AI workflow security scanner

Agentic Workflow Guard

Map every place a repository gives AI agents instructions, tools, secrets, or write power, then turn that map into findings, reports, and safer pull request checks.

GitHub npm Comparison Reports
AWGuard terminal demo showing inventory, score, migration, and graph reports

What It Scans

Agent Instructions

Finds AGENTS.md, Copilot instructions, custom agents, prompts, and reusable skills.

Automation Paths

Reviews GitHub Actions and other workflow files for unsafe agent execution boundaries.

MCP Trust

Flags unapproved MCP servers, package launches, command tools, and environment exposure.

Reports Built For Adoption

Inventory

--format inventory and inventory-json explain the agentic surface.

Risk Score

--format score gives teams a compact AWI score they can track over time.

Compare

--compare old.json new.json shows introduced and resolved findings between scans.

Copyable Launch Assets