Signal Scout

Research

Dated write-ups from our weekly observation of public AI projects. Each one is a single measurement, with its method and limits stated.

September 10, 2026

AI coding: common, fragmented, and some security improvements

A third of AI projects are set up for a coding agent. Those projects are about twice as likely to show basic security practices, and AI-specific protections are rare everywhere.

September 9, 2026

The AI security gap

How many public repositories building AI agents or MCP servers carry any AI security tooling, with the coverage audit and its limits.

September 7, 2026

Models with the safety removed or reduced

Open-weight models on HuggingFace with their refusal behavior deliberately removed or reduced, and how often they are downloaded.