Agent Safety Articles

Page 1 of 1 (4 articles)

8/10/2026 • EN

Build a Safe Repository Maintenance Agent with GitHub Copilot and Microsoft Agent Framework in Python

Build a safe repository maintenance agent in Python using GitHub Copilot and Microsoft Agent Framework, with explicit approval for writes and shell commands.

Agent Safety Github Copilot Microsoft Agent Framework Python Repository Maintenance

Sahan Serasinghe

7/10/2026 • EN

Ensuring Agent Safety in AI Development

Overview of Microsoft's best practices for ensuring AI agent safety, including data security and prompt injection prevention.

Agent Safety ai development Data Security Microsoft Learn prompt injection

Jesse Liberty

6/8/2026 • EN

Appropriateness is what safety cannot mechanise

Explores limitations of safety checks in AI agents, focusing on contextual harm that passes all structural gates.

Agent Safety AI Safety Contextual Integrity Harm Prevention Product Safety

Jonathan Kingston

11/4/2025 • EN

MCP Colors: Systematically deal with prompt injection risk

A method using color-coding (red/blue) to classify MCP tools and systematically mitigate prompt injection risks in AI agents.

Agent Safety ai security mcp prompt injection Tool Classification

Simon Willison

Agent Safety Articles

Build a Safe Repository Maintenance Agent with GitHub Copilot and Microsoft Agent Framework in Python

Ensuring Agent Safety in AI Development

Appropriateness is what safety cannot mechanise

MCP Colors: Systematically deal with prompt injection risk

Select Language

We use cookies