Agent Skills

An Agent Skill is a named capability with its own instructions that you teach an Agent. Think of it as training material: you write the instructions once, then attach the same Skill to as many Agents as you need.

Caution

GreenVee has two different objects both called “Skill”. This page is about Agent Skills — capabilities for AI Agents. There are also HR Skills in the People module, which are attached to Employees and Project Members — see Skills (People module). They share a word but are completely separate objects. Changing an Agent Skill does not affect HR Skills, and vice versa.

Agent Skill fields

Field	What it does
Name	A short label for the capability (e.g. “Summarize meeting notes”)
Description	A note for your team — when this Skill is useful
Instructions	The detailed guidance the Agent follows when using this Skill
Status	Whether the Skill is draft, active, or archived
Trigger	How the Skill activates — see below
Run mode	What happens once the Skill activates — see below
Visibility	Who can attach this Skill to an Agent: private, team, or org
Use in Polaris	An opt-in switch — when on, GreenVee’s shared Polaris assistant can also use this Skill

Only active Skills are available to be attached and used; draft and archived Skills are hidden from an Agent’s day-to-day behaviour.

Trigger: how a Skill activates

Trigger	Meaning
Auto-detect	The Agent decides on its own, based on the conversation, when this Skill is relevant.
Slash command	The Skill only runs when explicitly invoked, e.g. by typing a command.
Mention only	The Skill only comes into play when it’s directly referenced.

Run mode: what happens when it activates

Run mode	Meaning
Reply	The Skill’s instructions shape the Agent’s answer — it doesn’t take any action on its own.
Auto-run	The Agent carries out the Skill’s instructions immediately, without asking first.
Approval	The Agent asks you to confirm before carrying out the Skill’s instructions.

What good Skill instructions look like

The Instructions field is the most important part. This is what the Agent actually reads and follows. Write instructions as if you were briefing a new colleague:

• Be specific about the task (“When given a list of project updates, produce a three-sentence summary covering status, blockers, and next steps.”)
• Say what to include and what to leave out
• Give examples if the output format matters

Vague instructions produce vague results. The more precise you are, the more consistent the Agent’s output will be.

Reuse across Agents

One of the key benefits of Agent Skills is reuse. You do not have to write the same instructions twice.

For example, if you have a “Summarize meeting notes” Skill, you can attach it to:

• A project coordinator Agent
• A client communication Agent
• An executive briefing Agent

Each Agent gets the same summarization behaviour. When you update the Skill’s instructions, all Agents that use it pick up the change.

Skills vs. Agents

	Agent Skill	Agent
What it is	A piece of know-how	A configured worker
Has instructions?	Yes — specific to the capability	Yes — general, always-on
Reusable?	Yes — attach to many Agents	Each Agent is its own configuration
Does work on its own?	No	Yes — when active

A Skill never does work on its own. It only comes into play when an Agent that has been given that Skill decides to use it.

Related pages

• AI Glossary — plain-English definition of Agent Skill vs. HR Skill
• Agents — how to attach Skills to an Agent
• Polaris — the shared assistant that can use Skills marked “Use in Polaris”
• Skills (People module) — HR Skills used for Employees and Project Members