Recall Gate is a cost-rationing safety monitor for cyber-capable agents: instead of running an expensive LLM monitor on everything an agent does, it reads the cheap system footprint of an action (crashes, restarts, refused connections) and escalates to the LLM only when that footprint is unfamiliar, remembering every verdict in MongoDB Atlas so monitoring gets cheaper over a session while a deterministic hard-signal override guarantees no real attack is ever waved through. The escalate-or-suppress decision runs as a single Atlas Vector Search aggregation (similarity plus a recency filter plus the suppress math in one server-side query), which is what turns the incident history into a live decision engine rather than a passive log.
More from The Harness Engineering & Model Wrangling Hackathon
Ripple designs circuit boards from plain English, grades them against hidden checks, and evolves its own harness from the failures, going from 0/4 to 3/4 on…

Offload learns the work you repeat, so next time you can hand it over. Its engine, REM (Replay · Evolve · Merge), is an agent harness with a day/night cycle…
Recursive Computer Use is a local-first desktop automation harness that enables AI agents to execute tasks on your computer, verify outcomes locally, and…

Da Vinci is a recursive improvement CAD harness designed to support physical engineering tasks through specialized AI agents, programmatic CAD and automated…

CyclopsDiary gives factory workers wearing Meta AI glasses a shared memory of the floor. Every worker's glasses feed one MongoDB memory that keeps what was…

A harness that teaches itself what context to show Jev, TypeSafe's fast decision model, proves the result on a storm it never saw, and can't cheat because…