I and some friends have been working on a remote relevance capability, so I'm sharing a screenshots of the work in progress with you.
So far we have a proof of concept working, but not all features are implemented (encryption and authentication).
A few key features that I'm aiming for, or have already included:
- no ports listening on the client (endpoints that will answer the relevance questions), connections are client initiated
- a server component provides both the web ui for operators and connection point for endpoints
- intent is to provide both non-context (fast), inspector-context (slower) answers as well as the ability to specify a Fixlet to assess on the endpoints.
- Operators will be able to launch from a dashboard (possibly even in the dashboard)
- Operator will initiate a fixlet action to the targets who will be given a key (room token) to connect to the remote QNA session create by the operator.
- operator may reconnect to a session and recall commands
- logging of all commands and to which endpoints they were sent
- operator (user/pass) and endpoint (token) authentication
- encryption of questions and responses
- Multi-line input