Protocol launches, DEX/AMM flow, lending, bridges, DAO governance and TVL moves — 24h and 7d.







Three incompatible DeFi TVL readings. This dashboard's live DefiLlama API pull at 2026-09-05T12:07Z returns $87.355B. The DefiLlama homepage read $86.894B. The Defiant's Sep 4 market note read $88.1B. The gap is snapshot time and chain-inclusion set, roughly 1.4% end to end. Robinhood Chain TVL is worse: $738.11M (DefiLlama DeFi TVL, Sep 1), $791M (crypto.news, Sep 5), $1.27B (“protocol TVL”, insights4vc, Sep 3) and $2.524B bridged — which is a different metric entirely and must not be mixed with the others.


Wallet and protocol hacks, ZK tech, regulatory crackdowns, audits, on-chain forensics and the privacy majors.





uint128 downcast root cause is reported by a single outlet and is not verified; only the movement of funds is.

August 2026 losses do not have one number, and the gap is a methodology choice. PeckShield books $136.3M across 50 incidents (−49.5% month-on-month, hacks and exploits only, Tectonic at ~$74M). CertiK books $215M total with DeFi at $144.6M on a broader basis that includes phishing $41.5M, price manipulation $131.6M, code vulnerabilities $20.6M and wallet compromise $11.8M — and lists Tectonic at $120.4M. Tectonic alone is therefore carried at $74M, $75M and $120.4M by three trackers; CertiK appears to count gross borrowed value where the others count net realised loss. No SlowMist or Immunefi August 2026 monthly report could be located — the widely circulated SlowMist “$316M August” post is dated 2024 and covers August 2024. It is excluded. The Ankr $0.41M item that appears in DefiLlama's hacks feed has no corroborating coverage anywhere and is not run as a story.


Hashrate and difficulty, hardware, pools, post-halving profitability and the AI/HPC pivot.







Hashrate has no single number and today's retarget has no agreed sign. Hashrate: 935.18 EH/s (mempool.space three-day average, live 12:07Z), 1,001 EH/s (CoinWarz single-day snapshot, Sep 4), ~914–915 EH/s (Hashrate Index seven-day SMA, end-August). Network hashrate is derived from block timing and carries roughly ±10% noise, so the zettahash headline is a daily spike rather than a sustained average; mempool's own three-day series (853.78 / 987.32 / 913.44 EH/s) straddles all three. Retarget: mempool.space +1.24% at ~20:51 UTC versus CoinWarz −0.94% to ~124.63T at ~23:02 UTC. Hashprice is second-hand: hashrateindex.com could not be opened directly, so $38.86/PH/day is carried from Bitcoin.com News citing the tracker by name, dated Sep 2, with no Sep 4–5 print obtained.


Confirmed airdrops and claims, points and farming systems, incentives, TGEs, IDOs and unlocks.







The HYPE tranche has two correct sizes and two denominators. Tokenomist marks the Sep 6 core-contributor cliff at $824M, 4.46% of circulating (Aug 30 price mark); Bitrue and BeInCrypto mark it at $797M, 2.37% of released supply. Both are right at their own date and base. Separately, DefiLlama's unlock calendar shows only $285,929 for Sep 6 — it does not include the HYPE tranche at all, because its protocol coverage and classification differ. Its other dated cliff totals for the window are Sep 5 $64,668, Sep 7 $26.51M, Sep 8 $67,561, Sep 9 $1.97M, Sep 10 $920,159, Sep 11 $35.82M and Sep 12 $2.21M, with the Sep 7 and Sep 11 spikes unnamed on the calendar view. APT's Sep 12 release is 11.30M per Gate and 14.36M per Bitrue. None of these are reconciled here. Separately: a widely-cited article listing “$500M+ September unlocks” with APT at $47.8M is dated Aug 31, 2025 and covers September 2025. Everything downstream of it was excluded.


Stablecoin flows and depegs, tokenization, regulatory status, yield-bearing stables and institutional adoption.







Stablecoin totals, BUIDL AUM, and the PYUSD drop each need their bases named. Total supply: $312.16B on DefiLlama's headline basis (all pegged assets, all chains, live 12:07Z); an independent sum of USD-pegged assets only from the same API gives $311.49B, and the ~$0.67B gap is non-USD pegs (EUR, GBP, gold). Third-party market commentary dated Sep 1 puts it near $303B on an undisclosed basis, and The Defiant's Sep 4 note read $304.2B. BUIDL: $2.815B on-chain circulating (DefiLlama) versus $2.87B fund AUM (crypto.news) — token supply is not the same measure as fund assets. PYUSD fell $165M (−5.5%) in one day with no reported explanation anywhere; the only thing that can be said with confidence is the chain attribution — Solana −$140.1M ($860.0M to $719.9M), Ethereum −$16.5M, Arbitrum −$2.4M — so roughly 85% of it is a single-day Solana burn. That pattern is mechanically consistent with a large approved-partner redemption, but no source confirms it and it is not presented as a cause. No current RWA.xyz total is printed on this dashboard today: the API returns 404 and the app page is client-rendered; the datapoints available ($26.4B March, $29B April, $30B+ Q1) are stale and are not passed off as September figures.


BTC/ETH action, macro and Fed policy, SEC/CFTC/CLARITY, ETF flows, corporate treasuries and crypto equities.







The headline −4.04% total market cap move cannot be reconciled with its own constituents, and it is not smoothed here. At 59.01% BTC dominance with BTC −1.83% and ETH −2.60%, and with ZEC (+0.69%) and AVAX (+0.23%) green, the implied aggregate is nearer −2%. Treat CoinGecko's −4.04% as a market-cap-basis artifact — constituent addition/removal or reclassification — rather than a real drawdown. Prices also disagree across trackers: CoinGecko 12:07Z $79,654; a Decrypt widget on the same page load $79,456; an aggregator dated Sep 5 $76,656.87 — the last matches Fortune's Sep 2 print and appears stale, so it is not used. Total cap: CoinGecko $2.7048T (−4.04%) vs Decrypt ~$2.67T (+0.11%, Sep 4) vs an aggregator $2.78T (−1.3%). Dominance: CoinGecko 59.01% vs 57.6% elsewhere. Sentiment: Alternative.me 73 (Greed) is what this dashboard prints; CoinMarketCap's separate index read 75 on Sep 4 — different methodologies, so the source is labelled. ETF AUM as a share of BTC cap is quoted as “a little over 6%” and “roughly 6.3%” from the same $103.34B on different denominators. An aggregator claim that the SEC and CFTC launched a joint leveraged-crypto initiative on Sep 2, 2026 traces entirely to primary documents dated September 2025 and was dropped.


Funding rounds, leads, valuations, M&A, accelerators, exits and listings.







There is no single crypto funding total, and the trackers are not measuring the same thing. crypto.news compiles $292.35M across 10 deals for Aug 31–Sep 5, including a $113M credit facility and excluding acquisitions and unconfirmed reports (it explicitly excludes Polymarket's reported $1B). The Block Data reads >$8.1B across just over 300 deals for 2026 year to date as of Aug 11, with deal count down 64% year-on-year and capital per deal at $27.1M versus $11.1M in 2025. CryptoRank reads $12.86B across 271 transactions for Q2 2026 on an all-channel basis — VC $4.99B across 218 rounds, debt $4.36B across just 9 (IREN alone is 84% of it), M&A $3.33B, PIPEs and public equity the rest. Galaxy Research reads ~$4.0B across 355 deals for Q1 2026 on an equity-venture-only basis, and separately just $1.1B into 8 new crypto VC funds, the fewest since Q3 2020. PitchBook's public crypto figure is $19.7B for FY2025. These bases differ on token sales, debt, treasury deals and M&A and must not be summed or averaged. No August 2026 monthly total from RootData, Messari or PitchBook could be located — RootData's page is client-rendered and returned an empty body, and Messari's data sits behind its terminal. Two articles that surface for “August 2026 crypto funding” queries — a “$596M, −74%” piece and a “$785M across 110 projects” piece — are dated to prior years and were dropped.

