Semantic Kernel transitions to Microsoft Agent Framework
What changed: Semantic Kernel transitions to Microsoft Agent Framework 1.0, which Microsoft now identifies as its enterprise-ready successor.
Why it matters: Teams with Semantic Kernel builds need to establish whether the successor preserves their integrations and deployment assumptions. The same operational widening appears elsewhere today: Augment Code now centers Cosmos across the software delivery lifecycle, while Coarena adds pre-action intent registration plus verification and guard APIs for safety evaluation.
Do this: Inventory each Semantic Kernel workload’s dependencies and compare them against the Microsoft Agent Framework 1.0 migration path before committing new framework work.
See what changed on Semantic Kernel's listing →
What changed on the tools' own sites
|
Augment Code repositions around Cosmos as an always-on software factory Augment Code now centers its product on Cosmos for coordinating agents across software delivery, monitoring, and incident response.
Matters if you manage agent workflows from tickets through production incidents.
|
|
Coarena expands from computer-use arena into an agent safety evaluation platform Coarena added structured safety evaluations with intent registration, verification APIs, practice runs, MCP tools, public results, and a local starter.
Matters if you evaluate agents before they take actions in external systems.
|
|
Pilot Protocol reframes itself as an open network for AI agents and launches… Pilot Protocol redesigned its product and documentation around an open agent network with identities, encrypted tunnels, trust handshakes, and access tools.
Matters if you build agents that exchange data through trusted peer connections.
|
|
WebRun expands from browser automation infrastructure into a business AI… WebRun now presents itself as a business automation platform for scheduled agent workflows across existing tools.
Matters if you schedule agents to automate recurring work across business software.
|
Changes we spotted on each tool's own website since the last issue.
Biggest movers this week
Listings with the largest week-over-week jump in outbound-click interest
- Claude Code · +18.5% week-over-week. Claude Code is Anthropic's agentic coding tool that lives in your terminal. It understands your entire codebase, makes multi-file edits, runs commands, manages git workflows, and uses MCP for tool integration. Built with a Unix philosophy — it reads, plans, edits, and verifies in a loop. The fastest-growing product in the coding agent category.
- Sinatra · +17.2% week-over-week. Sinatra is a cloud coding agent that picks up assigned Linear tickets or labeled GitHub issues, writes code, runs tests, and opens a draft pull request. It works through the tools engineering teams already use and can revise the same PR in response to review comments.
- Orbi · +12.3% week-over-week. Orbi is an open-source, self-hostable AI coding agent that takes GitHub Issues from implementation through independent review, merge, and release. It can run against your own model endpoint or through Orbi’s managed cloud service.
|