Observed arrival · 2026-09-07
Cairn puts AI-agent actions on the record
A control layer that intercepts AI-agent tool calls, applies policy, routes risky actions to human approval, and records each verdict in a hash-chained ledger.
Field notes
Cairn places an interception point between an AI agent and its connected tools, presenting the agent with an ordinary MCP endpoint while inspecting the requested action. Its sample console exposes ledger, policy, approval, and range views; one displayed rule holds payments.transfer requests above 10,000 GBP for a treasury approver, while each ledger entry carries a link to the preceding hash. The page labels the material illustrative and says no live customer data is included.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue