Observed arrival · 2026-08-27
Aggrete Gives AI Assistants a Handbook-Shaped Conscience
Aggrete places a policy layer between workplace AI assistants and company systems, checking requests against an employee handbook before retrieving information.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
Field notes
The product translates handbook language into retrieval rules, illustrated by clause 7.1 and a refusal that cites the clause rather than returning a generic error. Its demo tracks requests such as “Q3 headcount plan” and “On-call gaps,” distinguishing allowed, warned, and refused combinations. The page says the system can sit across Claude, Microsoft Copilot, ChatGPT, and Cursor, with an audit trail recording decisions.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
$PaidCommerce or pricing visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue