Observed arrival · 2026-09-27
Foo Guard: a deterministic security gate for AI agents
Foo Guard scans AI-agent configurations and MCP-style tool definitions for risky permissions, exposed credentials, and other security findings, then can enforce policy in GitHub and CI.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Teams reviewing AI-agent configs in GitHub or CI
- Worth noticing
- More than 23 rules assess identity, credentials, permissions, tools, autonomy, and auditability; findings and grades are not LLM-generated.
Field notes
The analyzer presents three sample configurations, ranging from a low-risk FAQ assistant to a deliberately vulnerable agent, alongside JSON and YAML input options. For CI policy, the page describes failOn and minGrade settings, baseline comparisons, and a rule that the strictest policy source wins. The site labels the product beta; its listed plan features are described by the site, not independently verified here.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
$PaidCommerce or pricing visible
✦PrettyNotable craft visible
◎NicheUnusually specific use
ƒJavaScriptBrowser-side code central
One card from the complete issue