We shipped a new batch of skills and agents this week. The catalog grows as we find more work worth packaging.
The focus this round was on the tasks people run most often. Research helpers that fan out across sources and come back with cited answers. Content skills that draft in your voice instead of a generic one. A few agents that lean on MCP servers to reach real data, not just text.
Every entry goes through the same bar. It has to save real time on a task you actually do, and it has to be honest about what it needs to run. If an agent depends on an MCP server, the runbook says so and walks you through setup. New here? Start with what is a skill.
Check the catalog for the latest. If something is close to what you want but not quite right, that is useful signal. Tell us, and it helps shape what gets built next.