Skills
A skill is a folder with one SKILL.md. Type /name in a chat, or let the model load it on its own.
Shipped with Exe AI
Six skills, ready to use.
They load on their own when the task fits. You can also call any of them by name.
Inspects a folder before removing unnecessary files.
Schedules future actions: one-off reminders ("remind me in 10 minutes", "Friday at nine"), recurring jobs, scheduled scripts, or monitoring that reports only when the specified event occurs.
Examines a plan through focused questions until its decisions are settled.
Summarizes the conversation so another session can continue the work.
Explains Exe AI using its own manual: where settings are located, what controls and indicators mean, and why the program behaves as it does. Use for questions about the program, not the user’s work.
Generates an image from the rest of the user’s message without asking follow-up questions.
Write your own
One folder. One file.
A short header says what the skill does and when to use it. Everything below it is the instruction the model follows.
/skills in the CLI.--- description: Summarizes the conversation so another session can continue the work. auto: true --- Write a handover for whoever continues this work, including a future session with no access to this conversation. Review the conversation from the beginning. Do not rely on a general recollection; seemingly minor details may be essential. Include the following sections where relevant, omitting any empty headings: - **Context** — explain the purpose of the work in one or two sentences. - **Still open** — identify unfinished work, unanswered questions, and external dependencies. - **Decisions** — record settled choices that must not be reopened, with a brief reason for each. Without that reason, a later reader may reverse a decision without understanding its basis. …