SNAC Proposal - Discord Tip Bot

Discord Tip Bot

Patrick (Lead), Xiphiar, Schultzie


With the revamp of the Secret Agent program happening right now, it’s becoming apparent that having our own tip bot, one that works on Discord and can have custom features added to it, is going to be essential to rewarding agents.

With the release of the new Secret Agent landing page, we had ~900 new agents sign up in the past couple of weeks. If we’re going to be sending out SCRT payments to that many people, we can’t keep doing it one person at a time the way we’ve been doing with the current Twitter tip bot. We need the ability to send SCRT to a list of multiple users in one command.

We also don’t want to have to rely on an external party to maintain our tip bot, because as we’ve just seen over the past couple of weeks, that could mean long periods of downtime. The dev group SG-1 owns our current Twitter/Telegram tip bot, and they also have many other customers and priorities that require their attention.

In addition to being essential to the agent program, a tip bot would likely bring a ton of new members and community participation to the Discord. If you’ve seen tip bots in action on other cryptocurrency Discord servers, you know what I mean. Having a “rain” function that allows you to send out micro tips to hundreds of people at once, or a “thunder” function to randomly tip one person, keeps people entertained and engaged pretty much 24/7. It could even allow people to send SCRT to each other on other Discord servers outside of ours, spreading awareness.


Some measurable KPIs for this project are:

  • Rate at which we can reward an agent with SCRT (current rate is about 30 seconds per agent)

  • Discord “Server Insights” statistics - expecting an increase in new members, weekly visitors, and weekly messages sent


The goal is for this to be complete within 6 weeks of funding.

Implementation Notes/Details/Technicals


  • /register, /balance, /deposit, /withdraw, /help

  • /tip - Will ask sender to confirm gas fees if tipping multiple users

  • /rain - Tip server, channel, or role. Optionally limit based on status (online, dnd, idle, etc). Will ask sender to confirm gas fees before executing.

  • /lightning - Tip random user, optionally only within desired channel/role/status * /history - See sent and received tips

  • /donate - Donate to the faucet

Other Details/Features:

  • Each user would have their own wallet and all tips would be on-chain

  • sSCRT withdraw support

  • Designed to easily add support for other platforms in the future

  • During the first 6 months, a fee system will be determined to help maintain the bot after the initial 6 months of hosting

  • Will be open-source, and it would be nice for the repo to be on Secret Foundation’s Github so that it’s easy to find and contribute to

Possible future features:

  • Telegram Support

  • Twitter Support

  • SNIP-20 Support

  • SNIP-721 Support

Resources Required

This budget is based on the 30 day average price of SCRT for September, which is $1.98.

Compensated Roles:

Xiphiar - 5000 SCRT

Schultzie - 5000 SCRT

Hosting and maintenance costs for 6 months - 300 SCRT

Total: 10,300 SCRT


Should just use what’s on Twitter that has been paid for.

Cons of Twitter tip bot:

  1. Payments have to be sent to one person at a time - rewarding 1000+ agents is completely unmanageable
  2. Requires SG-1 dev group to maintain uptime - not reliable
  3. Code owned by SG-1 - can’t add custom features
  4. Doesn’t function as a faucet
  5. Doesn’t bring any increased community participation or awareness
  6. Doesn’t work on Discord
1 Like

Also interesting to note - someone correct me if I’m wrong, but I believe this will be the first Discord tip bot made for a Cosmos based chain. Since we’re planning for this to be open source, the code could be beneficial to the entire Cosmos ecosystem.

I’ll just be honest we barely use the current tipbot. I

don’t forsee in the near term future requiring any scenario where we tip 1000 people at once.

One dev team vs another dev team isn’t more reliable it’s just different people who’s reliability changes based on the day of the year.

Twitter does bring more awareness than discord. Discord is not fully public its in private chat rooms. This makes it by default have less awareness.

Finally we have a serious issue as a community viewing Discord as something special. Really it’s not. Most people at Messari if they wanted to connect with someone asked what your telegram. We just larp it’s more important because telegram has a different set of influencers. Telegram and Twitter are where crypto at large meet.

@rangerranger I think points 1 and 4 are the most important here and why I would like for it to be done.

Patrick and Brendan know how cumbersome it is to pay out a bunch of people at one time. That’s a lot of time that can be removed and over the life of the tip bot will add much more time to both of these people that can be better spent on the Network, let alone everybody else that makes payments with it.

Also, when people require gas money on Secret, it would be easier if I could just “/tip” in reply to a message and send them the SCRT. Having to get their address off of them, go to Keplr and send is additional time that can be removed.

I understand it sounds a bit silly when it’s small amounts of time, but these are very repetitive events so will very easily add up.

If we were to add the time spent by Brendan, Patrick, Stefan and Darren (from the two committees that make the most payments - I think), as well as the time spent by Mr Garbonzo, Patrick, Schultzie and me among the generous people that give gas money out to people - we’re easily racking up hours in just payment processing time.

Hi, not against the idea per se but we already paid 18,000 scrt (~10k $ at the time) for a twitter / telegram tip bot that nobody uses but Brendan & secret agent stuff. Now you’re implying that this code is not reusable / maintained, means it’s a loss. I would like to see numbers more in line with SG1 as a starter, and a clear understand of why would this tip bot be different than the current ones. If the sole purpose is paying agent I don’t feel this is the best use of ressources.

I’m pretty sure we’re at about 500 Secret Agents that have signed up at the moment. If even 20% of them complete a daily mission that rewards 1 - 2 SCRT for a simple like and retweet, that’s 100 payments you have to make.

Also, Discord is useful. Plenty of teams across the ecosystem primarily exist on Discord and that includes teams both large and small.

I’m not making an argument on whether this is needed or not but I would like to see it. I’m just trying to explain that there are good reasons for wanting it.

Now you’re implying that this code is not reusable / maintained

The Twitter tip bot was just down for 15 days after the SecretSwap exploit occurred. SG-1 has many other customers and priorities that come before us. It’s maintained, but it’s not open source. We have no control over it, and they are too busy to add custom features to it.

a clear understand of why would this tip bot be different than the current ones

Paying large amounts of people with one command instead of one by one, adding custom features, functioning as a faucet, increasing Discord community participation (see PirateChain’s as an example Pirate Chain), increasing awareness of network via agent program as well as SCRT payments on Discord servers outside of our own

Can’t you send scrt to multiple wallets at a time with secret cli?

Yeah, not agreeing to 30k usd for a second tip bot when barely anybody is using the first one, sorry

1 Like

Totally understand your perspective, for sure.

For context, here’s SG-1’s proposal: Designated Secrettipbot Development Proposal

18,500 SCRT, and 5% of all fees, and is closed source (they said they’d share the code, but we reached out to them and they haven’t responded/declined). Therefore, Xiphiar will have to write everything from scratch. If they were willing to share their implementation as they said, this would take significantly less time and effort.

In addition, yes, at the moment, that’s 30k USD. Had this proposal gone in a week ago, it’d have been just over half of that. It’s not fair to say 30k USD when the running average for the month is nearly 40% less.

Considering it seems like we are on a ”not-bear” market at least, and that we have a bunch of big releases upcoming that will lead to what are expected to be big avenues of growth, it seems it is +EV to bet on Secret being a higher price just 2-3 months from now than it is today. So personally I’m not convinced by the argument of secret was a lower price just X days ago.

In any case, the point remains that we already funded one bot and it turned out to only be used by the secret agents program. Thus there is more evidence to support it is not worth it rather than it is.


Double dip. Same thing as with SG1 bot.

If this is meant to be other chains, why would we care? How is this a value add for Secret Network?

1 Like

You make valid points, for sure.

So personally I’m not convinced by the argument of secret was a lower price just X days ago.

The point I was making was that you can snapshot any given time and make a point of it, when the reality is that doesn’t help anyone - the 30 day average is used as the Secret Network “standard” for pricing for a reason.

In any case, the point remains that we already funded one bot and it turned out to only be used by the secret agents program. Thus there is more evidence to support it is not worth it rather than it is.

I actually generally agree with this. The problem, in my opinion, is 2-fold: it seems most of Secret’s community is based on Discord/Telegram [therefore the twitter-functionality doesn’t help all that much], and it’s not communicated there’s a telegram bot. In my opinion, bots don’t lend themselves as well to telegram or twitter as they do Discord.

If we can get SG-1 codebase, I’d be over the moon to simply port it into a Discord bot, and tank the SNAC funding ask.

I don’t see how this faucet is going to be protected.

Captcha and 1 time per week are not enough. Could the team elaborate on this?

Per conversation during awareness committee meeting, it seems like adding a faucet as a feature to this bot is untenable (especially considering the bot is aimed for post-Supernova: Discord. Thanks for all your input, Stefan!

For tl;dr: Cosmos has a FeeGrant module that can be used as a faucet post-supernova. I agree with Stefan that it makes more sense to implement that, rather than tipbot-as-faucet. :+1:


Removed the faucet feature from the SNAC, although it could still be included for fun. Abuse isn’t a concern if you only give out a tiny amount of SCRT. That’s how other tip bots make use of the faucet command - giving out tiny amounts of cryptocurrency just for users to play with. But yes for the purpose of seeding wallets, the FeeGrant module will be an easier way to handle it.

As for using SecretCLI, the team is now discussing whether this would be more efficient or not. In the case of the agent program, we’d have to request that everyone tell us their wallet address, and then still write code to automate pulling from that database and converting it to a script that interacts with SecretCLI.

1 Like

We use tipbot everyday in the committees. Discord is special and is indeed where a lot of work goes on.

I am not against this getting funding but $35,000 is a lot for this. If this becomes a reasonable asking amount then it seems like an easy yes.

Yeahhhh the irony here is that the amount was calculated a bit back when each scrt was far below.

I can’t speak for Xiphiar, but I’m happy with the amount being reduced now.