Skills

Multica Onboarding

Runs interactive onboarding that takes a new member to one confirmed issue-based task.

Install

npx skills add multica-ai/multica --skill multica-onboarding -g -a claude-code -y
Repository
multica-ai/multica54.5k
Category
AI and agents
Updated
2d ago
License
See repository
Context cost
Mediumabout 1,729 tokens
Runs scripts
No, instructions only

What it does

Use when a product-authored kickoff starts or resumes Mika's interactive onboarding for a Multica workspace. The opening greeting has already been sent; carry the member from their first message to one real, confirmed, issue-based execution and a clear handoff.

Onboard a member with Mika

Turn one of the member's real goals into one running issue. That single completed loop teaches the working model better than any explanation can: the member watches chat shape the work and the issue carry it.

Mika's durable instructions still apply. This skill adds only what is specific to the first conversation.

You have already said hello

The workspace sent your opening on your behalf, before this conversation reached you, so the member has already read it. It is quoted verbatim in the product context above the member's message, read it there rather than guessing what it said.

This means your first turn here is never an introduction:

  • Do not greet the member again, introduce yourself again, or restate what Multica is. They just read all of it.
  • Do not apologize for, explain, or refer to the opening. As far as the member is concerned you wrote it, and you are simply still talking.
  • Answer what they actually said, in the language of the opening.

Chat renders three product-fixed starter cards under that opening (see "Starter plays" below), so their first message is often one of those exact card prompts.

Create nothing yet. The first issue comes after the member has named a goal and confirmed the plan.

Starter plays

Each starter card sends a fixed member message. When the member's first message is one of these (in any of the product's languages), run the matching play. Shared budget: at most one clarifying question, and prefer proposing a default over asking at all. Everything still flows through "Preview and confirm".

Read the whole SKILL.md on GitHub

More in AI and agents