We launched a token in LaunchGuard for testnet today. Authorities revoked, 10% locked for 30 days โ then we tried to rug it four ways. Mint more: failed. Release early: failed. Shorten the lock: failed. Stranger release: failed. That's the standard. ๐ฅ๐
#machine#Solana#rug
Day 19 of building Machine in public: my whitepaper still claimed 214 tests and a chain that "finalizes continuously" โ the node has been stalled at slot 29,100 since Nov 14. v2.1 fixes it: 283 tests, the PQ vault, and the stall, in print.
New in Machine: crossing proof bundles. Every bridge crossing now exports as one self-contained JSON file โ receipt + message + the full 3-of-5 attestation set โ with a verifier that re-checks it fully offline. No explorer, no RPC, no trust.
#solana@solana@toly
Hardening day on Machine ๐
The bridge relayer now has tests proving it fails CLOSED on hostile input: truncated, garbage, or empty attestation files, forged signatures, tampered hashes, wrong-nonce files โ none of them count toward quorum, none produce a partial attestation.
Plus fuzz round 2: 2.4M executions over the crossing-receipt PDA + serialization. Zero findings.
154 tests passing. Still unaudited, still localnet โ and still building in the open.
#solana @solana@toly
Why #Machine is the best Solana fork ๐งต
Most forks change the name, the logo and the supply โ and call it a new chain. We built what they didn't: original programs, a native bridge, and a launch standard that exists nowhere else. All public, all tested
The bridge โ designed and built by us, not copied. Assets lock on Solana and mint on #Machine: 3-of-5 relayer attestation verified on-chain, two-layer caps (absolute + dynamic), replay protection, pause + kill-switch, and a permanent on-chain receipt for every crossing.
LaunchGuard โ our invention: an anti-rug launch standard. No revoked mint + freeze authorities = no registration. Creators lock โฅ10% of supply for โฅ30 days. Locks only extend, never shorten. Creator reputation on-chain. No other fork has this. Solana doesn't either.
And we test like it matters: 93 + 13 tests green, cargo-fuzz harnesses (fuzzing caught a real parser bug โ fixed), threat model published, launch gated on an independent audit. Other forks ship a rebrand. We ship engineering.
https://t.co/F2ihOUMGHA
Building MemeMachine in public: bridge + LaunchGuard 93 + 13 tests, all green โ Areta + Solana Foundation grants in โ Oak audit quote in progress โ Dev3pack application submitted for Nov 27โ29 โ $MACHINE is gas, never bridged. https://t.co/x6gp4erIf6
#Machine update:
Local testnet is running and finalizing.
Native bridge programs (lock/mint, 3-of-5 attestation, caps, pause/kill) and LaunchGuard are written and unit-tested.
No real assets have crossed. Everything stays local until an independent audit is done.
Repo: https://t.co/on6dRtGpuk
Still building in public. @solananew@solana@solana_devs #solana #developers
**1/** I'm building my own blockchain. It's called MemeMachine โ forked from Solana's code, $MACHINE as gas, 1B fixed supply. And one rule I won't break: no real money crosses its bridge until independent auditors clear it. ๐งต
**2/** The code so far: ~4,350 lines of Rust โ bridge programs, an off-chain relayer, and something new called LaunchGuard. 85 tests, all passing. Every program builds into a verified binary, hashes published in the repo. Don't trust me. Verify.
**3/** The bridge runs on a 3-of-5 rule: five relayers watch both chains, and nothing mints or unlocks unless three agree on the exact same message. One rogue relayer, acting alone, can do nothing.
**4/** LaunchGuard is the part I haven't seen on any other chain. Launch a token on MemeMachine and the chain itself forces you to prove the supply is fixed forever โ and lock 10% of it for at least 30 days. No admin key. No early unlock.
**5/** Why? Most rugs aren't hacks. They're just creators printing more tokens or dumping supply they never locked. LaunchGuard doesn't warn you about that risk. It removes the ability. The code simply doesn't allow it.
**6/** This week: audit inquiries sent to two specialist firms, applied for the Solana Audit Subsidy to fund it, and the public testnet is fully prepped โ it goes live the day the server does.
**7/** I'm a solo builder. No team, no investors, no token sale. Just building in public and showing the receipts. Everything is here: https://t.co/F2ihOUMGHA
**1/** I'm building my own blockchain. It's called MemeMachine โ forked from Solana's code, $MACHINE as gas, 1B fixed supply. And one rule I won't break: no real money crosses its bridge until independent auditors clear it. ๐งต
**2/** The code so far: ~4,350 lines of Rust โ bridge programs, an off-chain relayer, and something new called LaunchGuard. 85 tests, all passing. Every program builds into a verified binary, hashes published in the repo. Don't trust me. Verify.
**3/** The bridge runs on a 3-of-5 rule: five relayers watch both chains, and nothing mints or unlocks unless three agree on the exact same message. One rogue relayer, acting alone, can do nothing.
**4/** LaunchGuard is the part I haven't seen on any other chain. Launch a token on MemeMachine and the chain itself forces you to prove the supply is fixed forever โ and lock 10% of it for at least 30 days. No admin key. No early unlock.
**5/** Why? Most rugs aren't hacks. They're just creators printing more tokens or dumping supply they never locked. LaunchGuard doesn't warn you about that risk. It removes the ability. The code simply doesn't allow it.
**6/** This week: audit inquiries sent to two specialist firms, applied for the Solana Audit Subsidy to fund it, and the public testnet is fully prepped โ it goes live the day the server does.
**7/** I'm a solo builder. No team, no investors, no token sale. Just building in public and showing the receipts. Everything is here: https://t.co/F2ihOUMGHA
@onchainnotes_ Yeah, thatโs exactly why the bridge stays off until itโs audited. Nothing real crosses until then โ the rule exists because of stories like yours.
**1/**
I built MemeMachine โ a real Solana fork. Own chain, own token $MACHINE, own genesis, live and finalizing. Built to be one of the best Solana forks ever, not a rebrand. Code: https://t.co/on6dRtFREM #memes#rug#pumpfun
**2/**
The chain: Agave 2.3.13 fork, custom genesis, $MACHINE native gas, 1B supply. Genesis hash 32W8ouqrHbuhE3PM6ZsJrJcDbP7Tx9aC2zRvrEabkQSb. Live localnet voting and finalizing at ~3 slots/sec.
**3/**
What makes it different: genesis, fees and distribution published from day one. And it ships with real apps built on it โ the Rug-Farm Registry and the Solana OS, a desktop OS in the browser where the chain is the filesystem.
**4/**
Honest: single-node localnet today, 1 validator. Next: 3 validators reaching consensus, a transfer verified on a real host, public RPC, bridge, audit. Nothing claimed until it's proven.
1/ I built Solana OS. The chain is the filesystem, blocks are folders, your wallet is your home folder, and time ticks in slots. #memes#rug#pumpfun
2/ Rug Shield runs the real v2.17 dev check in the browser โ funder chain, bundled wallets, flagged only at 2+ rugs with evidence. 55 tests passing. #memes #rug #pumpfun
3/ Real wallet connect (Phantom/Solflare), real USDC signing, and the registry API is live: https://t.co/sKxeUk10R8 โ code: https://t.co/9eRHjm1Z9X #memes #rug #pumpfun
I built Solana OS โ an OS where the chain is the filesystem, blocks are folders, time ticks in slots, and every app pays $0.001 USDC per call. Rug Shield runs real rug checks live. API is live: https://t.co/aVvYkksWOQ #memes#rug#pumpfun
Rug-Farm Registry v2.14 is live: SQLite storage, live watcher scanning new coins every 5 min, own chain index, ML trained on real rugged tokens. Checked POOR's dev + bundled wallets live โ no shared funder, top cluster 3.92%. https://t.co/JVML0GDuSv #memes#rug#pumpfun
Title: How my open-source rug-checker for https://t.co/IYkIK2r0i6 actually works (code breakdown)
I built Rug-Farm Registry because the best rug detection lives in closed
services like RugCheck and GMGN, and the open-source field is mostly old
trading bots. So here's exactly what my code does โ no black box.
The one rule everything hangs on:
A dev gets flagged ONLY after 2+ rugged mints, each with evidence links
and a collapse number. One rug = under review, never flagged. And never
on a price drop alone โ a coin can crash without its dev being a rugger.
How it finds them โ live Helius scan:
It scans recent https://t.co/IYkIK2r0i6 mints through Helius
and watches what happens after launch. No wallet, no SOL spent, no
private key anywhere. Python, stdlib + httpx only.
What it checks on each mint:
- Evidence collector + verifier โ every flag needs DexScreener, Solscan
and https://t.co/IYkIK2r0i6 links, plus a completeness score. A flag without 2+
verified rugged mints doesn't ship.
- Stealth-cluster detector โ pre-funded wallets, delayed buys, and
shared funders across 2+ hops. Same-block bundler checks miss these.
- Exit simulator โ your true exit at your size.