Run and learn ywai: MCP, model profiles, agent groups, doc tours. Trigger: ywai mcp, ywai profile, /learn-ywai.
71
89%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Passed
No findings from the security scan
The human does not run these commands. You do.
Two jobs, one skill:
ywai … yourself. Do not paste the command for the user to copy.core. It stays installed.ywai mcp list
ywai mcp enable <id>
ywai mcp disable <id>
ywai mcp auth <id>disable keeps the config. auth is OAuth (figma, github, gitlab, or any catalog entry with AuthType=oauth).
ywai profile list
ywai profile use <name>list marks the active profile with *. Shipped names:
| Name | Effect |
|---|---|
inherit | No pinned models. Agents use the session/lead model. |
balanced | Cost/quality mix. DeepSeek v4 pro/flash on the former Grok/Minimax slots. |
fast | Flash model on every agent. |
deep | Top models for design/review. |
ywai groups
ywai groups enable <name>
ywai groups disable <name>Use the names from ywai groups. core is always on.
The command printed success (MCP … enabled, Active profile: …, Group … enabled) or you reported the error verbatim.
Reach for this when the user wants to learn, not mutate.
references/docs/ inside this skill. Every .mdx there is readable — index pages, guides, agent pages, all of them. Do not invent commands, flags, or agent names, and do not fetch the website.teach for courses. When the user asks for a long course, structured lessons, or the tour outgrows chat, load the teach skill and build its workspace (MISSION.md, HTML lessons). Use the references/docs/ pages as the source material there too. For a one-off question, stay in chat.references/docs/, plus any page the topic requires (e.g. the named agent page after agents/index.mdx).@agent examples from the page).If a page is missing or the read fails, say so and teach only what you successfully read.
/learn-ywai
/learn-ywai agentesYoizenSA/dev-ai-workflow
36f5af5
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.