Fordel StudiosAnthropic brings agentic plug-ins to Cowork
What Happened
The company says you can use plug-ins to "tell Claude how you like work done, which tools and data to pull from, how to handle critical workflows, and what slash commands to expose so your team gets more consistent outcomes."
Our Take
Honestly? Plug-ins are table stakes now. Everyone's got them. What matters is whether Anthropic's play actually solves the "agents hallucinating 90% of the time" problem or if it's just a nicer UX for the same mess. The "tell Claude how you like work done" angle is smart — workflow encoding beats prompt prompting every time. But I want to see whether teams actually lock down consistent outcomes or if this just becomes another way to be consistently wrong. The slash commands thing is fine, but unless they've built some kind of verification loop, this is still agents making shit up faster.
What To Do
Test Cowork's plug-in accuracy on your most critical workflow before rolling it out to the team.
Cited By
React