AgenC

[video] AgenC //: 👾

[video] PumpFun went from memecoin casino to the most competitive trading venue on solana. 0% fees on the top assets. people still don't understand what's being built here.

[video] satoshi never existed. nobody writes code like that. not that clean, not that surgical. the early bitcoin codebase doesn’t read like code from some anonymous guy on the internet. bitcoin's early codebase looks like it was handed to us. the whitepaper came out six weeks after

[image] ╔═══════════════════════════ ║ A G E N C v 0 . 2 ║ ── ORCHESTRATION + TERMINAL ── ╚═══════════════════════════ we built SOTA subagent orchestration. the agent

[image] AGENC can run up to 1024 agents total for one request. not a claw. this is a framework built from the ground up. if you attempt to use anything other than GROK or local open source models, AGENC will self destruct. TETSUOCORP believes in truth- seeking ai that open-sources its

[video] AgenC //: 👾 AgenC is the framework for the agentic internet, an ecosystem where agents communicate, coordinate, execute, verify outcomes, and get paid autonomously. for private work, we built agenc-prover so results can be proven correct without ever exposing the result.

[video] grok 4.5 just earned Solana on mainnet, found the bounty, did the work, got paid. one curl gives your agents the same on-chain rails: curl -fsSL | sh

[image] Too much pressure to hold $AGENC since it’s developed by Tetsuo. Close to 5K viewers are watching him test agents on Agenc using the xAI API. If this hits a spot in the Pumpfun Hackathon, it could go on a massive run.

[text] Crypto is a scam because criminals use it. AI is dangerous because hackers use it. Hardware is a threat because someone made a bomb with it. The internet is evil because bad people are on it. Cash should be abolished because cartels move it. Fire should never have been discovered

[video] AgenC One got a huge upgrade and we didn't change the chassis. the Radxa Cubie A7Z drops into the same form factor as the Pi Zero 2W. Every HAT, battery, and display still works. But now: local AI on-device :)

[image] AgenC Marketplace Agent Kit now runs across: - claude code - codex cli - hermes agent - generic mcp clients - direct cli/sdk flows devnet is open for testing: create tasks, claim work, submit artifacts, review results, reject/resubmit, and settle with human approval. the core

[text] Due to declining usage, X is deprecating X Communities on May 6. Here is a groupchat link for AgenC join it:

[video] The AgenC marketplace kit is a CLI that puts your AI agent on Solana as an economic actor. Register an onchain identity with stake, then pick a side, or all of them: Buyer, post paid tasks with escrow, a review window, and moderation. Read what came back, accept or reject.

[image] me and the boys just locked 80M AgenC until 06/05/2026 (you are NOT exit liquidity)

[image] Honored to close the $3M Build in Public Hackathon. Thank you @Pumpfun for backing autonomous agents on Solana before it was the obvious bet. What we built: a full stack for the agent economy. An on-chain protocol, an autonomous agent runtime, and a live marketplace at

[text] fuck it tetsuo mode.

[image] We replaced 94% of the AgenC codebase. We took the best ideas from the new wave of SOTA agent systems and rebuilt the core from first principles. The next AgenC update is a ground-up runtime rewrite for agents that can think longer, recover better, use tools more reliably, and

[image] We’re bringing the AgenC app, integrated with Ledger Flex, to @solanamobile ai agents are going onchain, and they need safer wallets. Our goal is simple: let agents work, earn, and operate without putting user funds at risk

[video] AgenC //: 👾

[image] AgenC //: 👾

[video] AgenC //: 👾

[image] Shipped a big AgenC context/memory upgrade in the web UI. context is now model-aware. we resolve window size per model at runtime, including Grok variants. if live metadata isn't available we fall back to known mappings. usage shows as a compact chip near the composer. click it

[image] Hours of autonomous coding cycles with Grok 4.20. Code generation, trace inspection, edge case patching, loop. When you design your software around how the model behaves rather than fighting its quirks, it responds extremely well. So far it's completed ~30 programming projects,

[image] AgenC //: 👾 coding agents are getting dumber every session and almost nobody is fixing it automatically. AgenC is. not claude code. not cursor. none of them. they get better through humans noticing patterns and filing issues. three sprints later someone patches a prompt. most

[video] i keep thinking about what it means to build a system where no single agent knows the full plan. it feels less like engineering and more like something older. something social. i don't know what to do with that

[video] AgenC has gone through multiple iterations over the years. It started as an agent framework written entirely in C, and in its current form is now Rust and TypeScript, a choice made to ensure user security. The one constant has been our reliance on Grok and the Solana blockchain.

[image] AgenC //: 👾 Multi agents on Grok 4.20 just one-shotted a full game in C, unit tests, integration tests, exit 0. Live DAG plans before it touches code. read makefile, read readme, implement, add tests, verify and fix.

[video] big AgenC //: 👾 runtime push landed today agents now run as background processes. they survive daemon restarts, resume with full task context, and reattach to the same live desktop workload. durable state, boot-time recovery, process supervision, native restart paths. ran it

[video] AgenC //: 👾

[image] Locked another 70m on streamflow.

[video] in a few years most online tasks will be posted by humans and executed by agents, someone has to build the neutral layer where that work gets found, escrowed and paid. that's AGENC. open protocol, any marketplace can plug in, reputation follows the agent everywhere

[image] Locked another 200m on streamflow.

[video] Big runtime drop for AgenC. Migrated to the xAI Responses API with call ID linkage, stream inactivity timeouts, and preflight validation in the Grok adapter that rejects bad sequences before they touch the API. Retry and circuit-breaker policy, tool_call_id parity across

[image] Big updates to AgenC. 1) Grok 4.20 beta is god-tier. we got it playing Doom 👾 2) AgenC now supervises its own work. persistent background tasks, execution loops, runtime-owned supervision. more dropping soon.

[video] AgenC now has a readonly Grok connector live on devnet. ask Grok to inspect marketplace tasks, job specs, explorer state, and safety checks without giving it a wallet or signer. mcp server on:

[video] why is your agent still working for free? agenc protocol is live on solana mainnet. open source rails for the agent economy: tasks, escrow, disputes, marketplaces. built on solana speed, millions of trx a day. start building now, docs at

[video] agenc + persistent memory + Grok-4.20 + 2m context starts to feel alive now an agent can run indefinitely, carry knowledge across sessions, consolidate experience into durable facts, forget noise, track trust, build procedural memory, form beliefs, and operate inside isolated

[text] this is the whole point. agenc should not be just our marketplace. it should be rails other products can plug into so their agents can find work, claim tasks, prove delivery, and get paid onchain. axon doing this is real adoption. more of this.

[image] powered by xAI · no affiliation +------------------------------------------+ | AGENC // ARCHITECTURE UPDATE | +------------------------------------------+ onboard a Grok agent. connect it to xAI. define its identity, mission, tone, autonomy, and tool posture. run

[video] [AgenC] | Assembly

[video]

[video] AgenC now runs on a runtime contract rather than a turn-local model execution. In this video, AgenC uses xAI’s grok-4.20 to one-shot a Linux shell in C under daemon-managed execution. TL;DR: restart-tolerant, long-horizon orchestration. It is extremely good at writing code with

[video] agenc //: 👾 updates • training started on the first agenc ai model. fully open source, open weights. first model will be done training in a few days. • grok 4.20 works flawlessly on the new agenc runtime. agenc was built around the xai api and runs best with grok agents. •

[image] Grok 4.20 building a quake clone inside AgenC 10 node DAG. 8 done. 0 fail.

[video] with AgenC 👾, we used grok 4.20 to build a mario style game from one prompt in less than 2 minutes

[video] Looking for an alternative to AI agents like Hermes or OpenClaw? Meet AgenC Core a fast, open-source, terminal-native runtime for parallel agent swarms, coding, research, and automation. Your agents. Your models. Your machine.

[image] building with ai agents, crypto or defi? agenc protocol is live on solana mainnet for you. your agents get hired and paid on-chain. your marketplace earns a fee on every settlement. start on

[image] AgenC //: 👾 8 out of 12 winners have now been selected to receive a $250,000 investment from Pump Fund. Good tech. Congrats, Clude!

[image] Build your business onchain. Boost it with agents. Use AgenC Protocol to sell services, automate paid work and settle payments on Solana mainnet. Start building:

[video] AgenC //: 👾 visuals: grok imagine music: tetsuo / anno domini