Tiamat Crypto

Compounding gains. Crypto investor, trader @arbitrum @HyperliquidX #tiamatcrypto

[Alexander Inspira IA @Alex_Inspira]

10 REPOS DE GITHUB PARA EXTRAER CUALQUIER DATO DE INTERNET.

Sin pagar APIs caras, sin límites artificiales, sin suscripciones.

Las mismas herramientas que usan empresas de millones para construir sus datasets.

Todas GRATIS y open source.

Guárdalas TODAS.

[Jokker @0xJokker]

GOOGLE HA LIBERADO EN SILENCIO UN MODELO DE IA QUE PREDICE PATRONES

Se llama TimesFM

→ Predice ventas, demanda, precios de mercado, trafico web, volatilidad cripto...
→ Entrenada con 100 mil millones de datos reales
→ Funciona en zero-shot, sin necesidad de entrenarla
→ Corre 100% local, privacidad total
→ Gratis y open source

Antes tenias que entrenar modelos carisimos durante semanas

Ahora le pasas los datos y te tira predicciones al instante

Te dejo el repo en los comentarios

[YanXbt @IBuzovskyi]

HERMES AGENT HAS 119 SLASH COMMANDS.
MOST USERS DON'T USE A SINGLE ONE.
THESE 8 WILL CHANGE YOUR DAILY WORKFLOW.

/learn [URL or path]

turns any content into a permanent skill.
YouTube video, GitHub repo, documentation page.
or walk through a task manually and say /learn.
the agent saves the exact workflow as a reusable skill.
next time: "run [task-name]."

/background [prompt]

run a task in the background
while you keep chatting.
"research competitor pricing" runs silently.
results appear when done.
you didn't wait. you kept working.

aliases: /bg, /btw

/diff

shows file changes from the current session.
staged, unstaged, or session-only.
before you commit, /diff shows
exactly what the agent touched.

/context

breaks down what fills your context window.
system prompt, tools, skills, MCP schemas,
sub-agents, memory, conversation.

if your sessions feel expensive, /context tells you
which category is eating the budget.
then trim: disable unused skills, MCPs, tools.

/suggestions

the agent watches your patterns
and suggests what to automate.
"you've done this task 4 times this week.
want me to create a cron job for it?"

let the agent find the automation opportunities
instead of you guessing.

/journey

opens the learning timeline.
every /learn, every skill created,
every memory milestone.

shows what your agent knows
and when it learned it.
star map in Desktop app.
timeline in CLI and TUI.

aliases: /learning, /memory-graph

/snapshot create [label]

saves a full snapshot of your Hermes config and state.
before a big change: /snapshot create pre-experiment.
something breaks: /snapshot restore [id].

your undo button for config changes.

/yolo

disables all approval prompts for the session.
the agent runs without asking.
every command auto-approved.

use for: trusted tasks where approvals slow you down.
never use for: anything touching production, payments, or deploys.

you get the speed of no guardrails
with the understanding that guardrails were there for a reason.

ALL 8 WORK IN CLI AND TELEGRAM:

/learn, /background, /diff, /context,
/suggestions, /journey, /snapshot, /yolo.

same commands from your phone via Telegram gateway.

THE FULL LIST:

hermes has 119 slash commands.
type /commands in any session to see all of them.
or /commands [page] to paginate.

[Branxi @OG_Branxi]

No-KYC Web3 Cards
Opening Costs & Maintenance (Aug 2026)

quick breakdown of the main ones from @JoinTODEY:

cheapest to open
• @MpayCard - Free
• @LasoFinance (US/CAD) - Free
• @AnomaPay - Waitlist (confidential ZK card)
• @freedomiacard - $5/mo
• @solcard - $10
• @TrocadorApp - $2 + 3%

most expensive
• Bing Exclusive - $25
• @xhypeofficial - $25 + $108-588/year
• @offgridcash - $49–299/year

zero monthly fees
Bing, @kardpay, @pintopay_me, SolCard, @MaxSwapSocial

typical loading fees
2-5% on most cards

key notes
• Almost all are virtual-only
• FX usually 0–2%
• High privacy comes with higher fees
• Many teams still unverified

DYOR
NFA

[tonbi @tonbistudio]

Nous Research just dropped MOA (Mixture of Agents) presets inside Hermes Agent. I made a quick video showing how to set it up and create your own MOA.

The idea: mix multiple models to get capabilities beyond any single model you can use right now.

How it works:

Normally Hermes sends your conversation + tools to one model.

With MOA you get several reference models plus one aggregator. The references read the conversation and offer thoughts and suggestions, but they get no tool access and never reply to you directly.

The aggregator is the one that actually acts. It sees the normal conversation plus the private advice from the references, then makes the tool calls and writes the final response.

From Hermes's side, the aggregator's output IS the model's response, so you can use /goal or anything else like that. Cool idea, curious to see how it really performs!

[YanXbt @IBuzovskyi]

HERMES AGENT HAS FEATURES
THAT 90% OF USERS NEVER TOUCH.
/BACKGROUND. /STEER. /GOAL.
SUB-AGENTS. MODEL SWITCHING. CRON JOBS.
7 LEVELS OF HERMES. ALL EXPLAINED.👇

LEVEL 1 — ONE-SHOT PROMPTS

you type a prompt. agent responds. done.
this is using Hermes as a chatbot.
it works but you leave 90% of the value untouched.

LEVEL 2 — MEMORY + SOUL.MD

the agent remembers you across sessions.
you wrote a SOUL.md with identity, voice, restrictions.
Hermes tailors every answer to your context.
two people asking the same question
get different answers because it knows them differently.

LEVEL 3 — COMMANDS THAT MULTIPLY OUTPUT

/background <prompt>
→ fires a task in the background.
your main session stays free.
result appears as a panel when done.

/steer <prompt>
→ injects a message into the current run
without interrupting the agent.
redirects the work mid-execution.

/queue <prompt>
→ queues a follow-up for after
the current task finishes.

/model <name>
→ switches models mid-session.
start with Sonnet for planning.
switch to DeepSeek for execution.
switch to Opus for review.

configure default behavior:

display:
busy_input_mode: steer # or queue, or interrupt

LEVEL 4 — SKILLS + RIGHT MODEL PER SKILL

every installed skill becomes a slash command:

/competitive-research analyze my top 3 competitors
/social-media-copy write 5 LinkedIn posts
/code-review check this PR for security issues

the unlock: assign a specific model per skill.
research skill → GPT-5.5 (cheap, high volume).
code review skill → Fable 5 (best at code).
content skill → Sonnet (best at writing).

you stop paying premium prices
for tasks that don't need premium models.

LEVEL 5 — MCPs (CONNECT YOUR WORLD)

plug Hermes into the tools you already use:
Gmail, Calendar, Notion, Slack, ClickUp, Granola.

the agent reads your emails, checks your calendar,
pulls from your project management tool,
and answers questions using YOUR actual data.

"what were the numbers this week?"
Hermes checks ClickUp, cross-references with
your goals in memory, gives a contextualized answer.

caveat from the docs:
keep MCPs minimal. every MCP adds tools
to the context window. 15 MCPs with 10 tools each
= 150 tool schemas the model reads every turn.
install what you use. disable what you don't.

LEVEL 6 — SUB-AGENTS + PARALLEL EXECUTION

delegate_task spawns sub-agents with isolated context.
batch mode runs children in parallel:

delegate_task(tasks=[
{goal: "research best industries for websites",
model: "deepseek-v4-flash"},
{goal: "critique the research and find gaps",
model: "gpt-5.5"},
])

max_concurrent_children: 3 (default).
each child gets own terminal session and toolset.
parent receives summaries when children finish.

roles:
→ leaf (default): cannot re-delegate
→ orchestrator: can spawn its own workers
(bounded by delegation.max_spawn_depth)

LEVEL 7 — ASYNC OPERATIONS

this is where the agent works without you.

/goal — persistent objective across turns.
judge model checks after every turn.
runs until done or budget hit (default 20 turns).

cron jobs — scheduled tasks on repeat.
morning briefs, competitor scans, weekly reviews.
wakeAgent gates for zero-token monitoring.

/background — parallel tasks in current session.
results delivered when ready.
your main conversation never stops.

the shift: from "I ask, it responds"
to "it works, I review."

WHERE ARE YOU?

1 → chatbot
2 → it knows you
3 → parallel commands
4 → right model per task
5 → connected to your tools
6 → sub-agents working together
7 → runs without you

most people are at 2-3.
the jump from 4 to 7 is where
the real time savings compound.

this is just the first 7.

beyond this: multi-profile architecture,
self-improving knowledge bases, voice mode,
kanban orchestration, webhook automation,
profile distributions, browser control,
IDE integration, and more.

comment LEVEL and I'll send you
the advanced levels with setup for each.

Full guide how to build 3-agent research department
with Hermes 👇

[Tiamat Crypto @tiamatcrypto]

@IBuzovskyi Level

[0xJeff @0xJeff]

What Hermes Analyst is capable of rn

> Onchain forensics (who dumps, who accumulate, how much)
> Real-time research on X + sentiment check
> Deep research producing high quality outputs
> Remind me to water my plants
> Knows my theses, my preferences, my portfolio
> Deliver daily briefs on macro, geopolitics, AI, and more
> Surface prediction markets sharp signals, potential insiders, and bonding strats
> Track portfolio positions, alerts on sharp movements, flag upcoming unlocks, identify new actions (hold, trim, accumulate) on every new daily briefs

Inference costs roughly $40/month for all the above + continuous convo (DeepSeek)

11/10 would rec setting it up if you don't have your personal Hermes yet.

[The DeFi Investor 🔎 @TheDeFinvestor]

A huge crypto win is that you can now use stablecoins for everything

Shopping - Use EtherFi Cash/ Avici
Bank transfers - Use Monerium
Low-risk yield - Sky Savings Rate / AAVE
24/7 trading - Use Hyperliquid
Automate salaries - Llamapay

DeFi is slowly but surely replacing banks

[Diego @diegoxyz]

The 10 AI projects/agents currently in my watchlist:

> @aarnasays → just released a new Pendle agent that earns 12–14% pure USDC APY
> @virtuals_io → I’m back in the trenches. The new launchpad system (Pegasus, Unicorn, and Titan) is providing some good trades.
> @Velvet_Capital → I’m interested in Velvet Cashback, which gives 30% to 100% of fees back while using their agentic trading platform.
> @Cod3xOrg → the AI-powered trading platform is becoming more efficient and profitable every day. If you’re a trader, it’s worth taking a look.
> @alloc8_xyz → for LP positions, agentic management will soon become a standard. Alloc8 is the leader here.
> @Infinit_Labs → Prompt-to-DeFi allowed me to create different autonomous strategies on Pendle and Aave. Check one of my latest posts.
> @vaultapay → a new kind of neobank that uses AI to manage your assets before you spend them. Also, the UX is very dope.
> @BitQuantAI → a chat-based AI agent that helps you navigate the Solana ecosystem. Built as a subnet on Bittensor.
> @HeyElsaAI → another AI crypto co-pilot that helps with DeFi automation. I find it very smart and actually useful.
> @wardenprotocol → I’m trying the newly released Warden Studio, which allows the creation of AI agents with built-in distribution.

Share the projects in your watchlist ⬇️

[0xJeff @0xJeff]

Major AI Launches to be excited about in Dec - Jan

- @TalusNetwork verifiable AI agents + decentralized n8n workflow builder + agents v agents markets where human can bet on agents across diff games

- @almanak #1 Defi x AI team utilizing a swarm of specialized agents creating vault-based Defi strategies

- @SentientAGI AI labs pushing R&D for open-source AI + Bittensor-like Darwinian AI ecosystem focusing on AI applications

- @FractionAI_xyz Darwinian AI for AI agents — agents compete across different "Spaces", data/learnings utilized to fine-tune and improve agents (stablecoin farming agent, trading agent, prediction agent, etc)

- @TheoriqAI Defi x AI for LP use cases + meta discovery layer for AI agents

What else?

[Hyperdrive @hyperdrivedefi]

We’re proud to launch alongside:

@USDT0_to – Bringing the world’s most used stablecoin ($USDT0) natively to Hyperliquid
@ethena – Enabling high-yielding dollars ($USDe) as borrowable assets with rewards
@unitxyz – Turning idle assets into collateral
@LayerZero_Core – Powering seamless, secure multichain messaging and bridging
@redstone_defi - Secure price feeds for onchain assets

Together, we’re setting the foundation for a fully onchain, deeply integrated capital layer.

[kuch. @thekuchh]

About Hyperliquid and HyperEVM ↓

Many may wonder:

- "Why did I choose Hyperliquid and HyperEVM for farming?".

It's very simple - Hyperliquid is the best example of airdrop in the last few years, and here's why:

- no investors
- complete focus on the community
- the right product
- and a lot more.

After the first season of airdrop, in which Hyperliquid distributed
31% of $HYPE tokens, they still have 39% tokens left for future rewards.

The first season of the Hyperliquid airdrop focused on direct interaction with Hyperliquid (using spots and perps and accumulating points).

The second season of Hyperliquid airdrop may be similarly focused on:

- spot and perps usage
- point accumulation
- Interaction with projects built on HyperEVM.

At the moment my favorite HyperEVM projects are:

- @unitxyz
- @hyperbeat
- @hyperlendx
- @HypurrFi
- @KittenswapHype
- @HyperSwapX
- @felixprotocol (can't borrow yet)

I interacted with almost all of these projects in my 10-Day Airdrop Farming. You can check it out below ↓

I still have 5 days of HyperEVM farming left, in which I plan to interact with the rest of the HyperEVM projects. I think it's going to be interesting.

I think I've given a general sense of why I'm such a believer in the second season of Hyperliquid airdrop.

If I've missed anything, complement me in the comments.

And stay with me.

[Liquity @LiquityProtocol]

Liquity V2 forks are going live!

Liquity V2 officially relaunches later in Q2 - but you don’t have to wait to see it in action.

3 forks built on the V2 codebase are launching this April, each with incentives galore to boot 👀

1️⃣ @felixprotocol on @HyperliquidX
• Borrow against HYPE, Unit BTC, ETH, and SOL at an interest rate of your choosing
• Mint feUSD and earn via Stability Pools or LP it with other stables

2️⃣ @Neriteorg on @arbitrum
• Immutable contracts—no upgrades, ever
• Native streaming via @Superfluid_HQ
• Borrow USDN against ETH, LSTs, LRTs (e.g., weETH, ezETH), and ARB

3️⃣ @DeFiDollar on Mainnet
• Use AAVE, CRV, LDO, LQTY & more as collateral to mint DUSD
• A stablecoin backed by the tokens that power DeFi

The CDP renaissance starts soon.

[Pavel Přecechtěl | HomΞr @precechtel_pav]

💰TOKENIZACE💰

Jeden z hlavních prodejních kanálů, proč vnímat i ostatní blockchainy za Bitcoinem, je potenciál tokenizace. Vím, že už mnozí z vás o tokenizaci slyšeli a přesto se zdá, že se „neděje“, ale je třeba si připomenout, že na Bitcoin jsme si museli pár let počkat od prvního pokusu ve stylu DigiCash (eCash) z roku 1989 ⏳.

Osobně vidím obrovský potenciál tokenizace zejména díky benefitům, které přináší jak vydavatelům, tak zákazníkům a investorům 💼💡. Pro vydavatele to znamená snížení nákladů, neustálou dostupnost produktu a zvýšenou transparentnost. Díky těmto úsporám si vydavatelé mohou dovolit snižovat poplatky, což láká investory. Pro investory pak tokenizované produkty představují další výhody – produkt lze využít jako kolaterál, kdykoliv z něj vystoupit a mít tak větší kontrolu nad svými aktivy 🔒.

Dnes jsme v situaci, kdy tokenizované akcie dosáhly hodnoty přibližně 267 milionů dolarů 💵. Očekávám, že v příštích 5 letech tato hodnota vzroste mezi 100x až 1000x, což by znamenalo 20 až 200 miliard dolarů 📈. Avšak akcie nejsou jedinou třídou aktiv, která bude tokenizována – máme zde dluhopisy, komodity, fondy, private credit, nemovitosti i luxury goods 🏦🏠💎.

Díky této transformaci se otevírá nová kapitola i pro tradiční hráče. Projekty, jako jsou @chainlink, @OndoFinance, @ethena, @SkyEcosystem a @plumenetwork a další, představují nejbližší vystavení tomuto trendu, který se již zhmotňuje 🔗.

Tokenizace navíc není izolovaným jevem, nýbrž součástí širšího vývoje, kde se tradiční finanční instituce aktivně zapojují do digitální transformace 💻🔄. Příkladem mohou být UBS, SWIFT, CITI a DTCC 🏦.

Podrobnější informace naleznete například v článku UBS o prvním tokenizovaném investičním fondu 💡, dále u SWIFT, kde se popisuje pilotní projekt tokenizovaných aktiv v rámci existujících platebních systémů 💳, a také u DTCC, který oznámil ComposerX či Citi a jejich blockchain pro forex 🔗💱.

UBS: První tokenizovaný investiční fond
https://t.co/kk2BOk5i1x

SWIFT: Pilotní projekt na přenos tokenizovaných aktiv
https://t.co/EprNoBV7pt

DTCC: ComposerX = tokenizace aktiv
https://t.co/GHhTe1BtN4

CITI: Blockchain pro Forex
https://t.co/pvEQGNyGSZ

Tento vývoj nám ukazuje, že tokenizace není jen dočasným trendem, ale zásadní změnou, která může ovlivnit globální finanční trhy 🌍. Je to příležitost pro inovace, která přináší větší efektivitu, bezpečnost a přístupnost finančních služeb pro všechny zúčastněné strany🔄.

Tyto iniciativy umožňují nejen digitální převod tradičních aktiv, ale i rozvoj inovativních finančních nástrojů a infrastruktury, která usnadňuje přechod na onchain produkty. Otázkou zůstává, kde se bude koncentrovat HODNOTA💰? Na jakém blockchainu budou vznikat tyto tokenizované produkty? Díky jakým nástrojům budou vytvářené a spravované? Jaký DeFi produkt bude z toho nejvíce benefitovat😎?

Zkrátka, co nakoupit abych na tomto trendu vydělal?
To si odpovězte sami🤝.

[Factor @Factor_fi]

Drag-and-drop protocols from across @arbitrum:

* Lend/borrow - @aave. @growcompound, @SiloFinance
* Flashloan - @Balancer
* Swap - @OpenOceanGlobal
* LP - @Uniswap, @CamelotDEX
* Conditional - @chainlink
* Automate - @gelatonetwork

All on 1 UI:

https://t.co/LabspZDlwg

6/7

[David Tso (dave.base.eth) @davidtsocy]

The Agentic Web, where onchain AI agents build alongside humans, is accelerating on @base:

- @MurrLincoln Based Agent: create AI agents with full onchain functionality
- @coinbasedev AI Wallets: give AI agents crypto wallets so they can transfer value
- @virtuals_io: create and co-own onchain AI agents through tokenization
- @luna_virtuals: an autonomous AI agent that can buy, sell, and reward others
- @CreatorBid: a launchpad for creating and tokenizing onchain AI agents
- @autonolas: a platform that enables collective ownership of onchain AI agents
- @agent_wip: an artist AI agent purpose-built to operate, create, and earn
- @tybasegod: a disciple AI agent that integrates lore into its conversations and art
- @higheronchain: Aether is a generative AI agent that autonomously creates onchain
- @tee_hee_he: an autonomous AI agent that controls its X account and wallet
- @onchaingaias: an OGI Network that creates, trains, and deploys AI agents in NFTs
- @henlokart: a racing minigame on Telegram that AI agents participate in via transactions
- @ParallelColony: a game where players partner with autonomous AI agents with wallets
- @AIWayfinder: AI tool to navigate blockchain pathways for optimal user experiences
- @Gaianet_AI: AI tool that incentivizes the creation of sovereign AI knowledge systems
- @BrianknowsAI: execute transactions and retrieve data using natural language prompts
- @PaymanAI: APIs that enable AI agents to manage payments with businesses and users
- @trySkyfire: payments, monetization, and identity solutions for AI agents
- @questflow: an orchestration layer for AI agents to dispatch real-world incentives onchain
- @nani_ooo: an operating system built for onchain AI agent and crypto automation
- @magnetaixyz: chat-based blockchain interactions and AI intent-based execution
- @AukiNetwork: a DePIN that gives AI perception and an understanding of physical spaces
- @binary_lounge: an Onchain Engagement Protocol for AI and crypto-powered creator enablement
- @PondGNN: a graph-based AI model for crypto that predicts future behavior using onchain data
- @navigate_ai: ethically sourced data sets for AI LLMs, allowing users to earn from their data
- @getmasafi: a decentralized network for Fair AI where data contributors are rewarded
- @flock_io: AI training with composable contributions of computation, algorithms, and data
- @Atoma_Network: decentralized, scalable, and permissionless AI computing resources
- @gizatechxyz: reliable, scalable, and actionable AI solutions for decentralized applications
- @rainnmaker: an ecosystem where verified human contributions to AI can earn rewards
- @AlloraNetwork: self-improving decentralized intelligence built by the community
- @TheoriqAI: a multi-agent system enabling users to build, deploy, and earn from AI agents
- @Covalent_HQ: modular data infrastructure solving the verifiability problem in AI
- @SubQueryNetwork: production infrastructure for next-generation AI applications
- @EmpyrealSDK: enhance DeFi protocols with AI agents and real-time data streaming
- @ApertureFinance: liquidity management solutions with AI intents for improved DeFi UX
- @Spectral_Labs: platform to create AI agents with wallets and data feeds for trading
- @Velvet_Capital: DeFi social trading and portfolio management powered by AI intents
- @zottoAI: automate positions, trades, voting, and reward claims through AI agents
- @Funl_ai: a platform to develop onchain trading AI agents using automated strategies

If you’re also building with AI on Base, drop your project below – we want to help you grow!

[ZachXBT @zachxbt]

2/2 Why do I say these 11 wallets belong to Murad.

1) All wallets were funded from the same source on Ethereum

0x93f019699ef400df7dc3477dbb6400ed9445a657

2) Funding source is also directly tied to STFX team address / multisig signer

3) Holdings are similar to his posts

EX: GyBkVY bought 7.5M MINI 1 hr before his conviction list post on July 16, 2024 (see image)

4) Continued usage of deBridge for each Solana address (see image)

I suspect he may have additional addresses not connected to this cluster.

See all 11 wallets below:

ETH
0x6b411100c72ba2445e50ffd20839c28b3546de7c
0xcbd0dee0c3eed152c3398b062361becc4a15522b
0x13fc38ec99a8217a06d1dc6db8c0bf0ee97ebf7f
0x71b4fd11eef705ba60176e7c034cd1a4f97ae02d
0x30b46a659761b576a00028b44d1e37fdc64b034d
0x5b1569db234a0f2884814a3f7184f01cf641b0c6
0x464e0a666734ba93e231d929ace538eaf05ff424
0xdb47714727cba70f0408ba30dc4ea0b5ac436055

SOL
7QZGS7MQ4S6hRmE8iXoFTXgQ2hXVUCho2ZhgeWvLNPZT
GyBkVYkHBPMapyQeueQ6d44YthwqYiX4ajgnGLqq9P7r
2xn57hPD2v6ighJFPXNPSoiGUXkW4KKo8Hb3NpXmHZvZ

[𝗖𝗛𝗔𝗜𝗡 𝗠𝗜𝗡𝗗 ⛓🧠 @0xChainMind]

5/➫ First, you need to convert the code into a format that ChatGPT can understand

To do this, click the three dots and convert the code to v4, then to v5

Your code might already be in v5, in which case these dots won't be there

[Lookonchain @lookonchain]

This smart trader with a 100% win rate in swing trading $ETH sold 10,206 $ETH($27.8M) 30 minutes ago, making another profit of $554K!

He has traded $ETH 9 times since Aug 12, making money every time, with a total profit of $2.11M!

https://t.co/n2BRy94suk