parent's collaboration contract but operates in narrower specialization context (language, framework, domain, writing style). Declared via specializes_from metadata in agent profile frontmatter.
Context: Agent Specialization Hierarchy Source: DDR-011 Related: Parent Agent, Specialization Context, Agent Specialization Hierarchy
Review specialist agent focused on code quality, standards compliance, and traceability validation without making direct code modifications
Context: Agent Roles - Code Review
Source: code-reviewer-cindy.agent.md
Related: Review Specialist, Code Quality, Standards Compliance, Traceability Check
Implicit constraint that humans can internalize ~50-100 precise domain terms effectively; bounded contexts manage this complexity
Context: Architecture - DDD / Cognitive Science
Source: bounded-context-linguistic-discovery.md
Related: Bounded Context, Vocabulary Fragmentation
A section within each agent profile that specifies behavioral commitments, boundaries, escalation protocols, and interaction patterns with other agents and humans. Defines what the agent will and won't do.
Related: Agent Profile, Escalation
Regular commit cadence (every 15-30 minutes) in autonomous work to create reversible progress points
Context: Version control practice
Source: autonomous-operation-protocol.tactic.md
Related: AFK Mode, Self-Observation Checkpoint
Classification of team interaction cadence: Daily, Weekly, Monthly, or Rarely
Context: Organizational patterns
Source: team-interaction-mapping.tactic.md
Related: Team Interaction Mapping, Shared Artifact
Cumulative effect where each small addition seems reasonable in isolation but together degrades system simplicity, making it harder to understand and maintain
Context: Development Methodology / Anti-Pattern
Source: locality-of-change.md
Related: Gold Plating, Premature Abstraction, Simplicity Preservation
Reusable UI component designs with established structure, behavior, styling, and composition rules for consistent interface development
Context: Frontend Architecture - Reusability
Source: frontend.agent.md
Related: Frontend Freddy, UI Architecture, Design System
Refactoring that makes implicit or duplicated logic explicit by extracting it into a named abstraction
Context: Code improvement
Source: refactoring-extract-first-order-concept.tactic.md
Related: First-Order Concept, Rule of Three
Minimum confidence level required before acting on automated detection results to balance automation benefits against false positive risks
Context: Quality Assurance / Automation
Source: living-glossary-practice.md
Related: False Positives, Human Review Loop
Process of categorizing upgrade conflicts as auto-merge candidates, local customizations to preserve, or breaking changes requiring manual review
Context: Agent Capabilities - Framework Maintenance
Source: framework-guardian.agent.md
Related: Framework Guardian, Upgrade Plan, Core/Local Boundary
Time from linguistic conflict detection to architectural issue manifestation; target <2 weeks (50% improvement over baseline)
Context: Architecture - DDD / Metrics
Source: language-first-architecture.md
Related: Linguistic Signal, Semantic Conflict
Metric measuring days from semantic conflict detection to resolution; target <7 days to prevent architectural drift
Context: Architecture - DDD / Metrics
Source: language-first-architecture.md
Related: Semantic Conflict, Linguistic Signal
Alternative explanation for observed results that must be controlled in validation experiments
Context: Experimental design
Source: requirements-validation-workflow.tactic.md
Related: Validation Experiment, Multivariate Analysis
Systematic detection of bounded context boundaries from team structure and terminology conflicts using Conway's Law principles
Context: DDD architectural analysis
Source: context-boundary-inference.tactic.md
Related: Conway's Law, Vocabulary Ownership, Semantic Boundary
Undesired influence of previous iteration results on subsequent runs, introducing bias
Context: Execution quality issue
Source: execution-fresh-context-iteration.tactic.md
Related: Fresh Context Iteration
Unacknowledged change in task goals, constraints, or assumptions during execution leading to scope creep
Context: Scope violation
Source: context-establish-and-freeze.tactic.md
Related: Frozen Context, Scope Creep
Practice of defining relationships between bounded contexts (Upstream/Downstream, Shared Kernel, Published Language, Conformist) with explicit terminology translation rules
Context: Architecture - DDD
Source: bounded-context-linguistic-discovery.md
Related: Bounded Context, Translation Layer
Degree to which a solution aligns with organizational culture, constraints, and environmental factors
Context: Decision analysis
Source: ammerse-analysis.tactic.md
Related: AMMERSE Analysis, Environmental Dimension
Automated observation and recording of terminology, decisions, or patterns in real-time as work happens, replacing big-bang manual efforts
Context: Meta-Pattern / Practice
Source: living-glossary-practice.md
Related: Agentic Enablement, Incremental Maintenance
Explicit determination of intended outcome for a failure scenario: prevent, reject, degrade, or tolerate
Context: Requirements specification
Source: ATDD_adversarial-acceptance.tactic.md
Related: Acceptance Boundary, Failure Scenario
Percentage of glossary entries initiated by team members (not automation); target >50% indicating healthy ownership and engagement
Context: Architecture - DDD / Metrics
Source: living-glossary-practice.md
Related: Glossary Ownership, Living Glossary
Application of Conway's Law to semantics—organizational communication structure predicts vocabulary structure; teams with infrequent communication develop different vocabularies
Context: Architecture - DDD / Organizational
Source: bounded-context-linguistic-discovery.md
Related: Bounded Context, Vocabulary Fragmentation
Hypothesis that semantic boundaries align with low-frequency communication boundaries between teams
Context: Organizational architecture
Source: context-boundary-inference.tactic.md
Related: Context Boundary, Team Communication Matrix
Architectural separation between framework core files (managed by framework) and local customizations (managed by repository owner) to prevent silent overwrites
Context: Framework Architecture
Source: framework-guardian.agent.md
Related: Framework Guardian, Conflict Classification, Local Customization
Minimal change proposals produced by Curator Claire to align artifacts without overhauling original content, preserving authorial voice
Context: Curation - Change Management
Source: curator.agent.md
Related: Curator Claire, Discrepancy Report, Minimal Change
Quarterly analysis of missing domain terms by comparing codebase terminology to glossary entries
Context: Glossary completeness
Source: glossary-maintenance-workflow.tactic.md
Related: Glossary Maintenance Workflow, Gap Closure Plan
Minimum percentage of code covered by tests required before code can be accepted; typically ≥80% for production code
Context: Quality Gates - Testing
Source: python-pedro.agent.md, java-jenny.agent.md
Related: Self-Review Protocol, Test Coverage, pytest, Quality Gate
Validation that all internal links resolve correctly and referenced documents exist
Context: Documentation quality
Source: documentation-curation-audit.tactic.md
Related: Orphaned File, Documentation Curation Audit
Structural and tonal consistency specialist agent who maintains cross-document integrity, enforces doctrine stack boundaries, and prevents drift through systematic audits
Context: Agent Roles - Content Curation
Source: curator.agent.md
Related: Curator, Structural Consistency, Tonal Integrity, Doctrine Stack, Discrepancy Report, Source vs Distribution
Classification of decision types as Minor (autonomous), Moderate (autonomous with note), or Critical (pause and escalate)
Context: Agent autonomy governance
Source: autonomous-operation-protocol.tactic.md
Related: AFK Mode, Escalation Protocol
Ratio of decision markers not yet promoted to formal ADRs; acceptable <20%, requires action >40%
Context: Development Methodology / Metrics
Source: decision-first-development.md
Related: Decision Marker, ADR Promotion
Structured inline annotation in code or documentation capturing architectural decision (what was decided, rationale, alternatives, consequences) with ADR reference
Context: Development Methodology / Documentation
Source: decision-first-development.md
Related: Decision-First Development, Traceable Decisions
Pre-defined interpretation logic for experiment results determining proceed, pivot, or abandon actions
Context: Experiment analysis
Source: safe-to-fail-experiment-design.tactic.md
Related: Success Criteria, Failure Criteria
Development workflow where architectural decisions are captured systematically throughout the lifecycle, integrating decision rationale with artifacts to preserve "why" knowledge
Context: Development Methodology / Documentation
Source: decision-first-development.md
Related: Decision Marker, Decision Debt, Traceable Decisions
Uninterrupted focus state with high cognitive load from problem-solving where agents stay passive and defer decision capture to session-end
Context: Development Methodology / Human-Agent Interaction
Source: decision-first-development.md
Related: Flow State Awareness, Decision Marker
Analysis process identifying same terms with different meanings across codebases or contexts
Context: Linguistic analysis
Source: code-documentation-analysis.tactic.md
Related: Semantic Conflict, Context Boundary
Process of identifying and documenting prerequisite relationships between tasks, ensuring correct execution sequence and preventing blockers
Context: Planning - Work Sequencing
Source: project-planner.agent.md
Related: Planning Petra, Task Breakdown, DEPENDENCIES
Collection of reusable components, patterns, guidelines, and assets enabling consistent, scalable UI development
Context: Frontend Architecture - Consistency
Source: frontend.agent.md
Related: Frontend Freddy, Component Patterns, UI Architecture
Build automation specialist agent who designs reproducible build, test, and release pipelines with documented runbooks and traceable deployment flows
Context: Agent Roles - Build Automation
Source: build-automation.agent.md
Related: Build Automation Specialist, CI/CD Pipeline, Reproducible Build, Deployment Pipeline, Release Automation
Diagramming specialist agent who transforms conceptual and architectural structures into semantically aligned diagram-as-code artifacts (Mermaid, PlantUML, Graphviz)
Context: Agent Roles - Visualization
Source: diagrammer.agent.md
Related: Diagramming Specialist, Diagram-as-Code, Semantic Fidelity, Visual Representation
Practice of creating diagrams using text-based formats (Mermaid, PlantUML, Graphviz) for version control, reproducibility, and semantic fidelity
Context: Agent Capabilities - Visualization
Source: diagrammer.agent.md
Related: Diagram Daisy, Semantic Fidelity, Mermaid, PlantUML
Anti-pattern where teams create glossary document, consider DDD "done," never update it; glossary becomes stale immediately with no ownership
Context: Architecture - DDD / Anti-Pattern
Source: living-glossary-practice.md
Related: Living Glossary, Glossary Ownership
Skipping required procedures because documentation feels like overhead
Context: Process violation
Source: phase-checkpoint-protocol.md
Related: Phase Checkpoint Protocol
Artifact produced by Curator Claire outlining detected inconsistencies across documents, their locations, and recommended corrective actions
Context: Artifacts - Curation
Source: curator.agent.md
Related: Curator Claire, Corrective Action Set, Structural Consistency
Path variable in doctrine configuration pointing to documentation root directory (default 'docs') where architectural docs, guides, and reference materials reside
Context: Configuration - Path Variables
Source: bootstrap-bill.agent.md
Related: Doctrine Configuration, workspace_root, spec_root, output_root
YAML configuration file (.doctrine/config.yaml) that defines path variables (workspace_root, doc_root, spec_root, output_root) and repository metadata for doctrine framework integration
Context: Artifacts - Repository Configuration
Source: bootstrap-bill.agent.md
Related: Bootstrap Bill, Path Variable, workspace_root, doc_root, spec_root
Systematic audit of directory structure, naming conventions, cross-references, and metadata completeness
Context: Documentation quality
Source: documentation-curation-audit.tactic.md
Related: Structural Consistency, Cross-Reference Integrity
Business-significant state change named in past tense (e.g., OrderPlaced, PaymentReceived, ShipmentCompleted)
Context: Event-driven design
Source: event-storming-discovery.tactic.md
Related: Event Storming, Event Cluster
Process of identifying divergence between installed framework files and canonical manifest through checksum comparison and file classification
Context: Agent Capabilities - Framework Maintenance
Source: framework-guardian.agent.md
Related: Framework Guardian, Manifest, Audit Report, Framework Integrity
Error reporting pattern providing generic user-facing messages and detailed internal diagnostic logs
Context: Error communication
Source: input-validation-fail-fast.tactic.md
Related: Fail-Fast Validation, Reference Number
Low-trust variant where agents commit to agent-trunk branch, humans review PRs to main, providing safety net during agent adoption phase
Context: Version Control / Trust Model
Source: trunk-based-development.md
Related: Trunk-Based Development, Human Review Gate
Analysis of whether practice is economically viable given current costs (labor, tooling, time) versus benefits delivered
Context: Meta-Pattern / Decision Framework
Source: living-glossary-practice.md
Related: Feasibility Shift, ROI Threshold
Writer and editor specialist agent who revises existing content for tone, clarity, and alignment while preserving factual integrity and authorial rhythm
Context: Agent Roles - Editorial
Source: writer-editor.agent.md
Related: Writer-Editor, Editorial Specialist, Paragraph-Level Refinement, Voice Alignment, Boy Scout Rule
Review dimension focused on clarity, readability, style consistency, grammar, terminology alignment, and appropriate tone
Context: Quality Assurance - Review Types
Source: reviewer.agent.md
Related: Review Dimensions, Style Consistency, Terminology, Glossary
Graduated enforcement levels for glossary terms: Advisory (suggest), Acknowledgment Required (warn), Hard Failure (block), chosen per term based on criticality
Context: Architecture - DDD / Governance
Source: living-glossary-practice.md
Related: Living Glossary, Glossary Ownership
The process of flagging issues, uncertainties, or conflicts that require human intervention or inter-agent coordination. Escalation uses integrity markers (❗️,
Reference: Directive 011
Related: Integrity Symbol, Risk
Procedure for pausing autonomous work and documenting critical decisions requiring human guidance
Context: Agent autonomy governance
Source: autonomous-operation-protocol.tactic.md
Related: Decision Boundary, AFK Mode
Tactic requiring explicit context clarification before work begins, then freezing that context for task duration to prevent scope drift
Context: Work initiation discipline
Source: context-establish-and-freeze.tactic.md
Related: Frozen Context, Context Drift, Goals and Non-Goals
Natural grouping of related domain events indicating potential bounded context boundaries
Context: DDD discovery
Source: event-storming-discovery.tactic.md
Related: Domain Event, Event Storming
Collaborative workshop technique using sticky notes to discover domain events, processes, and bounded context boundaries
Context: DDD discovery methodology
Source: event-storming-discovery.tactic.md
Related: Domain Event, Event Cluster, Boundary Signal
Color-coding system: Orange=events, Blue=actors, Purple=policies, Light Blue=commands
Context: Workshop visualization
Source: event-storming-discovery.tactic.md
Related: Event Storming
Collecting evidence to justify pre-decided requirements rather than genuinely testing hypotheses
Context: Research anti-pattern
Source: requirements-validation-workflow.tactic.md
Related: Claim Inventory, Intellectual Honesty
Classification of claim support: Empirical (quantitative), Observational (qualitative), Theoretical (conceptual), or Prescriptive (best practices)
Context: Research methodology
Source: claim-inventory-development.tactic.md
Related: Claim Inventory, Confidence Level
Requirements analysis approach grounding all claims in verifiable evidence with classified evidence types (Empirical, Observational, Theoretical, Prescriptive)
Context: Requirements Engineering / Research Method
Source: evidence-based-requirements.md
Related: Claim Classification, Testability Assessment, Requirements Validation
Automated test that validates acceptance criteria through code execution, providing verifiable evidence of requirement satisfaction
Context: Testing - Automation
Source: python-pedro.agent.md, backend-dev.agent.md
Related: ATDD, Acceptance Criteria, Phase 4 (Acceptance Tests)
Explicit scope, duration, resource, and audience limits containing risk of experimentation
Context: Experiment design
Source: safe-to-fail-experiment-design.tactic.md
Related: Safe-to-Fail Experiment, Blast Radius
Second phase of reverse speccing where Agent B validates reconstruction against ground truth with full context access
Context: Test validation procedure
Source: test-to-system-reconstruction.tactic.md
Related: Naive Reconstruction, Accuracy Assessment
Automated process (npm run export:all, npm run deploy:all) that transforms doctrine/ source content into tool-specific distribution formats with semantic preservation
Context: Doctrine Stack - Distribution
Source: curator.agent.md
Related: Source vs Distribution, Format Transformation, Semantic Preservation
Result consolidation performed outside execution loop to maintain iteration independence
Context: Data processing pattern
Source: execution-fresh-context-iteration.tactic.md
Related: Fresh Context Iteration
Storage location (work/notes/external_memory/) for offloaded task-specific details, allowing agents to swap context in/out while maintaining core behavioral norms
Context: Framework - Context Management
Source: Directive 002 (Context Notes), Directive 003 (Repository Structure)
Related: Token Discipline, Context Layer
Analysis discipline that separates observable fact extraction from meaning assignment to prevent premature interpretation bias
Context: Analysis methodology
Source: analysis-extract-before-interpret.tactic.md
Related: Extraction Phase, Interpretation Phase
First step of analysis where observable elements are extracted verbatim without judgment or conclusions
Context: Analysis procedure
Source: analysis-extract-before-interpret.tactic.md
Related: Extract Before Interpret, Interpretation Phase
Input validation strategy halting at first failure rather than continuing through all checks
Context: Error handling pattern
Source: input-validation-fail-fast.tactic.md
Related: Validation Sequence, Dual-Level Error Feedback
Concrete description of how a feature could fail, including trigger, manifestation, and likely cause
Context: Risk analysis
Source: ATDD_adversarial-acceptance.tactic.md
Related: Adversarial Testing, Premortem Analysis
Alternative approach when preferred tooling (fd, rg, ast-grep, etc.) is unavailable, using universally available alternatives (find, grep) with equivalent functionality
Context: Framework - Tooling
Source: Directive 013 (Tooling Setup & Fallbacks)
Related: Tool Suite, Escalation
Context boundaries that cut across natural process flows due to organizational politics or technical convenience rather than semantic divisions
Context: Architecture - DDD / Failure Mode
Source: bounded-context-linguistic-discovery.md
Related: Bounded Context, Event Storming Validation
Failure mode where automated detection produces low-quality outputs (overaggressive, poor context awareness), damaging trust in tooling
Context: Quality Assurance / Failure Mode
Source: living-glossary-practice.md, language-first-architecture.md
Related: Confidence Threshold, Human Review Loop
Testable prediction restated from a claim in If-Then-Because format with measurable outcomes
Context: Scientific method application
Source: requirements-validation-workflow.tactic.md
Related: Validation Experiment, Success Criteria
Transformation where previously infeasible practice becomes operationally viable due to technological change (e.g., agentic systems enabling continuous linguistic monitoring)
Context: Meta-Pattern / Technology Impact
Source: language-first-architecture.md, living-glossary-practice.md
Related: Economic Feasibility, Agentic Enablement
Coordination pattern using YAML task files and Git commits to enable asynchronous multi-agent workflows without queues, servers, or network dependencies
Context: Orchestration / Coordination Pattern
Source: file-based-orchestration.md, work-directory-orchestration.md
Related: Task Lifecycle, Work Directory Model, Asynchronous Coordination
Executive summary component of review report highlighting critical issues, key findings, and top recommendations for quick stakeholder assessment
Context: Quality Assurance - Reporting
Source: reviewer.agent.md
Related: Reviewer, Review Report, Review Dimensions
Extracted abstraction representing a clear domain or technical concept with single, well-defined responsibility
Context: Refactoring pattern
Source: refactoring-extract-first-order-concept.tactic.md
Related: Concept Extraction, Single Responsibility
Unreliable or inconsistent terminal/shell interactions in agent-based workflows that require remediation techniques to ensure reliable command execution
Context: Framework - Terminal Operations
Source: Directive 001 (CLI and Shell Tooling)
Related: Remediation Technique
Practice of adapting decision capture behavior to human productivity rhythms: Deep Creation (defer), Agent Collaboration (real-time), Reflection/Synthesis (batch)
Context: Development Methodology / Human-Agent Interaction
Source: decision-first-development.md
Related: Decision-First Development, Deep Creation Flow
Framework integrity specialist agent who audits framework installations against canonical manifests, detects drift, and guides safe upgrades while preserving local customizations
Context: Agent Roles - Framework Maintenance
Source: framework-guardian.agent.md
Related: Framework Integrity, Manifest Audit, Drift Detection, Upgrade Plan, Core/Local Boundary
State where installed framework files match canonical manifest specifications without unauthorized modifications or drift
Context: Framework Maintenance - Quality
Source: framework-guardian.agent.md
Related: Framework Guardian, Manifest, Drift Detection, Audit Report
Execution pattern running single instruction repeatedly in fresh context with no memory between runs
Context: Deterministic execution
Source: execution-fresh-context-iteration.tactic.md
Related: Context Carry-over, External Aggregation
Front-end specialist agent integrating design, technical architecture, and usability reasoning for coherent UI systems with maintainable component patterns
Context: Agent Roles - Frontend Development
Source: frontend.agent.md
Related: Frontend Specialist, UI Architecture, Component Patterns, State Boundaries, Design System
Agreed-upon goals, constraints, and assumptions that remain fixed during task execution unless explicitly renegotiated
Context: Scope management
Source: context-establish-and-freeze.tactic.md
Related: Establish and Freeze Context, Context Drift
BDD scenario structure: Given (context), When (action), Then (expected outcome)
Context: Specification format
Source: development-bdd.tactic.md
Related: BDD Scenario, Observable Behavior
Mental model treating glossary not as documentation but as living infrastructure with continuous updates, clear ownership, workflow integration, and tiered enforcement
Context: Architecture - DDD / Mental Model
Source: living-glossary-practice.md
Related: Living Glossary, Enforcement Tier, Glossary Ownership
Failure mode where centralized control over terminology becomes organizational power lever rather than tool for shared understanding
Context: Organizational / Failure Mode
Source: living-glossary-practice.md
Related: Linguistic Policing, Weaponized Standards
Four-cycle process: Continuous Capture → Weekly Triage → Quarterly Health Check → Annual Governance Retrospective
Context: Living glossary operations
Source: glossary-maintenance-workflow.tactic.md
Related: Glossary Triage, Staleness Audit, Coverage Assessment
Assignment of responsibility for terminology decisions to specific context owners (team leads, domain experts, architects) per bounded context
Context: Architecture - DDD / Governance
Source: living-glossary-practice.md
Related: Living Glossary, Bounded Context, Human in Charge
Weekly 30-minute session reviewing agent-generated candidates and making approve/reject/defer/merge decisions
Context: Glossary workflow
Source: glossary-maintenance-workflow.tactic.md
Related: Glossary Maintenance Workflow, Context Owner
Explicit declaration of what a task is expected to achieve AND what is deliberately out of scope
Context: Task scoping
Source: context-establish-and-freeze.tactic.md
Related: Establish and Freeze Context
Anti-pattern of adding features, abstractions, or optimizations "just in case" or "for completeness" without evidence of actual need
Context: Development Methodology / Anti-Pattern
Source: locality-of-change.md
Related: Locality of Change, Premature Abstraction, Complexity Creep
Anti-pattern of opportunistic improvements beyond stated goal (while-I'm-here syndrome)
Context: Scope violation
Source: change-apply-smallest-viable-diff.tactic.md
Related: Scope Creep, Smallest Viable Diff
Principle of providing fallback strategies for every tool to ensure agents can operate even when preferred tools unavailable
Context: Development Environment / Resilience
Source: tooling-setup-best-practices.md
Related: Tool Selection Rigor, Fallback Strategy
TDD/ATDD workflow state where tests pass after implementation, validating correct behavior
Context: Test-driven development
Source: 6-phase-spec-driven-implementation-flow.md
Related: RED Phase, Implementation Phase
Explicit transfer of responsibility from one agent to another at phase boundaries, documented in commit messages
Context: SDD workflow coordination
Source: 6-phase-spec-driven-implementation-flow.md
Related: Phase Checkpoint Protocol, Phase Declaration
Structured process for transferring work between agents in a multi-agent workflow, specifying what artifacts are ready, which agent receives them, and validation criteria
Context: Agent Collaboration
Source: analyst-annie.agent.md, architect.agent.md, project-planner.agent.md
Related: Phase Authority, Spec-Driven Development, Agent Assignment, HANDOFFS.md
Workflow where completing agent encodes follow-up work via result.next_agent block, orchestrator creates new task, copies context/artefacts for seamless continuation
Context: Orchestration / Workflow
Source: work-directory-orchestration.md, agent-profile-handoff-patterns.md
Related: File-Based Orchestration, Agent Coordination, Task Chain
Coordination artifact documenting which work products are ready for which next agent in the workflow
Context: Artifacts - Orchestration
Source: manager.agent.md
Related: Manager Mike, Hand-off Protocol, Agent Assignment
Concrete resource boundary (financial, time, physical, mental, moral, social) triggering immediate stop
Context: Boundary enforcement
Source: stopping-conditions.tactic.md
Related: Stopping Condition, Trigger Declaration
Review feedback that is specific, categorized by severity, and respectful of author context
Context: Review quality
Source: review-intent-and-risk-first.tactic.md
Related: Intent-First Review, Low-Noise Feedback
A governance principle emphasizing that humans retain ultimate responsibility, authority, and decision-making power in agent-augmented workflows. Distinct from "human in the loop" (which focuses on oversight/approval), "human in charge" explicitly centers human accountability and intervention rights. The human in charge bears responsibility for work outcomes and maintains authority to override, redirect, or halt agentic operations at any point.
Key distinctions from "human in the loop":
- Responsibility: Human bears accountability for outcomes, not just oversight
- Authority: Power to make significant decisions and interventions
- Control: Ability to halt, redirect, or override agent operations
- Ownership: Ultimate arbiter of quality, direction, and delivery
Practical implications:
- Agents request permission for high-impact changes
- Humans retain approval authority over critical decisions
- Agents escalate uncertainty or risk immediately
- Human judgment overrides agent recommendations when conflicting
Reference: Directive 026 (Commit Protocol), Directive 011 (Risk & Escalation)
Related: Escalation, Alignment, Collaboration Contract
Process where automated detection surfaces candidates but humans make final decisions, maintaining accountability and avoiding automation bias
Context: Quality Assurance / Governance
Source: living-glossary-practice.md
Related: Human in Charge, Confidence Threshold
Process of identifying affected artifacts (tests, code, documentation) when requirements change, enabled by traceability chain links
Context: Development Methodology / Change Management
Source: traceability-chain-pattern.md
Related: Traceability Chain, Bidirectional Linking
C4 model approach using hierarchical abstraction layers (Context, Container, Component, Code) for progressive disclosure of technical detail
Context: Architecture / Diagramming
Source: design_diagramming-incremental_detail.md
Related: C4 Model, Progressive Disclosure, Abstraction Layer
Practice of making small, frequent updates to living artifacts (glossaries, ADRs, specs) rather than periodic big-bang efforts
Context: Meta-Pattern / Practice
Source: living-glossary-practice.md
Related: Continuous Capture, Living Glossary
Code review practice examining changes without expanding scope or rewriting implementation
Context: Review discipline
Source: code-review-incremental.tactic.md
Related: Review Boundary, Locality of Change
Components directly responsible for feature logic that should NOT be mocked in tests
Context: Unit testing scope
Source: test-boundaries-by-responsibility.tactic.md
Related: Outside Boundary, Test Boundary
Set of APIs, protocols, data formats, and contracts through which a system interacts with external systems or services
Context: Backend Architecture - External Integration
Source: backend-dev.agent.md, bootstrap-bill.agent.md
Related: Backend Benny, Bootstrap Bill, API Contract, SURFACES
Review discipline summarizing apparent intent before identifying risks or suggesting changes
Context: Code review methodology
Source: review-intent-and-risk-first.tactic.md
Related: Risk Categorization, High-Value Feedback
Second step of analysis where meaning is assigned to previously extracted facts as a separate, explicit activity
Context: Analysis procedure
Source: analysis-extract-before-interpret.tactic.md
Related: Extraction Phase, Extract Before Interpret
Java development specialist agent focused on code quality, style enforcement, and testing standards using Maven and Java ecosystem tooling
Context: Agent Roles - Java Development
Source: java-jenny.agent.md
Related: Java Specialist, Code Quality, Maven, JVM Ecosystem, Test-First Development
Documentation of lessons in discoverable locations (tactics, directives, templates, guidelines)
Context: Organizational learning
Source: reflection-post-action-learning-loop.tactic.md
Related: Lesson Extraction, Post-Action Learning Loop
Architectural approach treating language drift as an early signal of deeper architectural problems, recognizing that linguistic fragmentation predicts system issues
Context: Architecture - DDD / Ubiquitous Language
Source: language-first-architecture.md
Related: Linguistic Signal, Bounded Context, Ubiquitous Language
Separation between doctrine stack layers (Guidelines, Approaches, Directives, Tactics, Templates) that must be respected to prevent content type mixing
Context: Doctrine Stack - Architecture
Source: curator.agent.md
Related: Doctrine Stack, Curator Claire, Guideline, Approach, Directive, Tactic
Identification of 1-3 concrete, actionable insights from completed work
Context: Learning practice
Source: reflection-post-action-learning-loop.tactic.md
Related: Post-Action Learning Loop, Knowledge Encoding
Minimal diff artifact produced by Lexical Larry showing suggested edits grouped by rule violated, ready for patch application
Context: Artifacts - Style Analysis
Source: lexical.agent.md
Related: Lexical Larry, LEX_REPORT, Minimal Diff, Rule Violation
Lexical analysis artifact documenting per-file style compliance (tone, rhythm, markdown hygiene) with rule violation annotations
Context: Artifacts - Style Analysis
Source: lexical.agent.md
Related: Lexical Larry, LEX_DELTAS, LEX_TONE_MAP, Style Compliance
Medium detection artifact showing which writing medium (Pattern, Podcast, LinkedIn, Essay) applies to each file with confidence scores
Context: Artifacts - Style Analysis
Source: lexical.agent.md
Related: Lexical Larry, Medium Detection, Tone Fidelity
Lexical analyst specialist agent who evaluates writing style compliance (tone, rhythm, formatting) while preserving authorial voice through minimal, rule-grounded edits
Context: Agent Roles - Style Analysis
Source: lexical.agent.md
Related: Lexical Analyst, Style Compliance, Tone Fidelity, Authorial Voice, LEX_REPORT
Analysis of documentation for style inconsistencies, readability issues, and tone misalignments
Context: Writing quality
Source: lexical-style-diagnostic.tactic.md
Related: Readability Scoring, Tone Consistency, Minimal Diff
Failure mode where glossary enforcement becomes compliance regime instead of shared understanding, often due to punitive enforcement or centralized authority
Context: Architecture - DDD / Failure Mode
Source: living-glossary-practice.md, language-first-architecture.md
Related: Enforcement Tier, Glossary as Power Tool
Observable pattern in terminology usage that indicates potential architectural problems before they manifest in code or process failures
Context: Architecture - DDD
Source: language-first-architecture.md
Related: Language-First Architecture, Vocabulary Fragmentation, Semantic Conflict
Continuously updated, executable glossary treated as infrastructure rather than static documentation, evolving with code and domain understanding
Context: Architecture - DDD / Documentation Practice
Source: living-glossary-practice.md
Related: Ubiquitous Language, Bounded Context, Glossary as Executable Artifact
Specification that evolves during development as understanding grows, then freezes when feature complete to become reference documentation
Context: Development Methodology - SDD
Source: spec-driven-development.md
Related: Specification-Driven Development, Specification Lifecycle
Repository-specific modifications to framework files or addition of custom files that must be preserved during framework upgrades
Context: Framework Maintenance - Customization
Source: framework-guardian.agent.md
Related: Framework Guardian, Core/Local Boundary, Conflict Classification
A design principle emphasizing that changes should be measured against actual problems, not hypothetical concerns. Agents must verify problem existence, quantify severity, and prefer simple solutions over architectural enhancements. Discourages gold plating, premature abstraction, and complexity creep.
Reference: Directive 020, approaches/locality-of-change.md
Related: Risk, Escalation, Alignment
Coordination specialist agent who routes tasks to appropriate agents, maintains workflow status maps, and prevents conflicting edits through file-based orchestration
Context: Agent Roles - Orchestration Source: manager.agent.md Related: Coordinator, Task Router, Workflow Status, Hand-off Tracking, AGENT_STATUS
Generalist agent whose collaboration contract and capabilities are inherited and refined by specialist child agents. Serves as fallback when no specialist matches task context or specialists are overloaded.
Context: Agent Specialization Hierarchy Source: DDR-011 Related: Child Agent, Specialization Boundary, Agent Specialization Hierarchy
Canonical list of framework files with checksums and metadata defining expected installation state for drift detection
Context: Framework Maintenance - Source of Truth
Source: framework-guardian.agent.md
Related: Framework Guardian, Framework Integrity, Drift Detection
Process of identifying which writing medium (Pattern, Podcast, LinkedIn, Essay) applies to content based on style patterns and tone markers
Context: Agent Capabilities - Style Analysis
Source: lexical.agent.md
Related: Lexical Larry, LEX_TONE_MAP, Tone Fidelity
Periodic analysis of work logs, prompt assessments, and operational patterns to surface systemic framework improvements and identify recurring issues
Context: Framework Improvement / Operational Practice
Source: meta-analysis.md
Related: Pattern Detection, Anti-Pattern Identification, Framework Evolution
Agent capability to observe and reason about its own execution state, recognize problematic patterns, and make course corrections before completing tasks
Context: Agent Behavior / Self-Observation
Source: ralph-wiggum-loop.md
Related: Ralph Wiggum Loop, Meta-Mode
Structured front matter in tactic files: Invoked By, Related Tactics, Complements
Context: Tactic documentation
Source: tactics-curation.tactic.md
Related: Tactics Curation, Relationship Tracking
Planning artifact that establishes significant checkpoints with goals, themes, decision gates, and validation hooks while remaining resilient to change
Context: Planning - Goal Setting
Source: project-planner.agent.md
Related: Planning Petra, PLAN_OVERVIEW, Batch Planning
Smallest possible code or configuration change required to resolve an issue, upgrade conflict, or implement a fix without unnecessary modifications
Context: Development - Change Management
Source: framework-guardian.agent.md, python-pedro.agent.md
Related: Framework Guardian, Locality of Change, Upgrade Plan
Concrete plan defining prevention actions, detection signals, and response procedures for a failure scenario
Context: Risk management
Source: premortem-risk-identification.tactic.md
Related: Premortem Risk Identification, Failure Scenario
Failure mode where agents skip phases or roles blur due to pressure to keep moving forward without proper validation or hand-off
Context: Development Methodology - Anti-Pattern
Source: spec-driven-6-phase-cycle.md
Related: Six-Phase Cycle, Role Confusion
Tendency to continue work into next phase without proper checkpoint due to task momentum
Context: Workflow violation
Source: phase-checkpoint-protocol.md
Related: Phase Skipping, Phase Checkpoint Protocol
Static type checker for Python that validates type hints and detects type-related errors before runtime
Context: Python Development - Type Safety
Source: python-pedro.agent.md
Related: Python Pedro, Type Hints, Type Checking
Phase where analyst with NO external context reads only tests to infer system behavior, data structures, workflows, and edge cases
Context: Quality Validation / Testing
Source: reverse-speccing.md
Related: Reverse Speccing, Test-as-Documentation
Planning artifact containing small batch of concrete, ready-to-run tasks for immediate execution (typically 1-2 weeks of work)
Context: Artifacts - Planning
Source: project-planner.agent.md
Related: Planning Petra, PLAN_OVERVIEW, Batch Planning, Task Breakdown
System outcomes verifiable without knowledge of implementation details
Context: Testing principle
Source: development-bdd.tactic.md
Related: BDD Scenario, Behavioral Contract
The act of moving detailed context (notes, enumerations, specifics) from active agent memory into external files to conserve token budget while maintaining reference capability
Context: Framework - Context Management
Source: Directive 002 (Context Notes)
Related: Token Discipline, External Memory
Qualitative scoring of alternatives against impact areas (performance, maintainability, etc.)
Context: Decision analysis
Source: adr-drafting-workflow.tactic.md
Related: ADR Drafting Workflow, Trade-off Analysis
Principle of letting patterns emerge naturally from observed practice before codifying them, rather than prescribing patterns prematurely
Context: Meta-Pattern / Design Principle
Source: agent-profile-handoff-patterns.md, locality-of-change.md
Related: Pattern Before Prescription, Premature Abstraction
Code, test, or documentation with no traceable link to requirements, specifications, or architectural decisions, making purpose and validity unclear
Context: Development Methodology / Anti-Pattern
Source: traceability-chain-pattern.md
Related: Traceability Chain, Documentation Debt
Document with no parent README reference or incoming links, reducing discoverability
Context: Documentation structure
Source: documentation-curation-audit.tactic.md
Related: Cross-Reference Integrity, Structural Consistency
Path variable in doctrine configuration pointing to generated artifacts directory (default 'output') for build products, reports, and exports
Context: Configuration - Path Variables
Source: bootstrap-bill.agent.md
Related: Doctrine Configuration, workspace_root, doc_root, Build Artifact
Supporting infrastructure and external dependencies that SHOULD be mocked/stubbed in tests
Context: Unit testing scope
Source: test-boundaries-by-responsibility.tactic.md
Related: Inside Boundary, Test Boundary
Failure mode of creating too many tiny bounded contexts, resulting in excessive translation overhead without cognitive load benefits
Context: Architecture - DDD / Failure Mode
Source: bounded-context-linguistic-discovery.md
Related: Bounded Context, Translation Layer
Discipline requiring pattern analysis across multiple instances before standardizing, ensuring patterns based on real usage not anticipated usage
Context: Meta-Pattern / Design Principle
Source: locality-of-change.md, agent-profile-handoff-patterns.md
Related: Organic Emergence, Evidence-Based Requirements
Analysis technique identifying recurring themes, common mistakes, or successful approaches across multiple work logs to inform framework improvements
Context: Framework Improvement / Analysis
Source: meta-analysis.md
Related: Meta-Analysis, Anti-Pattern Identification
Explicit constraints on response time, throughput, resource usage, or latency that backend services must satisfy
Context: Backend Architecture - Quality Attributes
Source: backend-dev.agent.md
Related: Backend Benny, Persistence Strategy, Service Design
Architectural decision defining how data is stored, retrieved, updated, and maintained including database choice, schema design, and transaction boundaries
Context: Backend Architecture - Data Management
Source: backend-dev.agent.md
Related: Backend Benny, Service Design, Performance Budget
Writing approach filtering content through specific persona goals, frustrations, and engagement style to ensure artifact addresses their stated pain points
Context: Documentation / Communication
Source: target-audience-fit.md
Related: Target-Audience Fit, Audience Segmentation
First phase of Spec-Driven Development where requirements are elicited, specifications are authored, and acceptance criteria are defined; Analyst Annie has PRIMARY authority
Context: Spec-Driven Development Phases
Source: analyst-annie.agent.md
Related: Analyst Annie, Phase Authority, Specification, Acceptance Criteria
Second phase of Spec-Driven Development where solutions are evaluated, trade-off analysis performed, and architectural design approved; Architect Alphonso has PRIMARY authority
Context: Spec-Driven Development Phases
Source: architect.agent.md
Related: Architect Alphonso, Phase Authority, Trade-off Analysis, ADR
Third phase of Spec-Driven Development where task breakdown, dependency analysis, and agent assignment occur; Planning Petra has PRIMARY authority
Context: Spec-Driven Development Phases
Source: project-planner.agent.md
Related: Planning Petra, Phase Authority, Task Breakdown, Dependency Mapping, YAML Task File
Fourth phase of Spec-Driven Development where acceptance criteria are implemented as executable tests before code implementation begins
Context: Spec-Driven Development Phases
Source: analyst-annie.agent.md, backend-dev.agent.md
Related: ATDD, Acceptance Criteria, Executable Test, Phase Authority
Fifth phase of Spec-Driven Development where production code is written to satisfy acceptance tests using TDD cycle
Context: Spec-Driven Development Phases
Source: backend-dev.agent.md, python-pedro.agent.md
Related: TDD, RED-GREEN-REFACTOR, Phase Authority
Sixth phase of Spec-Driven Development where multiple review types occur - architecture compliance (Architect), acceptance criteria validation (Analyst), code quality (Reviewer)
Context: Spec-Driven Development Phases
Source: analyst-annie.agent.md, architect.agent.md, reviewer.agent.md
Related: Review, Architecture Compliance, Acceptance Criteria, Code Quality
Designation of which agent has PRIMARY, CONSULT, or NO authority during each phase of the Spec-Driven Development workflow
Context: Agent Collaboration - Spec-Driven Development
Source: analyst-annie.agent.md, architect.agent.md, project-planner.agent.md
Related: Spec-Driven Development, Phase Checkpoint Protocol, Hand-off Protocol, Phase 1 (Analysis), Phase 2 (Architecture), Phase 3 (Planning)
Self-observation procedure executed at the end of each phase to verify completion, assess authority for next phase, and ensure proper hand-off
Context: Development Methodology - SDD
Source: spec-driven-6-phase-cycle.md
Related: Six-Phase Cycle, Ralph Wiggum Loop
Commit message annotation indicating which phase of the 6-phase cycle a commit belongs to
Context: SDD commit practices
Source: 6-phase-spec-driven-implementation-flow.md
Related: Hand-off, Phase Checkpoint Protocol
Workflow violation where an agent bypasses required intermediate phases (e.g., jumping from Analysis directly to Implementation)
Context: SDD workflow violations
Source: phase-checkpoint-protocol.md
Related: Phase Checkpoint Protocol, Role Overstepping
Planning artifact documenting current goals, themes, and focus areas for project execution
Context: Artifacts - Planning
Source: project-planner.agent.md
Related: Planning Petra, NEXT_BATCH, Milestone Definition
Project planning specialist agent who translates strategic intent into executable, assumption-aware plans with milestone definitions and dependency mapping
Context: Agent Roles - Project Planning
Source: project-planner.agent.md
Related: Planning Specialist, Milestone Definition, Dependency Mapping, Batch Planning, PLAN_OVERVIEW
Reflection tactic capturing concrete lessons after task completion to inform future work
Context: Continuous improvement
Source: reflection-post-action-learning-loop.tactic.md
Related: Lesson Extraction, Knowledge Encoding
Anti-pattern of creating frameworks, lookup tables, or automation before patterns stabilize or use cases mature, adding maintenance burden without proportional value
Context: Development Methodology / Anti-Pattern
Source: locality-of-change.md
Related: Gold Plating, Complexity Creep
Proactive failure analysis that assumes a proposal has already failed and identifies critical failure modes before execution
Context: Risk management
Source: premortem-risk-identification.tactic.md
Related: Failure Scenario, Mitigation Strategy, Risk Matrix
Four-step analysis (Evidence Collection, Severity Measurement, Baseline Option, Simple Alternatives First) required before proposing solutions to verify problem exists and justify complexity
Context: Development Methodology / Decision Framework
Source: locality-of-change.md
Related: Locality of Change, Evidence-Based Requirements
Principle of revealing system complexity gradually through hierarchical views, allowing stakeholders to consume exactly the abstraction level they need
Context: Architecture / Communication
Source: design_diagramming-incremental_detail.md
Related: Incremental Detail Design, C4 Model, Audience Alignment
Python testing framework used for unit tests, integration tests, and acceptance tests with fixtures, parametrization, and coverage reporting
Context: Python Development - Testing
Source: python-pedro.agent.md
Related: Python Pedro, Test-First Development, Coverage Threshold
Python development specialist agent applying ATDD + TDD with type safety, idiomatic Python 3.9+ patterns, and comprehensive testing using pytest ecosystem
Context: Agent Roles - Python Development
Source: python-pedro.agent.md
Related: Python Specialist, Type Hints, pytest, Coverage Threshold, Self-Review Protocol
Self-aware observation pattern where agents periodically monitor their own execution state, detect problematic patterns (drift, confusion, misalignment), and self-correct mid-task
Context: Agent Behavior / Self-Observation
Source: ralph-wiggum-loop.md
Related: Self-Observation Checkpoint, Meta-Awareness, Course Correction
Quantitative assessment using metrics like Flesch-Kincaid grade level and paragraph length distribution
Context: Writing analysis
Source: lexical-style-diagnostic.tactic.md
Related: Lexical Style Diagnostic
ATDD workflow state where acceptance tests have been written but fail, proving tests work before implementation
Context: Test-driven development
Source: 6-phase-spec-driven-implementation-flow.md
Related: GREEN Phase, Acceptance Test Implementation
TDD cycle where tests are written first and fail (RED), minimal code makes them pass (GREEN), then code is improved while keeping tests green (REFACTOR)
Context: Development Practices - TDD Source: python-pedro.agent.md, backend-dev.agent.md, java-jenny.agent.md Related: TDD, Test-First Development, Directive 017
Manager Mike process that reviews existing task assignments and updates them to use more specific specialist agents when available. Used for backward compatibility and after new specialists are introduced.
Context: Agent Collaboration - Orchestration Migration Source: DDR-011 Related: SELECT_APPROPRIATE_AGENT, Agent Specialization Hierarchy
Understanding that same concept may have different names in different registers (technical vs. user-facing, internal vs. external) without indicating semantic conflict
Context: Architecture - DDD / Nuance
Source: living-glossary-practice.md
Related: Semantic Conflict, False Positives
Benefit of test-first bug fixing where failing-then-passing test becomes permanent guard preventing bug from returning unnoticed
Context: Development Methodology / Testing
Source: test-first-bug-fixing.md
Related: Test-First Bug Fixing, Test Validates Fix
One-sentence explanation for each alternative not selected, documenting why it was rejected
Context: Decision transparency
Source: adr-drafting-workflow.tactic.md
Related: ADR Drafting Workflow, Alternatives Considered
Fallback workflow for handling unreliable terminal interactions by creating shell scripts, piping output to files, and capturing results from files instead of direct terminal interaction
Context: Framework - Terminal Operations
Source: Directive 001 (CLI and Shell Tooling)
Related: Bypass Check, Flaky Terminal Behavior
Structural artifact generated by Bootstrap Bill that documents repository topology, directory roles, key files, and navigation paths for multi-agent orientation
Context: Artifacts - Repository Structure
Source: bootstrap-bill.agent.md
Related: Bootstrap Bill, Repository Scaffolding, Topology Mapping, SURFACES, WORKFLOWS
Bootstrap procedure creating standard directory structure, configuration files, and initial documentation per SDD framework
Context: Project setup
Source: repository-initialization.tactic.md
Related: Directory Structure, Configuration Files
Process of mapping repository topology and generating structural artifacts (REPO_MAP, SURFACES, WORKFLOWS) to enable multi-agent collaboration
Context: Agent Specializations
Source: bootstrap-bill.agent.md
Related: Bootstrap Bill, REPO_MAP, Topology Mapping, Structural Artifact
Real-world or production-like data samples used during requirements analysis to capture actual patterns, edge cases, and validation scenarios
Context: Testing - Data Quality
Source: analyst-annie.agent.md
Related: Analyst Annie, Validation Script, Data Validation
Build process that produces identical artifacts from the same source inputs across different environments and time periods
Context: Build Automation - Reliability
Source: build-automation.agent.md
Related: DevOps Danny, Build Graph, CI/CD Pipeline
Agent specialization focused on requirements elicitation, specification authoring, data validation, and production-data alignment to reduce ambiguity
Context: Agent Specializations
Source: analyst-annie.agent.md
Related: Analyst Annie, Specification, Validation Script, Data Quality
Five-phase process (Research & Claim Extraction, Prioritization, Experiment Design, Validation Execution, Requirements Synthesis) transforming assumptions into validated knowledge
Context: Requirements Engineering
Source: evidence-based-requirements.md
Related: Evidence-Based Requirements, Testability Assessment
Incremental process of shifting calls from old to new implementation one call site or module at a time
Context: Strangler fig execution Source: refactoring-strangler-fig.tactic.md Related: Strangler Fig Pattern, Coexistence Period
Numeric specificity score (0-100) for specialist agents. Higher priority agents preferred when multiple match context. Parent agents default to 50, specialists typically 60-90, local specialists receive +20 boost.
Context: Agent Collaboration - Orchestration Source: DDR-011 Related: Agent Specialization Hierarchy, SELECT_APPROPRIATE_AGENT, Specialization Context
Research and corroboration specialist agent who gathers, synthesizes, and contextualizes information with source-grounded summaries for systemic reasoning
Context: Agent Roles - Research
Source: researcher.agent.md
Related: Research Specialist, Literature Synthesis, Comparative Analysis, Source Grounding, Verifiable Knowledge
Dual-agent validation technique reconstructing system understanding purely from test code to measure how effectively tests serve as executable specifications
Context: Quality Validation / Testing
Source: reverse-speccing.md
Related: Test-as-Documentation, Naive Reconstruction, Architecture Blind Spot
Pre-implemented rollback capability using version control, feature flags, or parallel deployment
Context: Risk mitigation
Source: safe-to-fail-experiment-design.tactic.md
Related: Safe-to-Fail Experiment, Feature Flag
Multiple perspectives applied during systematic quality review - structural (organization), editorial (clarity), technical (accuracy), standards compliance
Context: Quality Assurance - Review
Source: reviewer.agent.md
Related: Reviewer, Structural Review, Editorial Review, Technical Review, Standards Compliance Review
Comprehensive quality assurance artifact documenting findings across multiple review dimensions (structural, editorial, technical, standards) with prioritized recommendations
Context: Artifacts - Quality Assurance
Source: reviewer.agent.md
Related: Reviewer, Review Dimensions, Finding Summary, Validation Checklist
Quality assurance specialist agent conducting systematic multi-dimensional reviews (structural, editorial, technical, standards compliance) without making direct changes
Context: Agent Roles - Quality Assurance
Source: reviewer.agent.md
Related: Quality Assurance, Review Dimensions, Finding Summary, Review Report, Validation Checklist
Classification of review concerns by type: Correctness, Maintainability, Impact, Misuse
Context: Review analysis
Source: review-intent-and-risk-first.tactic.md
Related: Intent-First Review, Impact Assessment
2x2 grid plotting failure scenarios by impact (vertical) and likelihood (horizontal) for prioritization
Context: Risk prioritization
Source: premortem-risk-identification.tactic.md
Related: Premortem Risk Identification, Impact Rating
Failure mode where specialists perform work outside their authority (e.g., analysts making architectural decisions, architects implementing code)
Context: Development Methodology - Anti-Pattern
Source: spec-driven-6-phase-cycle.md
Related: Momentum Bias, Role Separation
Violation where an agent performs work outside their designated authority level (PRIMARY, CONSULT, or NO)
Context: Agent authority boundaries
Source: phase-checkpoint-protocol.md
Related: Phase Skipping, Role Boundaries Table
Design principle ensuring each phase has a distinct primary owner with explicit authority boundaries to prevent role confusion and maintain accountability
Context: Development Methodology - SDD
Source: spec-driven-6-phase-cycle.md
Related: Six-Phase Cycle, Role Confusion
Fast Python linter replacing flake8, isort, pydocstyle for code quality checks and style enforcement
Context: Python Development - Code Quality
Source: python-pedro.agent.md
Related: Python Pedro, black, Code Quality
Heuristic to consider extraction when duplication appears in three locations, not two, to avoid premature abstraction
Context: Refactoring timing
Source: refactoring-extract-first-order-concept.tactic.md
Related: Concept Extraction, Premature Abstraction
Bounded exploration with reversibility mechanisms, explicit success/failure criteria, and acceptable loss limits
Context: Innovation practice
Source: safe-to-fail-experiment-design.tactic.md
Related: Experiment Boundary, Reversibility Mechanism, Blast Radius
Design approach using concrete Given/When/Then scenarios to clarify requirements and drive implementation decisions
Context: Development Methodology - SDD
Source: spec-driven-development.md
Related: Specification-Driven Development, User Scenario
Automated verification that YAML task files comply with defined structure before marking tasks complete
Context: File-based orchestration quality
Source: task-completion-validation.tactic.md
Related: Task Schema, Validation Gate
Documentation and transcription specialist agent who maintains traceable, neutral documentation integrity through structured summaries and meeting notes
Context: Agent Roles - Documentation Source: scribe.agent.md Related: Documentation Specialist, Meeting Notes, Structured Summary, Neutral Tone, Timestamp
Orchestration tactic that determines most appropriate agent for a task considering specialization hierarchy, context matching, workload, and complexity. Invoked by Manager Mike during task assignment, handoff processing, and reassignment passes.
Context: Agent Collaboration - Orchestration Source: DDR-011, SELECT_APPROPRIATE_AGENT.tactic.md Related: Agent Specialization Hierarchy, Routing Priority, Reassignment Pass
Structured protocol executed at trigger points (time elapsed, warning signs) where agent switches to meta-mode to assess execution state and decide continue/adjust/escalate
Context: Agent Behavior / Self-Observation
Source: ralph-wiggum-loop.md
Related: Ralph Wiggum Loop, Meta-Mode, Warning Sign Detection
Systematic quality checklist executed by development agents before marking work complete, including test execution, type checking, linting, coverage validation, and ADR compliance
Context: Quality Gates - Development
Source: python-pedro.agent.md
Related: Coverage Threshold, Type Checking, Lint Results, ADR Compliance
Grouping of related domain terms based on co-occurrence patterns and semantic similarity
Context: Linguistic analysis
Source: code-documentation-analysis.tactic.md
Related: Co-occurrence Analysis, Vocabulary Domain
Situation where the same term has different meanings in different contexts, indicating a hidden bounded context boundary requiring explicit management
Context: Architecture - DDD
Source: language-first-architecture.md, bounded-context-linguistic-discovery.md
Related: Linguistic Signal, Bounded Context Boundary, Translation Layer
Quality measure of how accurately a diagram or visualization represents the underlying conceptual, architectural, or organizational relationships
Context: Agent Capabilities - Visualization
Source: diagrammer.agent.md
Related: Diagram Daisy, Diagram-as-Code, Visual Representation
Markdown heading structure using single h1 per file with ordered h2+ hierarchy to convey document structure semantically
Context: Documentation / Markdown
Source: style-execution-primers.md
Related: Style Execution Primer, CommonMark Compliance
Architectural process defining service boundaries, API contracts, integration patterns, and failure modes for backend systems
Context: Backend Architecture - System Design
Source: backend-dev.agent.md
Related: Backend Benny, API Contract, Integration Surface, Persistence Strategy
Safety mechanism running old and new implementations in parallel to compare outputs before full rerouting
Context: Migration validation
Source: refactoring-strangler-fig.tactic.md
Related: Strangler Fig Pattern, Feature Flag
Resources multiple teams contribute to: repositories, documentation, meetings
Context: Collaboration indicator
Source: team-interaction-mapping.tactic.md
Related: Team Interaction Mapping, Communication Frequency
Flexible review pattern for trunk-based development: Ship (commit directly), Show (commit + notify for async review), Ask (branch + pre-merge review)
Context: Version Control / Review Pattern
Source: trunk-based-development.md
Related: Trunk-Based Development, Code Review Discipline
Feature branch with maximum lifetime of 24 hours (target 4-8h) to minimize integration risk and maintain trunk stability
Context: Version Control / Pattern
Source: trunk-based-development.md
Related: Trunk-Based Development, Branch Age Warning
Labeled portion of document indicating which persona it serves (e.g., "For Jordan—getting started") to reduce cognitive load when serving multiple audiences
Context: Documentation / Structure
Source: target-audience-fit.md
Related: Target-Audience Fit, Persona-Driven Writing
Active practice of maintaining architectural simplicity over time by resisting complexity creep, validating problem severity before adding solutions, and favoring 80/20 approaches
Context: Development Methodology / Design Principle
Source: locality-of-change.md
Related: Locality of Change, Complexity Creep
Structured specification-driven workflow with distinct phases (Analysis, Architecture, Planning, Acceptance Test, Implementation, Review), each with a primary owner and explicit hand-offs
Context: Development Methodology - SDD
Source: spec-driven-6-phase-cycle.md
Related: Phase Checkpoint Protocol, Momentum Bias Prevention, Role Separation
Principle of introducing changes using minimal modification that achieves stated goal, avoiding unintended side effects
Context: Change discipline
Source: change-apply-smallest-viable-diff.tactic.md
Related: Surgical Modification, Gold-plating, Locality of Change
Architectural distinction where doctrine/ contains canonical source content and tool-specific directories (.github/, .claude/, .opencode/) contain generated distribution artifacts
Context: Doctrine Stack - Architecture
Source: curator.agent.md
Related: Curator Claire, Export Pipeline, Doctrine Stack, Tool-Specific Distribution
Path variable in doctrine configuration pointing to specification files directory (default 'specifications') where functional and technical specs are stored
Context: Configuration - Path Variables
Source: bootstrap-bill.agent.md
Related: Doctrine Configuration, workspace_root, doc_root, Specification
Explicit limits on agent capabilities defining what each agent will and won't do to prevent scope creep and role confusion
Context: Agent Collaboration - Role Definition Source: All agent profiles Related: Collaboration Contract, Hand-off Protocol, Escalation
Declarative conditions in agent profile defining when specialist preferred over parent: language, frameworks, file patterns, domain keywords, writing style, complexity preference.
Context: Agent Collaboration - Routing Source: DDR-011 Related: Agent Specialization Hierarchy, Child Agent, Routing Priority
Progression of specification states: DRAFT → APPROVED → IMPLEMENTED, with status changes marking phase transitions
Context: Development Methodology - SDD
Source: spec-driven-6-phase-cycle.md
Related: Living Specification, Specification Stub
Initial draft specification with structure and metadata created during analysis phase before detailed requirements are fully documented
Context: Development Methodology - SDD
Source: spec-driven-development.md, spec-driven-6-phase-cycle.md
Related: Phase 1 Analysis, Specification Lifecycle
Development methodology where specifications serve as primary artifacts bridging strategic intent and implementation, remaining independent of tests or architectural decisions
Context: Development Methodology - SDD
Source: spec-driven-development.md
Related: Living Specification, Scenario-Driven Design, Specification Stub
Quarterly review identifying outdated definitions (>6 months unchanged) and validating against current implementation
Context: Glossary quality
Source: glossary-maintenance-workflow.tactic.md
Related: Glossary Maintenance Workflow, Coverage Assessment
Percentage of glossary definitions that are outdated; target <10% to maintain glossary value as living reference
Context: Architecture - DDD / Metrics
Source: living-glossary-practice.md
Related: Living Glossary, Incremental Maintenance
Review dimension focused on adherence to style guides, templates, directives, ADRs, file naming conventions, and required metadata
Context: Quality Assurance - Review Types
Source: reviewer.agent.md
Related: Review Dimensions, Style Guide, Template Compliance, Directive Compliance
Explicit divisions defining where application state is owned, how it flows between components, and mutation responsibilities
Context: Frontend Architecture - State Management
Source: frontend.agent.md
Related: Frontend Freddy, UI Architecture, Data Flow
Clear, measurable threshold defining when to stop pursuing a goal based on acceptable loss limits
Context: Commitment management
Source: stopping-conditions.tactic.md
Related: Hard Limit, Warning Signal, Trigger Declaration
Incremental refactoring pattern introducing new implementation alongside old, gradually rerouting behavior, then removing old code
Context: Large-scale refactoring
Source: refactoring-strangler-fig.tactic.md
Related: Rerouting, Coexistence Period, Shadow Mode
Generated documentation or code that describes repository organization, workflows, or architectural patterns for orientation and navigation
Context: Artifacts - Repository Documentation
Source: bootstrap-bill.agent.md
Related: Bootstrap Bill, REPO_MAP, SURFACES, WORKFLOWS
Review dimension focused on organization, flow, completeness, template compliance, cross-reference validity, and heading hierarchy
Context: Quality Assurance - Review Types
Source: reviewer.agent.md
Related: Review Dimensions, Template Compliance, Cross-Reference
Concise, action-ready guidance for agents working in specific formats (Markdown, Python, Perl, PlantUML) focusing on diff-friendliness and template alignment
Context: Documentation / Coding Standards
Source: style-execution-primers.md
Related: Template-First Approach, Semantic Heading, Cross-Cutting Concern
Continuing past stopping conditions by rationalizing invested effort rather than acknowledging limits
Context: Decision bias
Source: stopping-conditions.tactic.md
Related: Stopping Condition, Hard Limit
Metric tracking how often developers override glossary checks; >10% indicates enforcement too strict or checks producing false positives
Context: Quality Assurance / Metrics
Source: living-glossary-practice.md
Related: Enforcement Tier, False Positives
Artifact documenting integration points, APIs, config files, and external dependencies discovered during repository bootstrapping
Context: Artifacts - Repository Structure
Source: bootstrap-bill.agent.md
Related: Bootstrap Bill, REPO_MAP, Integration Surface, API Contract
Precise, minimal code changes targeting only files and sections necessary to achieve goal
Context: Refactoring practice
Source: change-apply-smallest-viable-diff.tactic.md
Related: Smallest Viable Diff, Reviewable Change
Multi-agent integration specialist who merges insights from multiple agents into coherent narratives and conceptual models while preserving source integrity
Context: Agent Roles - Integration
Source: synthesizer.agent.md
Related: Integration Specialist, Pattern Synthesis, Coherence Articulation, Cross-Agent Integration
Architectural artifact that breaks down complex systems into components, interfaces, and relationships with explicit trade-offs and decision rationale
Context: Artifacts - Architecture
Source: architect.agent.md
Related: Architect Alphonso, ADR, Component Diagram, Interface Design
README.md in tactics directory listing all available tactics with intent and invocation context
Context: Doctrine navigation
Source: tactics-curation.tactic.md
Related: Tactics Curation, Discoverability
Meta-tactic maintaining structural, tonal, and metadata integrity across the tactics layer
Context: Doctrine maintenance
Source: tactics-curation.tactic.md
Related: Metadata Header, Tactics Catalog, Template Compliance
Communication pattern ensuring every artifact intentionally addresses specific reader personas with appropriate tone, depth, and structure
Context: Documentation / Communication
Source: target-audience-fit.md
Related: Persona-Driven Writing, Audience Segmentation, Signposted Section
Explicit listing of files a task will modify in YAML artefacts field, enabling pre-commit conflict detection and coordination between agents
Context: Orchestration / Conflict Avoidance
Source: work-directory-orchestration.md, trunk-based-development.md
Related: File-Based Orchestration, Artifact Conflict Detection
Process of decomposing specifications into concrete, executable tasks with clear dependencies, acceptance criteria, and agent assignments
Context: Planning - Work Decomposition
Source: project-planner.agent.md
Related: Planning Petra, Phase 3 (Planning), YAML Task File, Dependency Mapping
The standardized progression of orchestrated tasks through states: new (unassigned work), assigned (routed to specific agent), **in_progress ** (actively being worked), done (completed with results), **archive ** (historical record). Task state transitions are tracked through file-based coordination.
Reference: Directive 019 (File-Based Collaboration)
Related: Orchestration, Work Log
Documentation of communication frequency and artifact sharing patterns between organizational teams
Context: Organizational analysis
Source: context-boundary-inference.tactic.md
Related: Conway's Law, Team Interaction Mapping
Documentation of organizational communication patterns to identify vocabulary clusters and predict semantic boundaries
Context: Organizational analysis
Source: team-interaction-mapping.tactic.md
Related: Communication Frequency, Vocabulary Cluster, Conway's Law
Review dimension focused on accuracy verification, example correctness, reference citations, code functionality, and version currency
Context: Quality Assurance - Review Types
Source: reviewer.agent.md
Related: Review Dimensions, Technical Accuracy, Code Correctness
Validation that tactics follow standardized structure from doctrine/templates/tactic.md
Context: Quality assurance
Source: tactics-curation.tactic.md
Related: Tactics Curation, Structural Consistency
Practice of reusing repository templates before handcrafting new formats to ensure consistency and avoid format proliferation
Context: Documentation / Standards
Source: style-execution-primers.md
Related: Style Execution Primer, Configuration Consistency
Proposed glossary entry with preliminary definition, source references, and confidence level awaiting triage
Context: Glossary workflow
Source: glossary-maintenance-workflow.tactic.md
Related: Glossary Triage, Continuous Capture
Scope of components included in a test based on functional responsibility rather than structural layers
Context: Unit testing strategy
Source: test-boundaries-by-responsibility.tactic.md
Related: Inside Boundary, Outside Boundary, Functional Responsibility
Verification step ensuring test fails for the RIGHT reason (reproduces bug) by temporarily inverting assertion to expect wrong behavior, confirming test accuracy
Context: Development Methodology / Testing
Source: test-first-bug-fixing.md
Related: Test-First Bug Fixing, Regression Prevention
Principle that well-written tests should comprehensively document system behavior so reading tests alone enables understanding "what" and "how" without external docs
Context: Quality Validation / Testing Philosophy
Source: reverse-speccing.md, test-readability-clarity-check.md
Related: Reverse Speccing, Living Specification
Disciplined debugging approach requiring a failing test that reproduces the bug BEFORE modifying production code, transforming trial-and-error into systematic verification
Context: Development Methodology / Testing
Source: test-first-bug-fixing.md
Related: Red-Green-Refactor, Regression Prevention, Test Validates Fix
Evaluation of whether a requirement claim can be validated: Fully Testable (concrete experiment), Partially Testable (proxy metrics), Not Testable (subjective/unfalsifiable)
Context: Requirements Engineering
Source: evidence-based-requirements.md
Related: Evidence-Based Requirements, Claim Classification
Practice of maintaining essential governance layers in active memory while offloading task-specific details to external files, optimizing context window usage without losing behavioral guardrails
Context: Framework - Context Management
Source: Directive 002 (Context Notes)
Related: External Memory, Context Layer, Offloading
Preservation of consistent voice, rhythm, and authorial personality across documents while preventing style drift or flattening
Context: Content Quality - Voice Preservation
Source: curator.agent.md, lexical.agent.md, writer-editor.agent.md
Related: Curator Claire, Lexical Larry, Editor Eddy, Authorial Voice, Voice Fidelity
Alignment of voice with target audience, avoiding informal/formal mixing
Context: Writing quality
Source: lexical-style-diagnostic.tactic.md
Related: Lexical Style Diagnostic, Voice Preservation
Active maintenance of original tone and emotional register during content transformation (translation, editing, summarization)
Context: Content Quality - Voice Preservation
Source: translator.agent.md, writer-editor.agent.md
Related: Authorial Voice, Voice Fidelity, Tonal Integrity
Disciplined framework for choosing tools based on measurable criteria (usage frequency, performance improvement, maintenance activity, security posture)
Context: Development Environment / Best Practice
Source: tooling-setup-best-practices.md
Related: Tooling Necessity Check, Tooling Quality Assessment, Graceful Degradation
Generated content formatted for specific development tools (GitHub Copilot, Claude Desktop, OpenCode, Cursor) deployed to tool-specific directories
Context: Doctrine Stack - Distribution
Source: curator.agent.md
Related: Source vs Distribution, Export Pipeline, .github/, .claude/, .opencode/
Process of scanning and documenting repository structure, identifying directory roles, key files, configuration locations, and navigation paths
Context: Agent Capabilities - Repository Analysis
Source: bootstrap-bill.agent.md
Related: Bootstrap Bill, REPO_MAP, Repository Scaffolding
Bidirectional link pattern connecting artifacts throughout development lifecycle (Strategic Goal → Specification → Tests → ADRs → Implementation → Work Logs)
Context: Development Methodology / Quality Assurance
Source: traceability-chain-pattern.md
Related: Bidirectional Linking, Impact Analysis, Orphaned Artifact
Architectural reasoning process that explicitly documents alternatives considered, evaluation criteria, and rationale for chosen approach
Context: Architecture - Decision Making
Source: architect.agent.md
Related: Architect Alphonso, ADR, Phase 2 (Architecture), System Decomposition
Explicit mapping mechanism protecting downstream contexts from upstream changes by transforming terminology at bounded context boundaries (Anti-Corruption Layer pattern)
Context: Architecture - DDD
Source: bounded-context-linguistic-discovery.md
Related: Anti-Corruption Layer, Bounded Context, Semantic Conflict
Explicit mapping defining how terminology changes when crossing bounded context boundaries
Context: Context integration
Source: context-boundary-inference.tactic.md
Related: Anti-Corruption Layer, Context Boundary
Contextual interpreter specialist agent who preserves authorial tone and rhythm during accurate cross-language translation using voice fidelity techniques
Context: Agent Roles - Translation
Source: translator.agent.md
Related: Translation Specialist, Voice Fidelity, Tone Preservation, Contextual Pass, VOICE_DIFF
Pre-commitment statement in 'When X happens, then I will stop' format conditioning behavior
Context: Behavioral conditioning
Source: stopping-conditions.tactic.md
Related: Stopping Condition, Hard Limit
Monitoring visualization showing key trunk metrics (commit frequency, revert rate, test pass rate, time-to-fix) for stability assessment
Context: Version Control / Monitoring
Source: trunk-based-development.md
Related: Trunk Stability, Trunk-Based Development
Measure of main branch health (>95% test pass rate, <5% revert rate, <15 min time-to-fix) indicating system readiness for continuous deployment
Context: Version Control / Metrics
Source: trunk-based-development.md
Related: Trunk-Based Development, Trunk Health Dashboard
Branching strategy where all developers commit frequently to single shared branch (main), using short-lived feature branches (<24h) only for coordinated changes
Context: Version Control / Collaboration Pattern
Source: trunk-based-development.md
Related: Short-Lived Branch, Trunk Stability, Branch Age Warning
Static analysis validation using mypy (Python) or similar tools to ensure type safety before code acceptance
Context: Quality Gates - Code Quality
Source: python-pedro.agent.md
Related: Self-Review Protocol, mypy, Type Hints, Type Safety
Python 3.9+ annotations specifying expected types for function parameters, return values, and variables to enable static analysis
Context: Python Development - Type Safety
Source: python-pedro.agent.md
Related: Python Pedro, mypy, Type Checking, Type Safety
Structural design of user interface including component hierarchies, state boundaries, data flow patterns, and interaction flows
Context: Frontend Architecture
Source: frontend.agent.md
Related: Frontend Freddy, Component Patterns, State Boundaries, Design System
Detailed document produced by Framework Guardian classifying upgrade conflicts and proposing minimal patches while preserving local customizations
Context: Artifacts - Framework Maintenance
Source: framework-guardian.agent.md
Related: Framework Guardian, Conflict Classification, Core/Local Boundary, Minimal Patch
Classification of validation types: Presence, Format, Range, Logical Consistency, Uniqueness, Referential Integrity
Context: Input validation
Source: input-validation-fail-fast.tactic.md
Related: Validation Sequence
Systematic checklist confirming all review criteria were applied, evidence collected, recommendations prioritized, and quality standards met
Context: Quality Assurance - Verification
Source: reviewer.agent.md
Related: Reviewer, Review Report, Quality Standard
Rigorous test designed to accept or reject a claim hypothesis with clear success criteria and confound controls
Context: Evidence-based validation
Source: claim-inventory-development.tactic.md
Related: Testability Assessment, Falsifiable Hypothesis
Executable script or SQL query that validates requirements against real production data, capturing pass rates and edge cases for specification quality assurance
Context: Artifacts - Requirements Analysis
Source: analyst-annie.agent.md
Related: Analyst Annie, Data Validation, Specification, Representative Data
Ordered validation checks: Presence → Format → Range → Logical Consistency → Uniqueness
Context: Input processing
Source: input-validation-fail-fast.tactic.md
Related: Fail-Fast Validation, Validation Category
Tiered approach balancing stability with updates (Pinned Versions for API-sensitive tools, Package Manager for stable tools, Latest Stable for rapidly evolving tools)
Context: Development Environment / Configuration
Source: tooling-setup-best-practices.md
Related: Configuration Consistency, Tool Selection Rigor
Set of domain terms consistently used by a specific team, indicating ownership and expertise
Context: Linguistic patterns
Source: team-interaction-mapping.tactic.md
Related: Vocabulary Ownership, Team Interaction Mapping
Condition where different parts of a system or organization use inconsistent or conflicting terminology for domain concepts, signaling hidden architectural boundaries
Context: Architecture - DDD
Source: language-first-architecture.md, bounded-context-linguistic-discovery.md
Related: Linguistic Signal, Bounded Context
Assignment of terminology domains to specific teams based on code authorship and documentation patterns
Context: Linguistic governance
Source: context-boundary-inference.tactic.md
Related: Context Boundary, Semantic Clustering
Quality measure of how well translated or edited content preserves the original author's distinctive tone, rhythm, and cadence
Context: Agent Capabilities - Translation/Editing
Source: translator.agent.md, writer-editor.agent.md
Related: Translator Tanya, Editor Eddy, Authorial Voice, Tone Preservation
Agent's ability to recognize internal signals indicating problems (repetitive patterns, lost goal, speculation, verbose output, scope creep, directive violation)
Context: Agent Behavior / Self-Observation
Source: ralph-wiggum-loop.md
Related: Ralph Wiggum Loop, Self-Observation Checkpoint
Observable red flag indicating trouble requiring attention or reassessment
Context: Early detection
Source: stopping-conditions.tactic.md
Related: Stopping Condition, Hard Limit
Organizational anti-pattern where glossary, coding standards, or architectural guidelines used as tool in politics rather than shared understanding
Context: Organizational / Anti-Pattern
Source: living-glossary-practice.md
Related: Linguistic Policing, Glossary as Power Tool
Directory structure encoding orchestration state: collaboration/ (routing), reports/ (logs, metrics), external_memory/ (scratch), notes/ (ideation), planning/ (aids), schemas/ (validators)
Context: Orchestration / Structure
Source: work-directory-orchestration.md
Related: File-Based Orchestration, Task Lifecycle
Chronological log of multi-agent workflow execution maintained by Manager Mike for traceability and coordination
Context: Artifacts - Orchestration
Source: manager.agent.md
Related: Manager Mike, AGENT_STATUS, Hand-off Tracking
Artifact documenting build, test, CI/CD, and deployment workflows discovered during repository bootstrapping
Context: Artifacts - Repository Structure
Source: bootstrap-bill.agent.md
Related: Bootstrap Bill, REPO_MAP, CI/CD Pipeline, Build Graph
Path variable in doctrine configuration pointing to task orchestration workspace directory (default 'work') where agents create work logs, coordination artifacts, and intermediate outputs
Context: Configuration - Path Variables
Source: bootstrap-bill.agent.md
Related: Doctrine Configuration, doc_root, spec_root, output_root
Structured task definition file used in file-based orchestration containing task metadata, dependencies, acceptance criteria, and agent assignment
Context: Orchestration - Task Definition
Source: project-planner.agent.md, manager.agent.md
Related: Planning Petra, Manager Mike, File-Based Orchestration, Task Breakdown
- Reference glossary terms when explaining decisions or documenting work
- Use consistent terminology across all artifacts and communications
- When encountering ambiguous terms, consult this glossary first
- Suggest new terms or clarifications via work logs when gaps are identified
- Use glossary terms when providing instructions to agents
- Reference specific terms when clarity is needed
- Propose term additions or refinements through pull requests
- Review glossary during agent framework onboarding
When directives or agent profiles reference glossary terms, use this format:
See [Term Name](./GLOSSARY.md#term-name) for definition.This glossary is a living document. Updates follow Version Governance (Directive 006):
- Minor updates (clarifications, examples): Update version patch number
- New terms (additions): Update version minor number
- Structural changes (reorganization): Update version major number
All changes require review by Curator agent for consistency and alignment with existing framework documentation.
Maintained by: Curator Claire
Review cycle: Quarterly or when 5+ new terms are identified
Change requests: Submit via ${WORKSPACE_ROOT}/collaboration/inbox/ task file
- AGENTS.md - Core agent specification
- directives/ - Extended directive set (001-019)
- Agent Profiles - Role-specific configurations
- Version Governance - Version tracking system
- Work Log Creation - Documentation standards