Observed arrival · 2026-10-12
AIGuard puts an approval checkpoint between AI agents and shell commands
AIGuard is an open-source command guardrail for Kimi Code CLI and Claude Code, with decisions to allow, ask for approval, or block.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Developers using Kimi Code CLI or Claude Code
- Worth noticing
- Unrecognized commands default to ask; absolute blocks cannot be overridden, even by user policy.
Field notes
The homepage describes a local policy file at ~/.aiguard/policy.toml, with rules evaluated in order and user allow rules unable to override an Absolute Block. It also says command decisions are appended to ~/.aiguard/audit.jsonl, with restrictive file permissions. The site labels the tool beta and explicitly warns that it is not a sandbox or security boundary.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue