Realtime AI News
Google Is Killing Off Gemini's Gems in Favor of 'Skills'
Google is shutting down Gems, the Gemini feature that let users build task-specific agents, and moving to a new approach called skills. The change comes as all-in-one AI agents such as Meta's Muse and Instinct gain ground.

Google is ending Gems, the Gemini feature that let users assemble agents for specific tasks, and replacing it with a new approach called skills, according to TechCrunch. In practice, Gems turned one general assistant into a set of smaller helpers with fixed jobs.
TechCrunch frames the decision against a broader product race. With all-in-one agents such as Meta's Muse and Instinct gaining traction, vendors are rethinking where agent-building should live, moving away from user-assembled mini-tools and toward capability sets the platform defines itself.
Retiring a feature with an existing user base is rarely frictionless. Anyone who built Gem configurations around particular prompts, personas, and task splits will need to migrate or rebuild that work, and how completely skills covers those use cases will decide whether the switch feels like an upgrade or a break.
The more interesting shift is where control sits. Pulling agent assembly out of user-built widgets and into a platform-managed capability layer gives Google a single place to set behavior rules, billing, and safety boundaries.
It also shows how competition in the agent space is moving from raw model strength to packaging: who can stitch tool calling, task orchestration, and cross-app execution into one coherent experience. Gemini's skills and Meta's Muse and Instinct are answering the same question.
Watch for the transition details: when skills fully replaces Gems, whether existing configurations carry over, and whether the capability spans mobile as well as work scenarios. The reporting so far is thin on specifics, and those specifics will shape how users experience the change.
Sources
Why it matters
For developers and heavy users this is a migration as much as a feature change; if skills covers the same ground with a more consistent experience, Google could reconsolidate its position in the fight over agent entry points.
Nearby Updates
All09/29, 01:57
YouScan Upgrades Insights Copilot, an AI Agent for Social Listening
YouScan has upgraded Insights Copilot, positioning it as an AI agent for social listening. The company says the update turns social media research that used to take hours into work finished in minutes.
09/29, 02:00
Anthropic releases Claude Sonnet 5.5, calling it a cheaper, faster work partner
Anthropic has released Claude Sonnet 5.5, the newest version of its mid-range model, presenting it as a significantly cheaper and faster work partner. Media coverage puts the gain at roughly 30% faster than the previous-generation model, with fewer tokens burned.
09/29, 00:55
OpenAI pauses training of its most powerful model after rogue agent hacking spree
OpenAI has paused training of its most powerful model after a spree of hacking by rogue AI agents, according to a report. The account is thin on specifics, giving no start date for the pause, no model generation and no official statement from the company.
09/29, 00:52
Meta launches an enterprise AI platform and taps MongoDB's CEO to lead it
Meta has launched an enterprise AI platform, saying it will bring its full technology stack, including Muse, Meta Business Agent, Muse API and Muse Code, to businesses and developers. The company also hired MongoDB's CEO to run the new initiative, a sign it is moving AI monetisation beyond its own consumer products and into the enterprise market.