How to Audit MCP Servers and Third-Party Tools Connected to Your AI Coding Agent
Learn how to audit MCP servers and third-party AI coding tools: verify identities, permissions, OAuth tokens, tool definitions, data access and revocation.
SaveAI Coding Agents
Safe AI-Assisted Development: Permissions, Code Review and Deployment Approval Workflows
Learn safe AI-assisted development workflows: restrict agent permissions, review code independently, protect branches and require approval before deployment.
Save
AI Coding Agent Security Checklist: Protecting Secrets, Repositories and Production Data
Protect secrets, repositories and production data when using AI coding agents. Follow this 12-point checklist for safer agent permissions and workflows.
Save
PixelLeak: AI Coding Agents Exposed 13,000 Screenshots on Public GitHub Repositories
Glow Security researchers say AI coding agents exposed more than 13,000 internal screenshots linked to 343 organisations in public GitHub repositories.
Save
Make more of this page
Follow the connection between stories.
This topic brings together articles that share a subject, name or theme. The stories may cover different events and publication dates, so read each one in its original context rather than treating every headline as a current development.
Open an article for its explanation and source links. If you are researching a broader question, explore its main category as well: a tag is a useful connection, while a category gives you a wider view of the coverage.