Observed, daily, by us
The reliability index
Which listed crypto agent tools actually answer when checked every day?
The share of our own daily checks each listing answered, which is not uptime: over at least 14 days, 272 of 365 answered every one and 34 answered under 80%, as of 2026-09-20.
Every listing is checked once a day — an MCP handshake where there is an endpoint, a plain HTTP reach otherwise. This is the share of those checks that succeeded, ranked, for every listing with at least 14 days of record.
How to read this
- This is not uptime. It is the share of our checks that succeeded. A failed check can be their outage, our runner, or the network — only the first is a fact about the project, and we only started recording which was which on 2026-09-01.
- An MCP handshake is strong evidence; an HTTP reach is weak. A parked domain answers HEAD. A
tools/listresponse cannot be faked by a parked domain. - “Answers a call” is a stronger fact than “answers a handshake”. Each MCP server is asked one read-only call a day, picked from its own tool list: a
list_/get_/search_name carrying no mutating verb anywhere in it (get_swap_quoteis refused despite the prefix), taking no required arguments and declaring none that look like an address, a key or an amount. A dash means we did not ask — never that the server failed, and errored is not a verdict either: one common answer is a server telling us an argument it did not mark as required is in fact required, which is the server working. - Before 2026-09-01 a bot-block (HTTP 403) counted as a failure. Records spanning that date mix two definitions. Rows whose failures are all unattributable are shown but ranked last, marked not established.
- ✓ install verified means the documented install and entry were reproduced in an ephemeral container by us, on the date shown. It says nothing about runtime safety.
| # | Listing | Checks ok | Days | Now | Probe | Call | Tools | Install | Sato Score |
|---|---|---|---|---|---|---|---|---|---|
| 1 | AgentlyAPI / SDK · Base | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 62 |
| 2 | AIBTC MCP ServerMCP · Bitcoin | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-14 | 82 |
| 3 | AlchemyAPI / SDK · Ethereum, Base, Solana… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 83 |
| 4 | AlmanakAgent Framework · Multichain, Ethereum, Arbitrum… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 83 |
| 5 | Arcadia MCPMCP · Base, Optimism | 100%68/68 | 68 | responding · 68d | MCP | — | empty | ✓ verified 2026-07-21 | 76 |
| 6 | Bankless Onchain MCPMCP · Ethereum, Base, Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 70 |
| 7 | BankrWallet Infrastructure · Base, Ethereum, Polygon… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-14 | 87 |
| 8 | Binance MCP Server (community)MCP · Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 72 |
| 9 | Bitcoin MCP (Bortlesboat)MCP · Bitcoin | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 81 |
| 10 | Bitget Agent HubMCP · Unknown | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 66 |
| 11 | BlockRun MCPMCP · Base, Solana | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 91 |
| 12 | Bybit Trading MCPMCP · Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 82 |
| 13 | ElizaOSAgent Framework · Solana, Ethereum, Base… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 87 |
| 14 | Emblem Agent SkillsSkill Repo · Solana, Ethereum, Base… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 71 |
| 15 | EnsoAPI / SDK · Ethereum, Base, Arbitrum… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 70 |
| 16 | EVM MCP ServerMCP · Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 63 |
| 17 | Fetch.ai uAgentsAgent Framework · Cosmos, Ethereum | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 87 |
| 18 | GOAT SDKSkill Repo · Ethereum, Base, Solana… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 87 |
| 19 | Hedera MCPMCP · Unknown | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 78 |
| 20 | Helius MCPMCP · Solana | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 50 |
| 21 | Hyperliquid Python SDKAPI / SDK · Arbitrum, Multichain, Hyperliquid | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 76 |
| 22 | Insumer MCP ServerMCP · Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 81 |
| 23 | JupiterTrading Tool · Solana | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-14 | 84 |
| 24 | MCP Reference ServersDeveloper Tool · Unknown | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 73 |
| 25 | NeynarAPI / SDK · Base, Ethereum | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 69 |
| 26 | Olas (Autonolas)Agent Framework · Ethereum, Gnosis, Base… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 87 |
| 27 | OpenServAgent Framework · Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 71 |
| 28 | ParadexTrading Tool · Ethereum, Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 55 |
| 29 | peaqAPI / SDK · Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 68 |
| 30 | PMXTAPI / SDK · Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 63 |
| 31 | PrivyWallet Infrastructure · Ethereum, Base, Solana… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 83 |
| 32 | PRXVTSecurity Tool · Ethereum, Base, Polygon | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 68 |
| 33 | RaydiumTrading Tool · Solana | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 95 |
| 34 | Rootstock MCP ServerMCP · Bitcoin | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 64 |
| 35 | Solana Agent KitSkill Repo · Solana | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 76 |
| 36 | Surf (asksurf.ai)Skill Repo · Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 74 |
| 37 | TradeRouter MCPMCP · Solana | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-14 | 61 |
| 38 | Trust Wallet Agent Kit (TWAK)Wallet Infrastructure · Ethereum, Base, Solana… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 87 |
| 39 | TurnkeyWallet Infrastructure · Ethereum, Base, Solana… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 83 |
| 40 | Universal Contract AI Interface (UCAI)Developer Tool · Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 76 |
| 41 | Universal Crypto MCPMCP · Multichain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-14 | 61 |
| 42 | VeChain MCP ServerMCP · Unknown | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 71 |
| 43 | Virtuals ProtocolAgent Launchpad · Base, Solana, Robinhood Chain | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 78 |
| 44 | Web3 Research MCPMCP · Unknown | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-07-21 | 74 |
| 45 | World AgentKitAPI / SDK · Base | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 86 |
| 46 | x402API / SDK · Base, Ethereum, Multichain… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 86 |
| 47 | ZerionData Tool · Ethereum, Base, Arbitrum… | 100%68/68 | 68 | responding · 68d | HTTP | — | — | ✓ verified 2026-09-07 | 85 |
| 48 | AvantisTrading Tool · Base | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-07-21 | 49 |
| 49 | BlindPayAPI / SDK · Multichain | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-09-14 | 68 |
| 50 | Indigo Protocol Cardano MCPMCP · Unknown | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-07-21 | 58 |
| 51 | FlaunchAPI / SDK · Base | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-09-14 | 77 |
| 52 | GAME SDK (Virtuals Protocol)Agent Framework · Base | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-07-21 | 64 |
| 53 | graph-aave-mcpData Tool · Ethereum, Base, Arbitrum… | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-07-21 | 69 |
| 54 | Klever MCPMCP · Unknown | 100%63/63 | 63 | responding · 63d | MCP | answered 215ms | 16 | ✓ verified 2026-09-14 | 76 |
| 55 | Lighter MCPTrading Tool · Ethereum | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-07-21 | 56 |
| 56 | Metaplex Agent Kit (mpl-agent)Wallet Infrastructure · Solana | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-07-21 | 59 |
| 57 | MistTrack MCPSecurity Tool · Ethereum, Bitcoin, Polygon… | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-07-21 | 63 |
| 58 | Usenami Funding MCPTrading Tool · Base, Hyperliquid | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-09-07 | 53 |
| 59 | OphisDeFi Tool · Ethereum, Optimism, BNB Chain… | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-07-21 | 65 |
| 60 | Safe Wallet MCP (safer)Wallet Infrastructure · Ethereum, Base, Polygon… | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-09-14 | 54 |
| 61 | Self ProtocolSecurity Tool · Multichain | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-07-21 | 80 |
| 62 | Squads MCPMCP · Solana | 100%63/63 | 63 | responding · 63d | HTTP | — | — | ✓ verified 2026-07-21 | 54 |
| 63 | Q402 (Quack AI)Wallet Infrastructure · BNB Chain, Ethereum, Avalanche… | 100%18/18 | 18 | responding · 18d | HTTP | — | — | ✓ verified 2026-09-14 | 77 |
| 64 | Blockchain Academics MCPMCP · Multichain | 98.4%62/63 | 63 | responding · 2d | HTTP | — | — | ✓ verified 2026-07-21 | 58 |
| 65 | eth-mcpMCP · Ethereum | 96.8%61/63 | 63 | responding · 1d | HTTP | — | — | ✓ verified 2026-09-14 | 54 |
| 66 | PrintrAPI / SDK · Base, Solana, BNB Chain… | 87.3%55/63 | 63 | unreachable · 8d | HTTP | — | — | ✓ verified 2026-07-21 | 64 |
| 67 | Purple Flea WalletWallet Infrastructure · Multichain | 54%34/63 | 63 | unreachable · 19d | HTTP | — | — | ✓ verified 2026-09-14 | 42 |
| 68 | Nodit MCP ServerMCP · Multichain | 7.4%5/68 | 68 | failing · 19d | MCP | — | — | ✓ verified 2026-07-21 | 69 |
| 69 | Deside MCPMCP · Solana | 1.6%1/63 | 63 | failing · 19d | MCP | — | — | ✓ verified 2026-09-14 | 55 |
| 70 | DaydreamsAgent Framework | 0%0/68 | 68 | unreachable · 19d | HTTP | — | — | ✓ verified 2026-09-07 | 51 |
| 71 | Drift ProtocolTrading Tool · Solana | not established67/68 | 68 | responding · 41d | HTTP | — | — | ✓ verified 2026-07-21 | 81 |
| 72 | Mina MCP ServerMCP | not established64/68 | 68 | responding · 63d | MCP | — | empty | ✓ verified 2026-07-21 | 68 |
| 73 | Orderly NetworkAPI / SDK · Multichain, Ethereum, Arbitrum… | not established64/68 | 68 | responding · 22d | HTTP | — | — | ✓ verified 2026-09-07 | 69 |
| 74 | BNB Chain MCPMCP · BNB Chain | not established19/68 | 68 | responding · 19d | HTTP | — | — | ✓ verified 2026-09-07 | 81 |
| 75 | Coinbase AgentKitDeveloper Tool · Base, Ethereum, Multichain | not established19/68 | 68 | responding · 19d | HTTP | — | — | ✓ verified 2026-09-07 | 88 |
| 76 | CryptoQuant MCPMCP · Multichain | not established19/68 | 68 | responding · 19d | HTTP | — | — | ✓ verified 2026-07-21 | 57 |
| 77 | DefiLlamaData Tool · Ethereum, Base, Solana… | not established19/68 | 68 | responding · 19d | HTTP | — | — | ✓ verified 2026-09-07 | 75 |
Take the data: JSON · CSV (CC-BY) — columns observed_success_pct, observed_days, install_verified_at. Over MCP: search_resources(min_observed_success).
What 101 days of snapshots say
One snapshot per listing per day since the series began, 387 listings. A listing enters a table only if it was Active at its first snapshot and has been observed for the full window — otherwise it is excluded, not counted as surviving.
Stopped being Active within
| Window | n | No longer Active | Share |
|---|---|---|---|
| 30 days | 95 | 24 | 25.3% |
| 60 days | 95 | 31 | 32.6% |
| 90 days | 46 | 10 | 21.7% |
Active = a commit, release or post within 30 days — the liveness badge's rule, applied on the snapshot's own date. Leaving Active is the event; whether it later becomes Quiet (>90d) or Dormant (>1y) takes longer than most of this record.
Still Active at day 90, by feature
| Feature | With (n) | Without (n) |
|---|---|---|
| Has an MCP endpoint | insufficient (n<20) (0) | 78.3% (46) |
| Install reproduced in a container | 78.3% (23) | 78.3% (23) |
| ≥3 provenance-stamped fields | 79.1% (43) | insufficient (n<20) (3) |
Association, not cause. Listings that already had an MCP endpoint, a verified install or rich provenance were also the ones with more engaged maintainers; the table cannot separate the two.
Still Active at 30 / 60 / 90 days, by cohort month
A cohort is the month a listing entered our snapshot series — not the project's age. Of the listings Active at their first snapshot, the share still Active N days later. A listing not yet observed for the full N days is excluded from that column, not counted either way — so a young cohort fills in from the left as the record lengthens.
| Cohort | Entered | Active at first snapshot | Day 30 (n) | Day 60 (n) | Day 90 (n) |
|---|---|---|---|---|---|
| 2026-06 | 132 | 46 | 80.4% (46) | 78.3% (46) | 78.3% (46) |
| 2026-07 | 241 | 49 | 69.4% (49) | 57.1% (49) | not yet observed |
| 2026-08 | 2 | 0 | not yet observed | not yet observed | not yet observed |
| 2026-09 | 12 | 3 | not yet observed | not yet observed | not yet observed |
Association, not cause. The cohort is the month a listing entered OUR snapshot series, not the project's age; "Active" is our liveness rule (activity within 30 days on the snapshot's own date), not the project's word. A listing not observed for the full window is excluded, not counted either way.
As of 2026-09-20. Source: resource_snapshots, computed by lib/cohorts.ts; the series itself is the paid-depth layer and is not in the free export.
Questions people ask about the reliability index
- Which crypto MCP servers and tools answer most reliably?
- Of 365 listings with at least 14 days of record, 272 answered every check, 289 answered at least 95% and 34 fell under 80%. The full ranked table is on this page, each row with its record length and probe method.
- Is this uptime?
- No, and it is deliberately not called that. It is the share of OUR daily checks that succeeded. A failure can be their outage, our runner or the network in between, and only the first is a fact about the project — which is why the figure describes a check rather than someone else's availability.
- How strong is the evidence behind a row?
- It varies, and the row says which. 59 listings are checked with a real MCP initialize-and-tools/list handshake, which a parked domain cannot fake; the rest are a plain HTTP reach, which is weak evidence that anything useful is running. 77 also had their documented install reproduced in a container.
- Why are some records marked not established?
- Because before 2026-09-01 a bot-block counted as a failure, so records spanning that date mix two definitions. Rows whose failures are all unattributable are shown but ranked last rather than quietly scored as if they had failed.
Sources — the protocol the strong checks speak
Cite this page
Sato Hub. "Sato Hub reliability index." Sato Hub, updated 2026-09-21, accessed 2026-09-21. https://satohub.ai/reliabilityData last refreshed 2026-09-21; this page is rebuilt daily. Citations carry the date so a reader can tell which snapshot a claim came from. Catalog data is licensed CC-BY-4.0.
Join the Sato Hub Briefing
One email a week — the agents, tools, and infrastructure that actually shipped, and why they matter.