Glossary

The glossary defines each term Athenode uses in the web app, the Athenode CLI and the documentation in one sentence and links to the page that covers it.

Terms for the product and its tools

Term	Definition	Covered in
Athenode	Athenode is a spec-driven development platform where you think a product through before your AI coding agent writes code.	Athenode documentation
spec-driven development (SDD)	Spec-driven development is a way of building software in which the work is written down as a specification before an AI coding agent writes the code.	The spec-driven development guide
web app	The web app is the Athenode interface in your browser, where you work with projects, specifications, setups, ToDo cards, members and the plan.	Core concepts
Athenode website	The Athenode website is athenode.com, which holds the documentation and the public pages of the setup library.	The setup library catalogue
Athenode CLI	The Athenode CLI is the command-line tool athenode from the package @athenode/cli, which connects a directory to a project, installs a setup and reads and changes the project's content.	Athenode CLI reference
Athenode MCP server	The Athenode MCP server is the server named athenode through which an AI agent reads and changes specifications, ToDo cards and setups.	Athenode MCP server
AI coding tool, AI tool	An AI coding tool is the product you write code with, such as Claude Code, Codex or Cursor.	Supported AI tools
AI coding agent, AI agent	An AI coding agent is what does the work in a session of your AI tool, from asking questions to implementing a specification.	The Athenode workflow

Terms for projects, members and plans

Term	Definition	Covered in
project	A project is the container you create in the web app for one specification tree, its agent setups, ToDo cards, members, project tokens and plan.	Projects
member	A member is a person with access to a project, in the role of owner, editor or viewer.	Invite and manage members
owner	The owner is the member who created the project and the one who invites members, changes the plan and deletes specifications or the project.	Roles and permissions
editor	An editor is a member who creates and changes specifications, setups and ToDo cards and can create project tokens.	The role matrix
viewer	A viewer is a member who can read everything in a project and cannot change its content.	What each role can do
invite	An invite gives a person access to a project by email, as an editor or a viewer.	Invites
plan	A plan is what a project is subscribed to (Free, Solo, Team or Business) and sets the project's limits.	Plans and limits
project token	A project token is a personal access token that lets the Athenode CLI, and your AI agent through it, act on one project.	Project tokens
local project directory	The local project directory is the folder on your computer, normally the repository root, where you run init and where .athenode/config.json is kept.	Install a setup

Terms for specifications

Term	Definition	Covered in
specification	A specification is one described piece of work, with a title, a summary, Markdown content, a status and its own questions and answers.	Specifications
specification tree	The specification tree is the hierarchy of a project's specifications, in which any specification can have sub-specifications to any depth.	The specification tree
sub-specification	A sub-specification is a specification that has another specification as its parent.	Specifications in the web app
leaf specification	A leaf specification is a specification with no sub-specifications, and leaves are what an AI agent implements.	Leaf specifications
deliverable	A deliverable is the word the web app uses for a leaf specification where it counts them, as in the Orbit view.	The Orbit view
specification id	A specification id identifies one specification in skills and commands, and Copy ID copies it in the web app.	Copying a specification id
status	A status says where a specification is in its life: Draft, Prepared, Processing or Completed.	Statuses
Draft	Draft is the status of a specification that is being shaped.	The four statuses
Prepared	Prepared is the status of a specification that is finished and ready to be decomposed or implemented.	What Prepared means
Processing	Processing is the status of a specification that an AI agent is planning or implementing.	Statuses during an apply run
Completed	Completed is the status of a specification that is implemented, and reviewed when review is on.	How statuses move
blocker	A blocker is a dependency between two specifications: a specification that is blocked by another is implemented after it.	Blockers
questions and answers	Questions and answers are the record of the clarifying questions an AI agent asked about a specification and the answers you gave.	Questions and answers
implementation plan	An implementation plan is the plan an AI agent records for a leaf specification before it implements the leaf.	The implementation plan
Tree view	The Tree view is the view of the Specifications page that lists the specification tree with a search box and a status filter.	Browsing the Tree view
Orbit view	The Orbit view is the view of the Specifications page that draws the specification tree as rings around a centre that shows how much is done.	Browsing the Orbit view
search by meaning	Search by meaning finds the specifications that match what a query means as well as those that contain its words.	Search by meaning

Terms for the workflow

Term	Definition	Covered in
mind map	A mind map is a session in which your AI agent asks rounds of questions and turns your answers into a specification.	Mind map: /atn-mindmap
decompose	To decompose a specification is to break it down into one level of stages, each a sub-specification of its own.	Decompose: /atn-decompose
stage	A stage is a sub-specification that decomposing a specification creates.	Stages of a decomposition
apply	To apply a specification is to have your AI agent plan and implement each leaf under it, in the order the blockers allow, with tests, review and git handled as your workflow settings say.	Apply: /atn-apply
apply run	An apply run is one execution of /atn-apply on a specification or a subtree.	An apply run
workflow settings	Workflow settings are a project's choices for gitflow, testing, code review and methodology, which the workflow skills follow.	Settings: /atn-settings
sub-project	A sub-project is a folder of a repository that has its own workflow settings.	Sub-projects in /atn-settings

Terms for agent setups

Term	Definition	Covered in
agent setup, setup	An agent setup is a named bundle of what your AI coding tools work with: skills, agents, rules, MCP servers and setup files.	Agent setups
standard Athenode setup	The standard Athenode setup is the setup with the workflow skills and their agents, of which every project starts with its own copy.	Standard Athenode setup
installed setup	The installed setup is the setup whose files are in a local project directory, chosen when you run init.	The installed setup
install	To install a setup is to have the Athenode CLI write it into your repository in the format each selected AI tool reads.	Install a setup
setup id	A setup id identifies one setup in CLI commands, and athenode setups list prints it with the setup's slug, which init --setup accepts as well.	The setups commands
skill	A skill is a reusable set of instructions that an AI tool loads when it is relevant, usually run as a slash command such as /atn-mindmap.	Skills in a setup
bundle files	Bundle files are the files a skill carries next to its instructions.	Bundle files of a skill
agent	An agent is a specialist that an AI tool can hand work to (a subagent in most AI tools).	Agents in a setup
rules	Rules are the instructions every agent in a setup follows.	Rules in a setup
Athenode rules block	The Athenode rules block is the part of a rules file such as AGENTS.md or CLAUDE.md, between two marker lines, where Athenode writes the rules of the installed setup.	The Athenode rules block
MCP server	An MCP server is a connection to an external tool server that the agents of a setup may use.	MCP servers in a setup
setup files	Setup files are the extra Markdown files stored with a setup, such as references that its skills and agents read.	Setup files and settings
settings files	Settings files are the setup files under settings/, which hold a project's workflow settings.	The settings files
clone	To clone a setup is to copy the whole setup under another name inside a project.	Create and organise setups
pull	To pull is to copy skills, agents, MCP servers, setup files or rules from another setup into a setup, once.	Pull from another setup
setup cards	Setup cards are the row at the top of the Setups page that shows one card for each setup of the project.	The Setups page
blueprint	The blueprint is the part of the Setups page that maps the rules, files, skills, agents and MCP servers of the selected setup.	The blueprint
inspector	The inspector is the panel of the Setups page that opens the part you select in the blueprint.	The inspector

Terms for the setup library

Term	Definition	Covered in
setup library	The setup library is the shared catalogue of agent setups that projects have published.	Setup library
library entry, entry	A library entry is a setup that a project published to the setup library, with a name, a description, tags, a publisher, versions and likes.	Library entries
add	To add an entry is to copy it into your project as a setup of its own.	Adding an entry
merge	To merge an entry is to bring its skills, agents, MCP servers, setup files and rules into a setup you have.	Merging an entry
update	To update a setup added from the library is to bring it to the latest version of its entry.	Updating from the library
publish	To publish a setup is to make it a library entry, or the next version of the entry it has.	Publish a setup
public display	Public display is the setting that shows a library entry on the Athenode website to people without an account.	Public display of an entry

Terms for ToDo cards

Term	Definition	Covered in
ToDo card, card	A ToDo card is a short note of work to do later: a follow-up, tech debt, a bug or an idea.	ToDo cards
card status	A card status is Open, Done or Dismissed, and every card starts as Open.	Card statuses
card type	A card type says what kind of work a card is: Bug, Tech debt, Follow-up or Idea.	Card types and priorities
open questions	Open questions are the decisions that have to be made before a card is carried out.	Open questions of a card
carry out directly	To carry out a card directly is to have your AI agent implement it without a specification.	Card queue: /atn-todo-dequeue