Threat explainer

Prompt injection

How hidden instructions trick coding agents into unsafe work.

A prompt injection is a trick note for an AI tool.

Model context can carry hostile instructions from files, pages, issues, or package metadata into privileged tool calls.

HOL Guard turns these moments into private receipts first, then public lessons only after redaction and moderation.

Harness setup guides

Protect the coding tools your team already uses without forcing everyone to become a security expert.

harness

Terminal-native coding agent with broad shell reach.

harness

Agentic coding harness with MCP and file access.

harness

IDE and CLI assistant across code and terminal flows.

harness

AI-first IDE with repo and terminal context.

Real protection moments, scrubbed for safety before becoming public learning pages.

codex

A redacted example of a local secret read attempt.

Practice attack patterns with static simulations. Nothing dangerous executes.

prompt-injection

See how hidden text tries to override your instructions.