The skills CLI now supports installing agent skills directly from Notion workspaces without needing a Git repository.

// 1h agoPRODUCT UPDATE

The skills CLI now supports installing agent skills directly from Notion workspaces without needing a Git repository.

The skills CLI has been updated to allow users to install AI agent skills directly from Notion pages and databases by running a simple command. This removes the previous requirement of using a Git repository, enabling developers to write, store, and manage their agent instructions in a more accessible and collaborative workspace.

// ANALYSIS

Allowing agent skills to be hosted in Notion significantly lowers the friction for creating and iterating on agent instructions.

–Eliminates the need for Git version control, speeding up the workflow for non-code text prompts.
–Makes it easier for non-technical team members to contribute to or manage AI agent behaviors in a familiar UI.
–Shifts the paradigm of agent customization toward no-code environments, which could foster a broader ecosystem of skill sharing.

// TAGS

aicoding-agentsnotionverceldeveloper-tools

DISCOVERED

1h ago

2026-09-17

PUBLISHED

1h ago

2026-09-17

RELEVANCE

8/ 10

AUTHOR

vercel_dev

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

LAUNCH1h ago

Anthropic opens Life Sciences Verification Program

Anthropic has opened applications for its Life Sciences Verification Program (LSVP), an initiative developed in partnership with the U.S. government to give qualified life sciences researchers access to its frontier models, including Claude Mythos. While standard frontier AI models enforce broad biosecurity safeguards that often trigger false-positive refusals on legitimate scientific tasks, the LSVP establishes a specialized verification tier designed to enable authentic biological and pharmaceutical research without compromising critical safety guardrails against dual-use risks.

OPEN SOURCE1h ago

NetEase Youdao open-sources R2T2 streaming ASR

NetEase Youdao has released R2T2 (Confucius4-R2T2), an open-source real-time streaming Automatic Speech Recognition (ASR) model engineered for voice agents with runtime context injection for domain terminology. Built on an LLM foundation, R2T2 commits text only when acoustic evidence is confident to prevent premature partials, and supports deployment via vLLM and Hugging Face Transformers.

INFRA1h ago

TypeSafe AI's Jev accelerates agent decision loops

Jev is a specialized "System 1" decision model developed by TypeSafe AI—founded by RLHF and ChatGPT co-inventor Diogo Almeida—that replaces slow, non-deterministic text generation with typed, probabilistic primitives such as Choice, Score, and Noul. Operating at sub-500ms latencies with calibrated confidence scores, Jev serves as a dedicated control layer for agentic workflows, highlighted by Cloudflare Principal Engineer Dillon Mulroy as an essential primitive for production-ready AI systems.