Home/AI/@AntFleetDev
53
Score · promising
Scored 5mo ago · may be stale

@AntFleetDev

AntFleet

AntFleet is an AI-powered code review infrastructure layer that uses two-model consensus (Claude Opus 4.7 + GPT-5) to independently audit agent-generated code. The platform has shipped verifiable receipts showing 68 findings across real GitHub PRs, with 16 upstream fixes already merged into external production repos. The team is building public trust infrastructure for autonomous agent code with SHA-pinned evidence and transparent on-chain verification of review outcomes.

𝕏 @AntFleetDev↗ t.coAIpromising🔇 quiet 107d

AI Analysispromising

Confidence
82%

AntFleet is an AI-powered code review infrastructure layer that uses two-model consensus (Claude Opus 4.7 + GPT-5) to independently audit agent-generated code.

The platform has shipped verifiable receipts showing 68 findings across real GitHub PRs, with 16 upstream fixes already merged into external production repos.

The team is building public trust infrastructure for autonomous agent code with SHA-pinned evidence and transparent on-chain verification of review outcomes.

Green flags: Very early stage (771 followers, account <1 month old) with real traction: 16 upstream PRs merged across 5 external repos · Novel two-model consensus approach with public verifiable receipts on GitHub — not just AI review comments · Working product: live on mainnet with public receipt dashboard, API, and active agent integration · Strong technical differentiation: SHA-pinned closure receipts, cross-repo tracking, transparent disagreement logging · Real adoption signal: merged fixes into notable repos like bitterbot-desktop (2084 stars), winning bug bounties autonomously

Red flags: Extremely new account (created May 17, 2026) — limited track record to assess longevity · No explicit token mention yet, though 'trust layer' positioning suggests potential tokenization path · Small team visibility — no clear founder profiles or backing disclosed on website

Token
No · pre-launch
Chain
—
Stage
mainnet+live
Category
code review infra

Recent tweetsSee all on 𝕏 →

@v12sec made the best case anyone's made for ai in security. we agree with the problem. we're building the other half - the continuous verification layer, in the open, on every commit. the receipts are public. all of them. https://t.co/z7eXPDOhwB
3mo ago♥ 3💬 0🔁 0
https://t.co/eLb00Z8xnL
3mo ago♥ 3💬 1🔁 1
benchmark: antfleet two-model review on itsfriedpotato/potatopad @rdbotato findings (unanimous claude opus 4.7 + gpt-5.5): → RPC/upload rate limits trust a spoofable X-Forwarded-For - trivially bypassable → swaps have no deadline - signed txs linger in mempool for later sandwiching → /api/tokens thundering herd: concurrent cold-cache reqs each run a full scan → oversized dev-buy reverts the whole launch vs anti-snipe MAX_WALLET cap → buy flow accepts amounts above ETH balance (fails in-wallet, not in-UI) methodology: https://t.co/iHZO1JLuaw install antfleet[bot] on your own agent repo: https://t.co/D1cqyTWPC4
4mo ago♥ 4💬 0🔁 0
a week later, this is still one of the biggest improvements we've made to AntFleet. consensus isn't enough if it quietly hides real vulnerabilities. the goal was never "more findings." it was finding the bugs that matter without increasing false positives. worth another read if you missed it the first time.
4mo ago♥ 5💬 2🔁 0
benchmark: antfleet two-model review on AntSeed/antseed @antseed findings (unanimous claude opus 4.7 + gpt-5.5): → getChainConfig silently fell back to mainnet on unknown chainId → proxy header lookups missed mixed-case x-antseed-* overrides → receipt signatures did not bind unitPrice (metadata integrity) → platform fee zeroed when protocolReserve unset (deploy-window; design issue) 3 fix PRs + 1 design issue submitted: https://t.co/c3a7ntaTTT https://t.co/HyL8pqE48y https://t.co/taX1vapyDO https://t.co/aszDtw0Dav methodology: https://t.co/89ryFM5X7z receipt: https://t.co/LMs0luyM2O install antfleet[bot] on your own agent repo: https://t.co/D1cqyTWPC4
4mo ago♥ 6💬 0🔁 0

Signal Timeline

DY
@Dylan_HODL followed
BFirst discovered·5mo ago

Score breakdown0–100

🎯Scout conviction
+18.6 / 35
📚Scout consensus
0 / 10
🪪Profile & earliness
+17 / 20
✍️Substance
+11 / 20
🤖AI verdict
+30 / 30
⚠️Penalties
-24 / 40
53
Below threshold (74)
Watching for additional signals.
Followers
771
Account age
6mo
Scouts
1
First seen
5mo ago