πŸ“– Claude Code Workspace Glossary

Comprehensive guide to terminology, concepts, and architecture

Version 1.0 | Last Updated: November 20, 2025

All terms visible

πŸ“‹ Table of Contents

πŸ’¬ Input Methods

β–Ό

Understanding the three ways data and instructions enter your Claude Code workspace.

Manual Input via Chat

β–Ό
Aliases: Chat Input, Conversational Input, Direct Chat
Definition: User types messages directly in the Claude Code chat interface to interact with the system.

Examples:

User: "Create a daily plan for today"
β†’ Triggers /daily-planning command

User: "I need to schedule Lucas's dentist appointment"
β†’ task-manager skill auto-detects and adds to active-tasks.md

User: "/dashboard lucas"
β†’ Executes dashboard generation command

Characteristics:

  • βœ… Most flexible and natural
  • βœ… Supports natural language understanding
  • βœ… Can invoke commands explicitly or implicitly
  • βœ… Interactive - Claude can ask clarifying questions
  • ⚠️ Requires user to be present and engaged

Direct Manual Input

β–Ό
Aliases: File Editing, Direct File Manipulation, Manual File Creation
Definition: User directly creates or edits files in the workspace using a text editor, without going through the chat interface.

Examples:

Example 1: Direct Task Entry
β†’ User opens: Personal-Productivity/daily/active-tasks.md
β†’ User manually adds: "| Review Dylan's homework | 2025-11-21 | 30min | Not Started |"
β†’ Saves file
β†’ Task now visible to Claude and dashboard systems

Example 2: Manual Log Entry
β†’ User opens: Homeschool/learning-log.md
β†’ User types: "**Nov 20, 2025** | Lucas solved fraction problem independently"
β†’ Saves file
β†’ Entry available for weekly review generation

Best Practices:

  • Follow existing file format/structure
  • Use consistent date formats (YYYY-MM-DD)
  • Maintain markdown table alignment
  • Add timestamps to log entries
  • Save changes frequently

Automated Input

β–Ό
Aliases: Automated Data Entry, System-Generated Input, Triggered Updates
Definition: System-generated data created automatically by scheduled tasks, triggers, or integrated tools without direct user intervention.

Types of Automation:

1. Time-Based Triggers:
   - Daily planning reminder (6 AM)
   - Weekly review reminder (Sunday 6 PM)
   - End-of-month goal review

2. Event-Based Triggers:
   - Task completion detection
   - File modification triggers
   - Threshold alerts (e.g., goal milestone reached)

3. Integration-Based:
   - MCP server data imports
   - External tool synchronization
   - API data feeds

4. Skill-Based Automation:
   - task-manager auto-detection
   - practice-logger automatic routing
   - progress-tracker calculations

πŸ”‘ Key Terms & Concepts

β–Ό

Core terminology used throughout the Claude Code workspace ecosystem.

Global

β–Ό
Aliases: Workspace-Level, System-Wide, Universal
Definition: Settings, configurations, commands, or skills that apply across the entire workspace, affecting all projects.

Examples:

Global Skills Location: e:\Lucas\skills/
- task-manager/ - Works across all 7 projects
- data-visualizer/ - Generates dashboards for any project
- practice-logger/ - Logs athletic activities
- meal-planner/ - Plans meals for Seth and Maria
β†’ These skills can be invoked from any project context

Project-Level

β–Ό
Aliases: Local, Project-Specific, Scoped
Definition: Settings, commands, or configurations specific to an individual project folder, not affecting other projects.

Examples:

Project-Specific Commands:

Location: Homeschool/.claude/commands/
- daily-planning.md β†’ Creates lesson plans (Homeschool-specific)
- quick-note.md β†’ Logs learning observations
β†’ These commands only work within Homeschool context

Location: Personal-Productivity/.claude/commands/
- daily-planning.md β†’ Creates daily task plans (different from Homeschool!)
- quick-capture.md β†’ Captures tasks/ideas
β†’ Same command name, different implementation per project

Agent

β–Ό
Aliases: AI Agent, Autonomous Agent, Primary Agent, Main Agent
Definition: An autonomous AI entity (instance of Claude) that can perform tasks, make decisions, and interact with files/tools independently within defined parameters.

Agent Capabilities:

  • File operations: Read, write, edit any workspace file
  • Tool usage: Bash commands, glob patterns, grep searches
  • Skill invocation: Activate task-manager, data-visualizer, etc.
  • Command execution: Run /daily-planning, /dashboard, etc.
  • Sub-agent spawning: Delegate complex tasks to specialized sub-agents
  • Context management: Remember conversation history and user preferences

Sub-Agent

β–Ό
Aliases: Specialized Agent, Task Agent, Spawned Agent, Delegate Agent
Definition: A specialized, temporary AI instance spawned by the main agent to handle specific complex tasks autonomously, then report back.

Sub-Agent Types:

| Type           | Purpose                  | Use Case                    |
|----------------|--------------------------|------------------------------|
| Explore        | Fast codebase exploration| Finding files, searching     |
| Plan           | Task planning            | Breaking down complex tasks  |
| General-Purpose| Multi-step tasks         | Complex workflows, research  |
| Custom         | Specific domains         | Dashboard gen, data analysis |

Custom Commands

β–Ό
Aliases: Slash Commands, Commands, Custom Slash Commands
Definition: User-defined shortcuts (slash commands) that execute predefined workflows when invoked in the chat interface.

Your Available Commands:

Productivity (3):
- /daily-planning - Generate today's plan with context
- /weekly-review - Comprehensive weekly review
- /quick-capture - Fast task/idea capture

Homeschool (5):
- /daily-planning - Create lesson plan (Homeschool context)
- /weekly-review - Homeschool accomplishment summary
- /quick-note - Log learning observations
- /log-activity - Internal activity logging
- /log-summary - Generate activity summary

Dashboard (1 command, 7 variants):
- /dashboard [person] - Generate dashboards
  (/dashboard lucas, dylan, maxi, seth, maria, family, productivity)

MCP (Model Context Protocol)

β–Ό
Full Name: Model Context Protocol
Definition: A standardized protocol that allows Claude to connect to external tools, databases, and services to extend its capabilities beyond file operations.

Potential Future Uses:

  • Fitness tracker integration: Auto-import workout data for Seth/Maria
  • School portal sync: Pull Dylan's grades automatically
  • Weather service: Check conditions for outdoor practices
  • Shopping API: Track prices for baseball equipment
  • Calendar sync: Integrate with Google Calendar for family events

Current Status:

⚠️ Not configured in workspace (all data currently file-based)

Skill

β–Ό
Aliases: Agent Skill, Capability, Auto-Skill
Definition: A reusable, context-aware capability that activates automatically or on-demand to perform specific functions across projects.

Your Available Skills:

Global Skills (8):
1. task-manager - Auto-detects and manages tasks
2. data-visualizer - Generates charts and dashboards
3. progress-tracker - Tracks goals and milestones
4. practice-logger - Logs athletic practices
5. meal-planner - Plans meals for Seth and Maria
6. weekly-family-report - Compiles family-wide summaries
7. meeting-notes - Structures meeting documentation
8. email-responder - Drafts email responses

Homeschool Skills (1):
1. update-calendar - Manages homeschool calendar

Key Difference from Commands:

Skills activate automatically based on context; commands require explicit /command invocation.

Hierarchy

β–Ό
Aliases: Structure, File Hierarchy, Project Structure, Organizational Hierarchy
Definition: The organizational structure of files, folders, projects, and configurations in your workspace, defining scope and relationships.

Hierarchy Principles:

  • Global > Project: Global settings apply everywhere unless overridden
  • Specificity wins: Project-level command takes precedence over global
  • Inheritance: Projects inherit global skills, can add project-specific skills
  • Scope boundaries: Project commands only work within that project

Logging

β–Ό
Aliases: Activity Logging, Event Logging, Data Logging, Tracking
Definition: The systematic recording of activities, events, metrics, and changes for tracking, analysis, and historical reference.

Types of Logging:

  • Activity Logs: Track actions and events as they occur
  • Productivity Logs: Track tasks, focus, energy, and time
  • Performance Logs: Track metrics and progress over time
  • Learning Logs: Capture observations and insights

Best Practices:

  • Timestamps: Always include date and time
  • Categorization: Tag entries for filtering
  • Context: Include relevant details
  • Consistency: Follow format conventions
  • Automation: Automate when possible

Context

β–Ό
Aliases: Conversation Context, Working Context, State, Session Context
Definition: The accumulated information, conversation history, file contents, and project state that Claude uses to understand and respond to requests.

Components of Context:

  • Conversation History: Previous messages in current chat session
  • File Contents: Files Claude has read in this session
  • Project State: Current working directory and project context
  • Workspace Structure: Knowledge of file locations and organization
  • Active Skills and Commands: Available capabilities in current context

Maintaining Good Context:

  • Start new chats for new topics
  • Refresh context when needed ("Re-read Lucas's profile")
  • Be specific about project
  • Reference previous context explicitly

Agentic System

β–Ό
Aliases: Autonomous System, Multi-Agent System, AI Agent Framework
Definition: An AI-powered system where autonomous agents can perceive their environment, make decisions, take actions, and achieve goals with minimal human intervention.

Agentic Characteristics:

  • Autonomy: Agents make decisions without constant human input
  • Perception: Agents sense their environment (read files, detect patterns)
  • Action: Agents take actions to achieve goals
  • Goal-Oriented: Agents work toward defined objectives
  • Learning/Adaptation: System improves through patterns and feedback

Your Workspace as Agentic System:

USER (provides goals)
    ↓
MAIN AGENT (Claude - orchestrates)
    ↓
    β”œβ†’ SKILLS (auto-activate capabilities)
    β”œβ†’ COMMANDS (execute workflows)
    β”œβ†’ SUB-AGENTS (handle complex tasks)
    β””β†’ FILE SYSTEM (reads/writes data)

πŸ—οΈ System Architecture

β–Ό

Visual representation of how components interact in your workspace.

Component Interaction Flow:

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                          USER                               β”‚
β”‚  (Provides goals, instructions, data)                       β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
             β”‚
             ↓ (via 3 input methods)
β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                    INPUT METHODS                            β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚  Manual Chat  β”‚  Direct File Edit  β”‚  Automated Triggers   β”‚
β”‚  "/dashboard" β”‚  Edit profile.md   β”‚  6 AM daily reminder  β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
         β”‚               β”‚                    β”‚
         ↓               ↓                    ↓
β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                      MAIN AGENT (Claude)                     β”‚
β”‚  β€’ Processes input                                           β”‚
β”‚  β€’ Maintains context                                         β”‚
β”‚  β€’ Orchestrates workflows                                    β”‚
β”‚  β€’ Makes decisions                                           β”‚
β””β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
      β”‚          β”‚            β”‚              β”‚
      ↓          ↓            ↓              ↓
β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β” β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β” β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β” β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚ COMMANDS β”‚ β”‚ SKILLS  β”‚ β”‚ SUB-AGENTS β”‚ β”‚ FILE OPERATIONS  β”‚
β”‚          β”‚ β”‚         β”‚ β”‚            β”‚ β”‚                  β”‚
β”‚ /daily-  β”‚ β”‚ task-   β”‚ β”‚ Explore    β”‚ β”‚ Read profiles    β”‚
β”‚ planning β”‚ β”‚ manager β”‚ β”‚ agent      β”‚ β”‚ Write logs       β”‚
β”‚          β”‚ β”‚         β”‚ β”‚            β”‚ β”‚ Update stats     β”‚
β”‚ /weekly- β”‚ β”‚ data-   β”‚ β”‚ Analysis   β”‚ β”‚ Generate         β”‚
β”‚ review   β”‚ β”‚ visual- β”‚ β”‚ agent      β”‚ β”‚ dashboards       β”‚
β”‚          β”‚ β”‚ izer    β”‚ β”‚            β”‚ β”‚                  β”‚
β”‚ /dash-   β”‚ β”‚ practiceβ”‚ β”‚ Custom     β”‚ β”‚ Edit trackers    β”‚
β”‚ board    β”‚ β”‚ logger  β”‚ β”‚ agents     β”‚ β”‚                  β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜ β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜ β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜ β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

πŸ’‘ Examples & Use Cases

β–Ό

Real-world examples demonstrating how the system works.

Example 1: Cross-Project Task Routing

β–Ό
Scenario: User mentions multiple tasks in different project domains

User Input:

"I need to plan next week's lessons for Lucas, schedule his baseball practice,
and create meal plans for Seth and Maria"

System Behavior:

Main Agent processes message
    ↓
task-manager skill auto-activates (detects "I need to")
    ↓
Parses 3 distinct tasks:
    β”‚
    β”œβ”€ Task 1: "Plan next week's lessons for Lucas"
    β”‚  Keywords: "lessons", "Lucas" β†’ Homeschool context
    β”‚  Routes to: Homeschool tasks
    β”‚
    β”œβ”€ Task 2: "Schedule his baseball practice"
    β”‚  Keywords: "baseball", "practice" β†’ Athletics context
    β”‚  Routes to: Lucas-Athletics tasks + calendar
    β”‚
    └─ Task 3: "Create meal plans for Seth and Maria"
       Keywords: "meal plans", "Seth and Maria" β†’ Health context
       Routes to: Health-Seth and Health-Maria tasks

βœ“ Single input β†’ 3 tasks β†’ Routed to 3 different projects automatically!

Example 2: Automated Dashboard Update Pipeline

β–Ό
Scenario: Practice logged β†’ Automatic chain reaction of updates

User Input:

"Lucas had practice today: 80 swings, 72% contact rate, worked on follow-through"

System Behavior:

practice-logger skill auto-activates
    ↓
Creates practice log β†’ practice-logs/2025-11-20-practice.md
    ↓
Updates batting stats β†’ metrics/batting-stats.md (new row: 72%)
    ↓
Triggers progress-tracker β†’ Recalculates progress (65% β†’ 72% = 90% to goal!)
    ↓
Triggers data-visualizer β†’ Regenerates lucas-dashboard.html
    ↓
Updates calendar β†’ 2025-11-baseball-schedule.md (βœ“ completed)
    ↓
Confirms to user:
    "βœ“ Practice logged for Lucas
     πŸ“Š 72% contact rate - New personal best! (+7% improvement)
     🎯 90% of way to 80% goal
     πŸ“ˆ Dashboard updated with latest data"

βœ“ One practice entry β†’ 5 files updated β†’ Dashboard regenerated β†’ Insights calculated!