Ping & Path Bots

Ping and path bots answer a trigger word in the #testing channel with the route your message took to reach them. They are how you find out whether your node can reach a region, how many repeaters it went through, and which ones.

A region can run more than one. All of them live in #testing - see Channels for how to join it.

Using a bot

Send !bot to #testing. Every bot in range answers with its own trigger word:

@[Your Node] Use "wre" for Whangarei

Then send that trigger word on its own to get a path report:

@[Your Node] 🦈=2 🦘=3 🛣️=1942, 0C37, E6D1

The markers are the bot owner’s choice. Some use text instead, with the same fields:

@[Your Node] hash=2 hops=3 path=1942, 0C37, E6D1

The values change with the path. From a sender using 1-byte paths:

@[Your Node] 🦈=1 🦘=3 🛣️=CF, AB, 59

Reading the reply

Every reply opens with @[Your Node] - your node name, so your client knows the reply is for you. Then three fields, in whichever marker set the bot uses:

Marker	Text	Meaning
🦈	hash=	Path hash size in bytes - how many bytes identify each repeater in the path. Matches the 1/2/3-byte prefix modes
🦘	hops=	Hops: how many repeaters the message passed through on its way to the bot
🛣️	path=	The path itself, in order - the hex prefix of each repeater the message went through

So 🦈=2 🦘=3 🛣️=1942, 0C37, E6D1, or hash=2 hops=3 path=1942, 0C37, E6D1, means three hops, identified by 2-byte prefixes, via repeaters 1942, 0C37, then E6D1.

Tip

The hex prefixes in the path are the same ones you can look up with the NZ Prefix Tool to find out which repeater is which.

Regional bots

Every path bot must be listed here, with the area it covers and who runs it - preferably with a way to reach them. An unlisted bot is one nobody can ask about when it misbehaves.

Triggers must be unique nationally. Beyond that they are free: most start from the IATA codes proposed for regions, but akl north, or anything else unclaimed, is equally fine. Agreeing a single national scheme has been tried and did not converge - unique and listed is what actually works.

Four to a region is about the practical limit.

Listed north to south.

Trigger	Region	Software	Owner
wre	Whangarei	OwlShack
akl	Auckland	OwlShack	🐶Wes
hlz	Hamilton	Custom	🏍️Leyton
tga	Tauranga	OwlShack
npl	New Plymouth	meshcore-ha
wlg hutt	Lower Hutt, Wellington	OwlShack	CJC
wlg	Wellington	OwlShack
wlg karori	Wellington	OwlShack	be1a
nsn mapua	Māpua, Nelson	OwlShack	Gav
chc	Cashmere, Christchurch	OwlShack	Gav

Note

Community stub - this list is incomplete. Running a regional bot that’s missing? Add it, in its place north to south.

Writing a bot

If you are building or configuring one, match this pattern - people rely on the fields being the same everywhere. The markers are yours to choose; the fields and their order are not.

Respond to !bot with your trigger word and region name:

@[<Sender Name>] Use "<Your Trigger>" for <Area You Cover>

Respond to your trigger word with the path report:

@[<Sender Name>] 🦈=<Path hash size in bytes> 🦘=<Hops from sender to bot> 🛣️=<Path the message took>

or with text markers:

@[<Sender Name>] hash=<Path hash size in bytes> hops=<Hops from sender to bot> path=<Path the message took>

Pick one set and stay with it, so your region’s replies look the same every time.

Reply at the same path hash size the sender used. A node testing on 1-byte paths needs a 1-byte answer to compare against; converting it to 2-byte makes the reply useless for the thing the sender was trying to measure. This is the single most important detail - check your bot software does it before you deploy.

Beyond that, the rules for every bot apply: one short reply per trigger, and nothing sent unprompted. Channel messages flood by nature, so every reply is carried by the whole mesh - that is what makes a chatty bot expensive.