Skills

A skill is a SKILL.md file: front matter, then instructions the model reads when it needs them. Orrery takes the agentskills.io format unchanged, so a skill written for another agent works here as it is.

#Add one

Put each skill in its own folder:

---
name: review-checklist
description: The checklist to run before calling a change done.
---
1. The tests for the change pass.
2. ...

Or type /skills add in the TUI. The model sees a new skill from the next message.

#Check them

orrery skills list                    # every skill found, and what its scripts may do
orrery skills list --agent reviewer   # only those this agent may load
orrery skills show review-checklist   # its front matter and body

#Scope and scripts

Two settings, both in configuration, never in SKILL.md:

[skills.review-checklist]
scope = ["main", "reviewer"]   # which agents may load it; empty means all

• scope limits which agents see the skill.
• grant says what the skill's scripts/ may do. Without one, its scripts do not run at all.

A skill cannot ask for privileges. It gets them from you, or it has none. A skill's scripts run as agent:skill:<name>, under the main agent's rules, so a skill is never wider than the agent that loaded it.

#In a profile

[profile.review]
skills = ["review-checklist"]