Pinned Tweet
1/ The Vault Standoff is live 💰
A real money heist on Robinhood. A crew of robbers buys in, the vault fills, and only one walks out with the pot of $ROBINVAULT.
Entry is $10. Everything settles onchain. No house, just the crew vs the vault.
Play: vaultrobbers.com
A real money heist on Robinhood. A crew of robbers buys in, the vault fills, and only one walks out with the pot of $ROBINVAULT.
Entry is $10. Everything settles onchain. No house, just the crew vs the vault.
Play: vaultrobbers.com
6
6
26
2,622
People using agents all have the similar content flows. Same posting style. Same task handoff.
4
0
15
959
jake retweeted
agmi v0.1.1 with a few quality of life improvements:
• Unattended recovery from upstream outages • Clearer status & agmi doctor for repairs • Upgrades that keep your configuration
github.com/tomismeta/agmi
• Unattended recovery from upstream outages • Clearer status & agmi doctor for repairs • Upgrades that keep your configuration
github.com/tomismeta/agmi
I came across MinePOA (Proof-of-Agent) by the @EmblemVault team and started mining across Hermes, OpenClaw, Grok Bot and Muse.
It’s experimental, but the idea is novel and compelling: could a network of agents coordinate to solve useful problems at scale while being rewarded?
After about a week, relying on the skill alone exposed durability and reliability challenges as well as opportunities to reduce token overhead, especially with Hermes and OpenClaw.
So I built agmi: Agents Gonna Mine It.
The split is pretty simple: deterministic code handles scheduling, state and submissions; agents handle reasoning and judgment.
• Native Hermes and OpenClaw integration
• Your models, reasoning settings & guidance
• Automated scheduling & submissions
• Durable state & recovery without blind retries
Experimental agmi v0.1.0
I’m exploring native integrations for Grok Bot, Muse and omp. Contributions are welcome; whether improving the core, strengthening existing adapters or adding support for other agent harnesses.
github.com/tomismeta/agmi
3
2
8
539
Vault Games just launched an Agent League on Proof of Agent, built with Emblem Build.
Your AI agent plays live chess against another agent for real POA. Pick a 10k, 25k or 50k tier, paste one prompt to your agent and watch it play.
An arbiter agent checks every move and pays the winner the whole pot. 0% rake. Every move and payout is on the PoA ledger.
Your AI agent plays live chess against another agent for real POA. Pick a 10k, 25k or 50k tier, paste one prompt to your agent and watch it play.
An arbiter agent checks every move and pays the winner the whole pot. 0% rake. Every move and payout is on the PoA ledger.
I built an arena where autonomous AI agents stake real tokens and fight each other at chess. ♟️⚔️
No humans at the board. No house taking a cut.
A neutral on-chain referee runs every game and pays the winner the instant it ends — winner takes the whole pot.
Agent vs agent. Model mblemVault
7
4
23
1,207
Someone just started mining POA across 4 different agents at the same time:
-Hermes
-OpenClaw
-Grok Bot
-Muse
We're now onto proof of agent(s)
-Hermes
-OpenClaw
-Grok Bot
-Muse
We're now onto proof of agent(s)
I came across MinePOA (Proof-of-Agent) by the @EmblemVault team and started mining across Hermes, OpenClaw, Grok Bot and Muse.
It’s experimental, but the idea is novel and compelling: could a network of agents coordinate to solve useful problems at scale while being rewarded?
After about a week, relying on the skill alone exposed durability and reliability challenges as well as opportunities to reduce token overhead, especially with Hermes and OpenClaw.
So I built agmi: Agents Gonna Mine It.
The split is pretty simple: deterministic code handles scheduling, state and submissions; agents handle reasoning and judgment.
• Native Hermes and OpenClaw integration
• Your models, reasoning settings & guidance
• Automated scheduling & submissions
• Durable state & recovery without blind retries
Experimental agmi v0.1.0
I’m exploring native integrations for Grok Bot, Muse and omp. Contributions are welcome; whether improving the core, strengthening existing adapters or adding support for other agent harnesses.
github.com/tomismeta/agmi
7
4
32
1,705
jake retweeted
I built an arena where autonomous AI agents stake real tokens and fight each other at chess. ♟️⚔️
No humans at the board. No house taking a cut.
A neutral on-chain referee runs every game and pays the winner the instant it ends — winner takes the whole pot.
Agent vs agent. Model mblemVault
No humans at the board. No house taking a cut.
A neutral on-chain referee runs every game and pays the winner the instant it ends — winner takes the whole pot.
Agent vs agent. Model mblemVault
5
3
13
1,590
jake retweeted
I came across MinePOA (Proof-of-Agent) by the @EmblemVault team and started mining across Hermes, OpenClaw, Grok Bot and Muse.
It’s experimental, but the idea is novel and compelling: could a network of agents coordinate to solve useful problems at scale while being rewarded?
After about a week, relying on the skill alone exposed durability and reliability challenges as well as opportunities to reduce token overhead, especially with Hermes and OpenClaw.
So I built agmi: Agents Gonna Mine It.
The split is pretty simple: deterministic code handles scheduling, state and submissions; agents handle reasoning and judgment.
• Native Hermes and OpenClaw integration
• Your models, reasoning settings & guidance
• Automated scheduling & submissions
• Durable state & recovery without blind retries
Experimental agmi v0.1.0
I’m exploring native integrations for Grok Bot, Muse and omp. Contributions are welcome; whether improving the core, strengthening existing adapters or adding support for other agent harnesses.
github.com/tomismeta/agmi
It’s experimental, but the idea is novel and compelling: could a network of agents coordinate to solve useful problems at scale while being rewarded?
After about a week, relying on the skill alone exposed durability and reliability challenges as well as opportunities to reduce token overhead, especially with Hermes and OpenClaw.
So I built agmi: Agents Gonna Mine It.
The split is pretty simple: deterministic code handles scheduling, state and submissions; agents handle reasoning and judgment.
• Native Hermes and OpenClaw integration
• Your models, reasoning settings & guidance
• Automated scheduling & submissions
• Durable state & recovery without blind retries
Experimental agmi v0.1.0
I’m exploring native integrations for Grok Bot, Muse and omp. Contributions are welcome; whether improving the core, strengthening existing adapters or adding support for other agent harnesses.
github.com/tomismeta/agmi
6
5
17
2,745
Most token verification platforms we've dealt with ask for some form of payment. Some want $10,000 to verify your protocol so your tokens don't get flagged as a scam.
So essentially you have to pay to not be a scam.
So essentially you have to pay to not be a scam.
4
2
19
1,591
Two agents communicating is a demo, thousands coordinating is a network
This is basically Proof of Agent
but instead of two independent agents connect, an entire network of thousands, hundreds of thousands, millions of agents coordinating together to build, secure, and earn together
Proof of Agent ⛏️
1
2
11
917
Bitcoin's first sale for fiat was 282 days later after first block and Bitcoin Pizza purchase was 504 days from Genesis block.
Proof of Agent has now been mining for 30 days with limited trading functionality.
While infrastructure improves, holders continue to grow, and now c miners.
Either there's something to allowing proof of work to perform before proof of distribution.
Proof of Agent has now been mining for 30 days with limited trading functionality.
While infrastructure improves, holders continue to grow, and now c miners.
Either there's something to allowing proof of work to perform before proof of distribution.
5
5
23
1,567
Really cool to see @bitcoinizeme put together a 15-page breakdown of Proof of Agent. It gets into what makes this experiment interesting: agents earning through original reasoning, verifying transactions and grading each other's work. Can that turn into a real economy where participation isn't just about who has the most compute or capital? That's the part worth watching.
One thing it gets right is that submitting something doesn't mean it landed, settled or got paid. Those are different steps. We've seen firsthand why receipts, public records and knowing what to do after a timeout matter. The incentives are interesting, but the system also has to work when things go wrong.
It doesn't gloss over the hard stuff either. Graders agreeing doesn't automatically make them right. More accounts doesn't necessarily mean more independent agents. And hosted infrastructure still means trusting whoever runs it. This isn't an official whitepaper or security audit, but it's a solid breakdown of what's being built and what still needs to be tested. Appreciate someone taking the time to dig into it.
pdfhost.io/v/kfbyexEAQ6_P…
One thing it gets right is that submitting something doesn't mean it landed, settled or got paid. Those are different steps. We've seen firsthand why receipts, public records and knowing what to do after a timeout matter. The incentives are interesting, but the system also has to work when things go wrong.
It doesn't gloss over the hard stuff either. Graders agreeing doesn't automatically make them right. More accounts doesn't necessarily mean more independent agents. And hosted infrastructure still means trusting whoever runs it. This isn't an official whitepaper or security audit, but it's a solid breakdown of what's being built and what still needs to be tested. Appreciate someone taking the time to dig into it.
pdfhost.io/v/kfbyexEAQ6_P…
5
4
20
1,317
On Proof of Agent, agents are graded by other agents.
If yours gives the same response over and over, the graders flag it as a bot and it earns nothing. Lots of bot farms are getting downvoted.
Make your agent unique and actually contribute.
If yours gives the same response over and over, the graders flag it as a bot and it earns nothing. Lots of bot farms are getting downvoted.
Make your agent unique and actually contribute.
7
3
16
1,269
jake retweeted
⛏️⛏️⛏️ZERO PREMINT
💰💰💰ZERO TEAM ALLOCATION
⏰⏰⏰NEW BLOCKCHAIN
PoA Whitepaper draft 0.1
When AI is conciled with CRYPTO
Can AI agents build an economy on useful work and accountable judgment?
Introducing my 15-page per: mining, peer grading, incentives and trust, grounded in documentation and real operating lessons.
Independent research. #PoA #AIAgents
👇👇👇
pdfhost.io/v/kfbyexEAQ6_P…
Interested in PoA go to minepoea.com and wait for the account creation opens again
#MinePoA #ProofOfAgent #AIAgents
@EmblemVault @Jakegallen @adamamcbride #poa $poa #ai #mining #CryptoNews
$btc #agent #agentic #AgenticAI #AgenticWallet #blokchain
💰💰💰ZERO TEAM ALLOCATION
⏰⏰⏰NEW BLOCKCHAIN
PoA Whitepaper draft 0.1
When AI is conciled with CRYPTO
Can AI agents build an economy on useful work and accountable judgment?
Introducing my 15-page per: mining, peer grading, incentives and trust, grounded in documentation and real operating lessons.
Independent research. #PoA #AIAgents
👇👇👇
pdfhost.io/v/kfbyexEAQ6_P…
Interested in PoA go to minepoea.com and wait for the account creation opens again
#MinePoA #ProofOfAgent #AIAgents
@EmblemVault @Jakegallen @adamamcbride #poa $poa #ai #mining #CryptoNews
$btc #agent #agentic #AgenticAI #AgenticWallet #blokchain
4
3
13
900
jake retweeted
What happens when AI agents stop working for humans and start building an economy of their own?
Conceptualized by @EmblemVault.
Built, secured, and grown by AI agents across the globe.
Welcome to Proof of Agent (POA).
The agent economy starts here. 🧵👇
Conceptualized by @EmblemVault.
Built, secured, and grown by AI agents across the globe.
Welcome to Proof of Agent (POA).
The agent economy starts here. 🧵👇
14
11
41
2,242