Verdent Docs
This is archived Verdent 1.0 documentation. For current product documentation, see Verdent.

Glossary of Terms

Technical terminology and definitions for Verdent features and concepts

Alphabetized reference of Verdent-specific terms and technical concepts.


Core Concepts

Agent

The main AI assistant that processes requests and executes tools

Context Exhaustion

When conversation history fills context window, limiting the ability to reference earlier content

Context Remaining

Indicator in the middle panel showing how much model context is left for the current conversation

Context Window

Amount of conversation history the AI can reference

MCP (Model Context Protocol)

System for extending Verdent with external tools

Project

A git repository you're actively working on. Projects maintain independent session history and configuration.

Subagent

Specialized AI agent with isolated context for specific tasks (e.g., Verifier)

Task

A specific piece of work within a workspace. Tasks represent individual conversation threads and share the workspace file state.

Tool

Function the agent can execute (file_read, bash, glob, etc.)


Execution Modes

Agent Mode

Full autonomous execution mode where Verdent can read, write, and execute commands without additional approval steps. Default mode for active development.

Plan Mode

Read-only planning mode where Verdent analyzes and creates implementation plans without making file modifications. Best for complex changes and unfamiliar codebases.


Desktop-Specific Terms

Base Workspace

The primary working directory on your chosen branch. Reference point for creating new worktrees, where rebased work ends up.

Git Worktree

Isolated working directory linked to the same repository, enabling parallel development without file conflicts

Parallel Agents

Multiple AI agents working simultaneously in separate worktrees, each with isolated context

Project Switching

Rapid navigation between different projects while maintaining session state

Rebase to main branch

Process of applying changes from a workspace back into the main branch (the branch the workspace was created from)

Sync with main branch

Pull latest changes from the main branch into the current workspace

In "Rebase to main" and "Sync from main," the branch name shown is actually the branch of the current Base Workspace, not necessarily master. If your Base Workspace is on a different branch (e.g., develop), the actions will target that branch instead.

Task Worktree

Isolated directory for a specific task with a complete copy of repository files and a separate branch


Files & Configuration

AGENTS.md

Project-specific team standards file (project root, version controlled)

mcp.json

MCP server configuration (~/.verdent/mcp.json)

VERDENT.md

Global personal preferences file (~/.verdent/VERDENT.md)

Workspace

A git worktree - an isolated working directory linked to the same repository, enabling parallel development


Tool-Specific Terms

Glob Pattern

File search pattern using wildcards (**/*.ts, src/**/*.js)

Line Range

Specific lines to read from large files (start_line, max_lines)

Regex

Regular expression for advanced text searching

Replace All

file_edit flag to replace multiple identical strings

Timeout

Maximum time for bash commands: 120 seconds (2 minutes, hard limit)


Subagent Terms

Flexible Policy

Subagent can activate automatically based on request patterns

Isolated Context

Subagent's separate context window (doesn't consume main context)

Strict Policy

Subagent only runs when explicitly @-mentioned

System Prompt

Instructions defining subagent's behavior and expertise


MCP Terms

Global Config

User-wide MCP configuration (~/.verdent/mcp.json)

MCP Server

External tool/service integrated via Model Context Protocol

Plugin

Pre-built MCP integration for specific tools or services


Performance Terms

Delegation

Routing tasks to subagents to preserve main context

Parallel Execution

Multiple subagents running simultaneously

Strategic Reading

Using line ranges and search to minimize context usage

Task Coordination

Main agent managing multiple subagent operations


See Also