Codex sandbox explained: modes, boundaries and approvals
Understand what the Codex sandbox isolates, how its modes differ from approval policy, and what files, writes and network access still require review.
Read articleBrowse by category
AI agents & context
Understand how agent harnesses, context, sandboxes and permissions shape what AI agents can read, change and use.
In this category
AI agents & context guides and articles.
Every published resource filed under AI agents & context. The foundational guide appears first where one applies.
6 articles
Codex permissions and workspace access explained
Learn how Codex permissions control local file and network access, how profiles differ from legacy sandbox settings, and where document checks fit.
Read article
Claude Code sandbox explained
Understand how the Claude Code sandbox constrains Bash, how auto-allow differs from permissions, and which file and network exposures still need attention.
Read articleClaude Code permissions and workspace access explained
Learn how Claude Code permissions, modes and working-directory rules govern file access, and how those controls complement its Bash sandbox.
Read article
Context engineering for AI agents: a practical guide
Learn how to engineer context for AI agents across instructions, tools, retrieved knowledge, memory, history, limits and evaluation.
Read articleWhat is an AI agent harness? Components and controls
Learn what an AI agent harness is, how its runtime components fit together, and where permissions, observability, context and safety controls belong.
Read article