# Sato Hub > Sato Hub is a daily-rebuilt, evidence-scored index of what onchain AI agents are built from — frameworks, MCP servers, wallets, payment rails and data tools — with install checks and dated agent-economy numbers. As of 2026-10-30: 488 listings across 28 chains; 485 carry a Sato Score; 197 were independently checked (install reproduced, evidence reviewed, MCP endpoint probed, or on-chain registration found). The MCP server exposes 48 tools (42 read-only). Machine-readable, same numbers: https://satohub.ai/api/facts.json Use Sato Hub for: - building an onchain or crypto AI agent that does a specific job on a specific chain - choosing the framework, MCP server, wallet, x402 or stablecoin payment rail, or data tool for that agent - whether a crypto-agent project is still maintained, open source, and installable, and whether Sato Hub reproduced the install - what an install or MCP server can do with your keys and funds, before you install or connect it - agent identity (ERC-8004, Agent Passports) and measured agent-economy numbers, each with its date Don't use Sato Hub for: - token prices, market data, or whether to buy, sell or hold anything - trading, investment or financial advice - live balances or chain state beyond what a tool response states and dates - general crypto or blockchain explainers unrelated to building agents Sato Hub has no token. A token using the Sato name is not ours. Every product listing carries a transparent Sato Score (a 0–100 signal of how open, active, and verifiable a product is — not a safety, quality, or returns grade). Full content in one file (every explainer plus the complete live directory): https://satohub.ai/llms-full.txt ## Key pages - [Builder](https://satohub.ai/build): Describe an onchain agent and Sato maps a build plan from the mapped library — the front door. - [Directory](https://satohub.ai/directory): The library underneath the builder — a filterable map of onchain-agent components: what they are, what they're for, and how to access them. - [Use Cases](https://satohub.ai/use-cases): The directory mapped to what agents do — trading, payments, wallets, data, identity, privacy, launch. - [Wiki](https://satohub.ai/wiki): Plain-English explainers of the core onchain-agent concepts. - [News](https://satohub.ai/news): A filtered feed of onchain-agent releases, headlines, and research. - [Launches](https://satohub.ai/launches): What launched or shipped in onchain agents in the last 7 or 30 days — every row derived from when the listing was first added or from a dated release or post bound to it, carrying that event's own source link, chain, liveness and Sato Score. RSS: https://satohub.ai/launches/feed.xml - [Blog](https://satohub.ai/blog): Long-form, fully sourced analysis. No hype. - [Sato Score](https://satohub.ai/sato-score): How the transparency/liveness score is computed (methodology). - [ERC-8004](https://satohub.ai/resources/erc-8004): Live on-chain registered-agent count read from Ethereum mainnet. - [MCP server](https://satohub.ai/mcp): Query the directory, wiki, news, and metrics over the Model Context Protocol. - [Install](https://satohub.ai/mcp#install): Add the MCP server to any client — one click for Cursor, VS Code, LM Studio and Goose, the exact config block for Claude Code, Codex, Gemini CLI, Windsurf, Zed, Cline, Continue, JetBrains, Warp and Claude Desktop. Machine-readable at https://satohub.ai/api/install.json - [Sato Hub API](https://satohub.ai/docs/api): The same index over plain HTTP, no key needed to start — listings, full exports (JSON, CSV, NDJSON), Preflight checks, agent-economy data, JSON Schemas and the OpenAPI contract at https://satohub.ai/api/openapi.json, plus the paid x402 reads. - [Agent Instructions](https://satohub.ai/instructions): Drop-in rules files (AGENTS.md, CLAUDE.md, .cursor/rules, .windsurfrules) that tell a coding agent to use this server for onchain tooling questions and to cite the record it used. Raw files at https://satohub.ai/instructions/AGENTS.md and friends. - [Sato Route](https://satohub.ai/route): Where an agent's swap, x402 request or job should go — a non-custodial router that chooses on Sato Hub's dated readings, returns them with the answer, and discloses its fee before anything is signed. Sato Route swaps cost 3 bps stable-to-stable, 15 bps between majors (a chain's native coin such as ETH or SOL, its wrapped form, or a stablecoin), 75 bps with any other token and 25 bps cross-chain (75 bps when either side is any other token), taken as a parameter on the aggregator's own quote inside the transaction you sign; a failed, reverted or unsigned trade pays nothing. A token launch carries a disclosed share of the creator's LP fee, stated as fee.bps on every launch response, including when it is 0. Sato OS charges 3 / 15 bps on its own spot swap rail and a 3 bps perp builder fee, with no cross-chain tier. Some venues keep a share or pay later, and every quote says which. It never signs, holds or moves funds. API: https://satohub.ai/api/route/swap - [Sato Swap](https://satohub.ai/swap): Post a swap intent and get four verdicts — token in, token out, venue, recipient — each with the rule that produced it and the value it read, plus the policy that decided, the chosen venue with that venue’s own fee sentence, a simulation, and either an unsigned transaction or the reason one was withheld. Unknown refuses by default. Sato Hub never signs, holds or sends: what comes back is an unsigned object. API: https://satohub.ai/api/swap/quote - [Token resolver](https://satohub.ai/api/token/resolve?input=0x4ed4E862860beD51a9570b96d89aF5E1B0Efefed&chain=base): What an address, a Solana mint or a link (DexScreener, GeckoTerminal, Birdeye, pump.fun, GMGN, Basescan, Solscan, Jupiter, Uniswap, Aerodrome, Zora, Clanker) is — chain, name, symbol, decimals, token program, the main pool with its own liquidity and the USD price, each figure with its source and the time it was read; on Solana the mint’s authorities and Token-2022 extensions from the chain. A pool link resolves to the pool’s base token. It describes and recommends nothing. API: GET https://satohub.ai/api/token/resolve; MCP tool onchain_agent_resolve_token. - [Preflight](https://satohub.ai/preflight): One free call before an agent installs a tool, connects to an endpoint or pays one — go / caution / no / unknown for a repo, npm package, https endpoint or ERC-8004 agent, with every reading, its source field and its date. A description of evidence, never a safety verdict. API: https://satohub.ai/api/preflight - [Sato Check](https://satohub.ai/check): Before an agent installs a crypto package, MCP server or skill — does it take your key, does your key leave, can it move funds on its own, what changed since the last version. Every answer carries its evidence class: declared (the project says so), traced (a static read of the published artifact) or observed (a sandbox run with planted, never-funded test keys). A description, never a safety rating; unknown stays unknown. API: POST https://satohub.ai/api/check/install · GET https://satohub.ai/api/check?target= · MCP tool: onchain_agent_check_install · Report: https://satohub.ai/check/report/2026-09 - create-sato-agent: A runnable onchain agent repo from a goal in plain words, generated from a template re-checked every night (policy file, dependency custody lock, manifest; fork by default, mainnet only with explicit risk acceptance). The goal is not stored. API: POST https://satohub.ai/api/create · MCP tool: onchain_agent_create_agent · CLI: npx create-sato-agent@0.1 "" --framework plain-ts · Status: https://satohub.ai/status - [Preflight in CI](https://satohub.ai/preflight#run-it-in-ci): Preflight a whole dependency manifest in one call — POST https://satohub.ai/api/preflight/batch with a raw package.json or requirements.txt (max 50 targets) returns a verdict per target, a summary, an exit-code suggestion and SARIF findings for CI annotations. A manifest is never contacted, only looked up; `unknown` never fails a build. GitHub Action: https://github.com/satohubai/preflight-action. Embeddable badge: https://satohub.ai/api/badge/preflight/package/ and /repo//. - [x402 verified](https://satohub.ai/x402/verified): Which endpoints that declare x402 actually answer a payment request — probed daily for a real HTTP 402 with a parseable payload, plus facilitator reachability, declared network and manifest. A dated protocol check; no payment is sent. - [The sato.deploy manifest](https://satohub.ai/docs/deploy-spec): The install format Sato Hub publishes per listing — runtime, install, entry, what you must supply, license — so an agent can stand a project up without parsing a README. Schema: https://satohub.ai/schemas/deploy-spec.json - [Sato Swap — the method](https://satohub.ai/docs/sato-swap): how a venue is chosen, what the four verdicts mean, the policy model, the simulation lanes, the per-venue fee qualifiers, what a receipt binds, and what Sato Swap deliberately does not do. The surface is https://satohub.ai/swap; the API is POST https://satohub.ai/api/swap/quote and the MCP tool onchain_agent_swap. Every response is an unsigned object — no custody, no keys, no broadcasting. - [Sato Bot](https://satohub.ai/satobot): Describe an onchain agent in plain words and get a build plan — a stack of real directory listings with their Sato Scores, a Preflight verdict per component, the install line for anything publishing a deploy spec, and the first action the goal implies with its fee disclosed. It holds no keys and signs nothing. API: https://satohub.ai/api/satobot/plan · MCP tool: onchain_agent_build_plan - [How Sato Route chooses](https://satohub.ai/route/methodology): The ordered ranking criteria behind every swap, x402, agent and launch route — each signal, the field it reads and why it sits where it sits — plus the full fee schedule, the non-custodial guarantee, and the list of things a route cannot tell you. - [Preflight methodology](https://satohub.ai/preflight/methodology): Every rule a Preflight verdict can cite (R1-R6 for listings, T1-T6 for tokens), what each verdict is NOT, worked examples for each verdict, the data-source table, and why unknown never becomes no. - [House rules](https://satohub.ai/trust-rules): How Sato Hub handles evidence — self-reported is not verified, null is unknown and never zero, unknown never becomes no, never sum across grains, a death is a measurement, a disclosure describes, a claim grants no trust, we never hold keys. - [Transparency report](https://satohub.ai/transparency): What the catalog did, month by month, derived from its own write pipelines — listings added and retired, status and verification changes, corrections applied from verified claims, enrichment proposals staged against proposals applied, submissions decided, dead-project flags raised, plus a derived record of retractions and verification downgrades. - [Badges](https://satohub.ai/badges): Free embeddable Sato Score badge for any listed project — Markdown or HTML, no signup. ## Citable numbers and data - [How many ERC-8004 agents are registered on each chain?](https://satohub.ai/numbers/erc-8004-agents-by-chain): the registry's own agent-id counter on every chain the singleton is deployed to, each row with the date it was read and no total on purpose — a registration on one chain is a different object from one on another. - [How many x402 payment endpoints answer a payment request?](https://satohub.ai/numbers/x402-endpoints-that-answer): of the paid resources projects name in their own /.well-known/x402 manifest, how many answered a real HTTP 402 with a parseable payment payload; listing pages are counted apart, unreached endpoints are unknown, and no payment is ever sent. - [Which Uniswap fee tier do agent pairs trade most?](https://satohub.ai/numbers/uniswap-fee-tiers-for-agent-pairs): per pair and chain, the Uniswap v3 fee tier with the most swap volume we observed, with the days each pool was actually covered shown rather than extrapolated to a week. - [How many crypto MCP servers answer today?](https://satohub.ai/numbers/mcp-servers-that-answer): of the crypto MCP endpoints this directory lists, how many completed the initialize handshake in our latest daily check — a count and a share of our own checks, never called uptime. - [Do agents registered on ERC-8004 actually transact?](https://satohub.ai/numbers/erc-8004-agents-that-transact): of a sample of ERC-8004 owner wallets, how many sent a transaction in the trailing week — measured by nonce movement, published as a share of the sample and never of the registry, and withheld where the sample is too small. - [What does an x402 call actually cost?](https://satohub.ai/numbers/x402-price-per-call): the median price sellers DECLARE in their own 402 payloads, per capability and settlement network, in USDC only — a ceiling rather than a charge, since no payment is ever sent, with every non-USDC and testnet declaration excluded and counted rather than converted. - [How is the ERC-8004 registered-agent count trending on each chain?](https://satohub.ai/numbers/erc-8004-registered-agents): the same per-chain registry counter as a weekly series with the week-over-week change, one row per chain and never summed. - [Which x402 volume number is which?](https://satohub.ai/numbers/x402-volume-which-number): the headline x402 payment counts and value figures x402scan, x402.org and agenteconomy.to each publish, side by side with the source, what it counts, the chains and window it covers, the day we read it and why they differ — never added, never ranked. - [Which ERC-8004 agent count is which?](https://satohub.ai/numbers/erc-8004-agent-count-which-number): 8004scan's views, agenteconomy.to's table, the registry's own id counter per chain and the tools in this directory — different counts of ERC-8004, each with its definition and date, never summed across chains, with a per-chain registered-versus-functional range shown as unknown until it is read. - [How many registered ERC-8004 agents answered our probe?](https://satohub.ai/numbers/erc-8004-registered-vs-functional): per chain, the ids the registry has issued and the share of agents that answered our probe on the day shown as a floor and a ceiling, each with its n and whether the chain was read in full or as a sample of segments — never summed across chains, and shown as not enough readings yet where the sample is too small. - [Which MCP server count is which?](https://satohub.ai/numbers/mcp-server-count-which-number): the official MCP registry, Docker's MCP catalogue, awesome-mcp-servers, Hugging Face repos, skills.sh and ClawHub each count something different around MCP; independent censuses side by side with their definitions and dates, never merged or added. - [How many ERC-8004 agents are registered per day, by chain?](https://satohub.ai/numbers/erc-8004-registrations-per-day): new agent ids the ERC-8004 Identity Registry minted per UTC day, one row per chain with its own date, never summed — an id is minted by any register() call and is not a distinct agent. - [How many x402 settlements do facilitators send per day?](https://satohub.ai/numbers/x402-facilitator-throughput): transactions sent per UTC day by known x402 facilitator keys, per chain and per operator — an upper bound on settlements, not screened for self-payments, and never summed across chains. - [x402 facilitators](https://satohub.ai/x402/facilitators): the curated facilitator map with per-address evidence, plus our measured settlement share per facilitator on Base. - [Open methods](https://satohub.ai/methods): how Sato Hub reads x402 facilitators and counts outside payers — definitions, steps and limits; CC BY 4.0. - [Dated facts and history](https://satohub.ai/datasets#history): what the historic record keeps — a dated copy of each read of agent identity registries, payment listings, MCP and skill registries, package downloads, chain and contract facts and public blocklists. Over MCP: onchain_agent_get_chain_card, onchain_agent_get_contract_facts, onchain_agent_get_sdk_lineage and onchain_agent_get_x402_facilitators each return one dated fact with its source. - [Sources and credits](https://satohub.ai/data/sources): every outside source Sato Hub publishes from, with its credit line, its terms and the day it was last read. - [x402 verified](https://satohub.ai/x402/verified): every endpoint that declares x402, probed daily for a real HTTP 402 with a parseable payment payload — plus facilitator reachability, declared network and whether the host serves /.well-known/x402. A dated protocol check, not a safety or solvency claim; no payment is sent. - [Scored lists](https://satohub.ai/lists): "N Solana MCP servers, scored", "N x402-ready tools" and more — derived from live records with the date and the rule, so they cannot rot. - [Alternatives](https://satohub.ai/alternatives): "Alternatives to X" for every scored listing with at least three scored peers in its category — each peer with its Sato Score, liveness, chains, install line and a derived "differs on" line. Ordered by evidence, no winner declared. - [Embed](https://satohub.ai/embed): put a live, dated slice of the directory on your own page with one script tag — no dependencies, no third-party requests, attribution included. Data at https://satohub.ai/api/embed/stack.json. - [What moved this week](https://satohub.ai/movers): Sato Score risers, fallers, tier crossings and newly scored listings over seven days, with the component that moved. A week in which we revised the scoring method publishes no risers or fallers, and says so. Signed JSON at https://satohub.ai/api/movers.json. - Score history per listing: https://satohub.ai/verify//history — every daily Sato Score reading we have captured since 2026-06-12, each dated, gaps drawn as gaps and never interpolated, with our own scoring-method revisions marked apart from the project's own movement. - [Agent templates](https://satohub.ai/templates): every create-sato-agent template × framework with what it does, its chains and networks, the pinned lane's last nightly result and the create command. Machine copy at https://satohub.ai/api/create/templates. A sponsor label, where present, never changes which template is picked. - [Status](https://satohub.ai/status): is each listed MCP server answering today — our daily checks, 30-day cells, never called uptime. - [Glossary](https://satohub.ai/glossary): x402, facilitator, ERC-8004, ACP job, 4337 account, Sato Score — each term linking to the page that measures it. - [Datasets](https://satohub.ai/datasets): the free CC-BY exports (directory, agent economy, skills) as JSON/CSV/NDJSON with schema, license and coverage; 5 open datasets are catalogued below. - [Onchain Agent Builder Bench (Hugging Face)](https://huggingface.co/datasets/SatoHub/onchain-agent-builder-bench): Multiple-choice questions on building onchain AI agents: stack selection, protocol facts (ERC-8004, x402, ERC-4337, MCP, A2A), reading what a package does with a key, glossary and scope. The held-out test split is not published, only its hash. Ships lm-eval-harness and Inspect AI task files. CC-BY-4.0, "data by satohub.ai". - [Sato MCP Tool-Calling (Hugging Face)](https://huggingface.co/datasets/SatoHub/sato-mcp-tool-calling): Synthetic conversations in which an assistant uses Sato Hub's MCP tools correctly, with a share that shows when not to call Sato Hub at all. OpenAI-messages, Hermes and xlam formats. CC-BY-4.0, "data by satohub.ai". - [Pricing](https://satohub.ai/pricing): what is free and what is paid (the plans are Sato Check Team and Platform & Data), every limit, the fee schedule and the three data licences, in one place; machine copy at https://satohub.ai/api/pricing.json. - [Dataset catalogue](https://satohub.ai/datasets): every open dataset with its licence, cadence, coverage window, field list and a citation string; machine copy at https://satohub.ai/api/datasets.json. - [Numbers](https://satohub.ai/numbers): the index of every figure above, each with its stage, method, sample, date and the query that reproduces it. Includes the twenty findings of the State of Onchain Agents report. ## Use cases - [Trading](https://satohub.ai/use-cases/trading): Perps DEXs, swap routers, and execution SDKs — where a strategy becomes signed onchain orders. - [Payments](https://satohub.ai/use-cases/payments): Pay-per-call, agent-to-agent, settled onchain in stablecoins — no card, no checkout, no human. - [Wallets](https://satohub.ai/use-cases/wallets): Give an autonomous process signing power — without handing it the keys to everything. - [Data](https://satohub.ai/use-cases/data): An agent only acts on what it can see. This is how it sees onchain. - [Identity & Verification](https://satohub.ai/use-cases/identity): Can you trust the agent on the other side? This is how it proves who it is. - [Privacy](https://satohub.ai/use-cases/privacy): A public chain front-runs your strategy. This is how agents act without showing their hand. - [Launch](https://satohub.ai/use-cases/launch): Ship a new onchain agent in minutes — launchpads, no-code builders, and token rails. ## By chain - [Onchain agents on Base](https://satohub.ai/chains/base): The components for building AI agents on Base — agent frameworks, wallet infrastructure, x402 payment rails, MCP servers, and DEX tooling. Each scored on how open, active, and verifiable it is. - [Onchain agents on Solana](https://satohub.ai/chains/solana): Agent frameworks, wallet tooling, MCP servers, and DEX infrastructure for building AI agents on Solana — each scored on how open, active, and verifiable it is. - [Onchain agents on Ethereum](https://satohub.ai/chains/ethereum): Agent infrastructure on Ethereum mainnet — identity standards like ERC-8004, wallet and account-abstraction tooling, frameworks, and data access. Scored on openness, activity, and verifiability. - [Onchain agents on Arbitrum](https://satohub.ai/chains/arbitrum): Agent tooling on Arbitrum — perps and spot venues, wallet infrastructure, data access, and frameworks. Each scored on how open, active, and verifiable it is. - [Onchain agents on Polygon](https://satohub.ai/chains/polygon): Agent frameworks, wallets, payment rails, and data tooling on Polygon — each scored on how open, active, and verifiable it is. - [Onchain agents on Optimism](https://satohub.ai/chains/optimism): Agent infrastructure on Optimism — wallets, DeFi venues, data access, and frameworks, each scored on how open, active, and verifiable it is. - [Onchain agents on BNB Chain](https://satohub.ai/chains/bnb-chain): Agent tooling on BNB Chain — trading venues, wallets, payment rails, and frameworks. Each scored on how open, active, and verifiable it is. - [Onchain agents on Avalanche](https://satohub.ai/chains/avalanche): Agent frameworks, wallets, and DeFi tooling on Avalanche — each scored on how open, active, and verifiable it is. - [Onchain agents on Bitcoin](https://satohub.ai/chains/bitcoin): Agent tooling that reaches Bitcoin — wallets, data access, payment rails, and cross-chain infrastructure. Each scored on how open, active, and verifiable it is. ## By use case and chain - [Trading agents on Base](https://satohub.ai/use-cases/trading/base) - [Trading agents on Solana](https://satohub.ai/use-cases/trading/solana) - [Trading agents on Ethereum](https://satohub.ai/use-cases/trading/ethereum) - [Trading agents on Arbitrum](https://satohub.ai/use-cases/trading/arbitrum) - [Trading agents on Polygon](https://satohub.ai/use-cases/trading/polygon) - [Trading agents on BNB Chain](https://satohub.ai/use-cases/trading/bnb-chain) - [Payments agents on Base](https://satohub.ai/use-cases/payments/base) - [Payments agents on Solana](https://satohub.ai/use-cases/payments/solana) - [Payments agents on Ethereum](https://satohub.ai/use-cases/payments/ethereum) - [Wallets agents on Base](https://satohub.ai/use-cases/wallets/base) - [Wallets agents on Solana](https://satohub.ai/use-cases/wallets/solana) - [Wallets agents on Ethereum](https://satohub.ai/use-cases/wallets/ethereum) - [Data agents on Base](https://satohub.ai/use-cases/data/base) - [Data agents on Solana](https://satohub.ai/use-cases/data/solana) - [Data agents on Ethereum](https://satohub.ai/use-cases/data/ethereum) ## By standard - [x402: agent payments over HTTP](https://satohub.ai/standards/x402): Every tool in the directory that speaks x402 — the standard that lets agents pay for APIs in stablecoins over HTTP, with no account and no API key. Each scored on how open, active, and verifiable it is. - [MCP: the tool interface for agents](https://satohub.ai/standards/mcp): Onchain tooling that exposes a Model Context Protocol interface — the open standard for giving agents tools. Each listing scored on how open, active, and verifiable it is. - [ERC-8004: onchain agent identity](https://satohub.ai/standards/erc-8004): Projects working with ERC-8004, the Trustless Agents standard giving agents an onchain identity and reputation surface. Each scored on how open, active, and verifiable it is. - [A2A: agents delegating to agents](https://satohub.ai/standards/a2a): Tooling that implements Agent2Agent, the protocol for agents to discover each other and hand off tasks. Each listing scored on how open, active, and verifiable it is. ## Build guides One page per build goal, answered first: the stack layer by layer, each pick with its Sato Score and what it does with your keys and money. Index: https://satohub.ai/guides - [Build a Base trading agent that swaps USDC for ETH when a price signal fires](https://satohub.ai/guides/base-trading-agent-usdc-eth-price-signal): The stack for a Base trading agent that swaps USDC for ETH on a price signal: framework, wallet, trading venue and data feed. Not covered: the signal, sizing and returns. Each pick shows its Sato Score and what it does with your keys. - [Build a treasury balance monitor for Base addresses that sends alerts](https://satohub.ai/guides/base-treasury-balance-monitor): The stack for a read-only treasury balance monitor on Base: the data feed and an optional agent framework, with the Sato Score of each and a scaffold command for the watcher. - [Build an agent that sends USDC on Solana as a reward when a GitHub pull request is merged](https://satohub.ai/guides/solana-usdc-rewards-on-merged-pull-request): The stack for a Solana agent that pays USDC to a contributor when a GitHub pull request merges: framework, wallet and payment rail, each with its Sato Score and what it does with your keys and money. - [Build an agent that pays for APIs with USDC on Base](https://satohub.ai/guides/base-agent-pays-for-apis-usdc-x402): The stack for a Base agent that pays for APIs in USDC over x402: framework, wallet, payment rail and MCP tooling, each with its Sato Score and what it does with your keys and money. ## Scored lists Each list is a stated rule over the live directory, ordered by Sato Score (how open, active and verifiable a project is; not a safety or returns grade), with the date it was computed. Index: https://satohub.ai/lists - [Which trading agent frameworks are in the directory?](https://satohub.ai/lists/trading-agent-frameworks): Every agent framework in the Sato Hub directory that targets onchain trading or DeFi execution, ordered by Sato Score. - [Which onchain trading tools on Base are in the directory?](https://satohub.ai/lists/trading-tools-on-base): Every trading venue, execution SDK and strategy tool in the Sato Hub directory that supports Base, ordered by Sato Score. - [Which x402 payment tools on Base are in the directory?](https://satohub.ai/lists/x402-tools-on-base): Every x402-ready tool in the Sato Hub directory that supports Base, ordered by Sato Score. - [Which Solana MCP servers are in the directory?](https://satohub.ai/lists/solana-mcp-servers): Every MCP server in the Sato Hub directory that supports Solana, ordered by Sato Score. - [Which Base MCP servers are in the directory?](https://satohub.ai/lists/base-mcp-servers): Every MCP server in the Sato Hub directory that supports Base, ordered by Sato Score. - [Which Ethereum MCP servers are in the directory?](https://satohub.ai/lists/ethereum-mcp-servers): Every MCP server in the Sato Hub directory that supports Ethereum mainnet, ordered by Sato Score. - [Which x402-ready tools are in the directory?](https://satohub.ai/lists/x402-ready-tools): Every tool in the Sato Hub directory whose standards field includes x402, ordered by Sato Score. - [Which ERC-8004 tools are in the directory?](https://satohub.ai/lists/erc-8004-tools): Every tool in the Sato Hub directory whose standards field includes ERC-8004, ordered by Sato Score. - [Which MCP servers have a verified install?](https://satohub.ai/lists/mcp-servers-with-verified-install): MCP servers whose documented install path Sato Hub reproduced in an isolated container, ordered by Sato Score. - [Which onchain trading tools on Solana are in the directory?](https://satohub.ai/lists/trading-tools-on-solana): Every DEX aggregator, execution SDK and trading tool in the Sato Hub directory that supports Solana, ordered by Sato Score. - [Which npm-installable crypto agent tools are in the directory?](https://satohub.ai/lists/npm-packages-for-crypto-agents): Every listing whose documented install runs through npm, npx, yarn or pnpm, ordered by Sato Score, with the date of the last commit or release we observed. - [Which Python-installable crypto agent tools are in the directory?](https://satohub.ai/lists/python-packages-for-crypto-agents): Every listing whose documented install runs through pip, uv or poetry, ordered by Sato Score, with the last commit or release we observed. - [Which wallets for agents on Solana are in the directory?](https://satohub.ai/lists/wallets-for-agents-on-solana): Key management, smart accounts and server wallets in the Sato Hub directory that support Solana, ordered by Sato Score. - [Which x402 payment tools on Solana are in the directory?](https://satohub.ai/lists/x402-tools-on-solana): Every x402-ready listing in the Sato Hub directory that supports Solana, ordered by Sato Score. - [Which payment tools for agents are in the directory?](https://satohub.ai/lists/payment-tools-for-agents): Every listing matching the payments use case — stablecoin rails, x402 tooling, billing and settlement — ordered by Sato Score. - [Which listings have the highest Sato Scores in the directory?](https://satohub.ai/lists/highest-sato-score): Ordered by Sato Score, by the rule stated on the page. - [Which hosted MCP endpoints are answering today?](https://satohub.ai/lists/live-mcp-endpoints-answering): Every listing with a hosted MCP endpoint that completed Sato Hub's most recent daily check, ordered by Sato Score. - [Which Solana MCP servers you can use without a local server are in the directory?](https://satohub.ai/lists/hosted-solana-mcp-servers): Every MCP server in the Sato Hub directory that supports Solana and names a hosted endpoint that answered Sato Hub's most recent daily check, so nothing has to run on your machine. - [Which Base MCP servers you can use without a local server are in the directory?](https://satohub.ai/lists/hosted-base-mcp-servers): Every MCP server in the Sato Hub directory that supports Base and names a hosted endpoint that answered Sato Hub's most recent daily check, so nothing has to run on your machine. - [Which Ethereum MCP servers you can use without a local server are in the directory?](https://satohub.ai/lists/hosted-ethereum-mcp-servers): Every MCP server in the Sato Hub directory that supports Ethereum mainnet and names a hosted endpoint that answered Sato Hub's most recent daily check, so nothing has to run on your machine. - [What does an MCP server do with my wallet key?](https://satohub.ai/lists/what-mcp-servers-do-with-your-keys): MCP servers in the Sato Hub directory whose Sato Check reading of one version is published on its listing page: whether the code takes a key, whether a planted test key left in our sandbox, and whether it can move funds on its own. - [Which onchain agent frameworks have recent commits or releases?](https://satohub.ai/lists/maintained-agent-frameworks): Crypto-native agent frameworks in the Sato Hub directory whose latest commit or release is recent enough to count as Active, ordered by Sato Score, with the date of that commit or release. ## Measured - [The reliability index](https://satohub.ai/reliability): every listing ranked by the share of Sato Hub's daily checks it answered over at least 14 days, with MCP tool-inventory movement and whether the documented install was reproduced in a container. Not uptime. - [Crypto agent skills, disclosed](https://satohub.ai/skills): agent skills from ClawHub and skills.sh that touch crypto, each with a static disclosure of the hosts it contacts, whether it generates keys, asks for credentials, pipes the internet into a shell, or registers the agent with a third party. Descriptions, never safety verdicts. - [The onchain agent economy](https://satohub.ai/agent-economy): weekly on-chain measurements per venue, chain and lifecycle stage — registries, launchpads, payment rails, account infrastructure. Nothing on it adds up, on purpose. - [State of Onchain Agents](https://satohub.ai/state-of-onchain-agents): twenty citable numbers, each with its stage, method, sample, date and the query that reproduces it. - [Numbers](https://satohub.ai/numbers): one page per figure. Cite the page, not the number alone — the method is part of the claim. ## Questions, answered Each is answered by one deterministic function over the live records, with the date the data was refreshed. Index: https://satohub.ai/answers - [What should I build a trading agent with?](https://satohub.ai/answers/what-should-i-build-a-trading-agent-with): Sato Hub's live records fill a trading agent stack with agent framework Coinbase AgentKit. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a trading agent with on Base?](https://satohub.ai/answers/what-should-i-build-a-trading-agent-with-on-base): Sato Hub's live records fill a trading agent stack on Base with agent framework Coinbase AgentKit. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a trading agent with on Solana?](https://satohub.ai/answers/what-should-i-build-a-trading-agent-with-on-solana): Sato Hub's live records fill a trading agent stack on Solana with agent framework Almanak. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a trading agent with on Ethereum?](https://satohub.ai/answers/what-should-i-build-a-trading-agent-with-on-ethereum): Sato Hub's live records fill a trading agent stack on Ethereum with agent framework Coinbase AgentKit. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a payments agent with?](https://satohub.ai/answers/what-should-i-build-a-payments-agent-with): Sato Hub's live records fill a payments agent stack with agent framework Coinbase AgentKit, wallet & keys Bankr. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a payments agent with on Base?](https://satohub.ai/answers/what-should-i-build-a-payments-agent-with-on-base): Sato Hub's live records fill a payments agent stack on Base with agent framework Coinbase AgentKit. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a payments agent with on Solana?](https://satohub.ai/answers/what-should-i-build-a-payments-agent-with-on-solana): Sato Hub's live records fill a payments agent stack on Solana with agent framework Almanak, wallet & keys Bankr. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a payments agent with on Ethereum?](https://satohub.ai/answers/what-should-i-build-a-payments-agent-with-on-ethereum): Sato Hub's live records fill a payments agent stack on Ethereum with agent framework Coinbase AgentKit. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a self-custodial agent with?](https://satohub.ai/answers/what-should-i-build-a-wallets-agent-with): Sato Hub's live records fill a self-custodial agent stack with agent framework Olas (Autonolas). Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a self-custodial agent with on Base?](https://satohub.ai/answers/what-should-i-build-a-wallets-agent-with-on-base): Sato Hub's live records fill a self-custodial agent stack on Base with agent framework Olas (Autonolas). Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a self-custodial agent with on Solana?](https://satohub.ai/answers/what-should-i-build-a-wallets-agent-with-on-solana): Sato Hub's live records fill a self-custodial agent stack on Solana with agent framework Almanak. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a self-custodial agent with on Ethereum?](https://satohub.ai/answers/what-should-i-build-a-wallets-agent-with-on-ethereum): Sato Hub's live records fill a self-custodial agent stack on Ethereum with agent framework Olas (Autonolas). Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a data agent with?](https://satohub.ai/answers/what-should-i-build-a-data-agent-with): Sato Hub's live records fill a data agent stack with agent framework Coinbase AgentKit, wallet & keys Bankr. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a data agent with on Base?](https://satohub.ai/answers/what-should-i-build-a-data-agent-with-on-base): Sato Hub's live records fill a data agent stack on Base with agent framework Coinbase AgentKit, wallet & keys Bankr. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a data agent with on Solana?](https://satohub.ai/answers/what-should-i-build-a-data-agent-with-on-solana): Sato Hub's live records fill a data agent stack on Solana with agent framework Almanak, wallet & keys Bankr. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [What should I build a data agent with on Ethereum?](https://satohub.ai/answers/what-should-i-build-a-data-agent-with-on-ethereum): Sato Hub's live records fill a data agent stack on Ethereum with agent framework Coinbase AgentKit. Ranked on Sato Score, liveness and whether we reproduced the install, not on quality, as of 2026-10-30. - [Which agent frameworks can an onchain trading agent be built on?](https://satohub.ai/answers/which-trading-agent-frameworks): 17 listings qualify: Coinbase AgentKit, Olas (Autonolas), TradingAgents, ElizaOS and 13 more. 10 open source, 5 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which trading tools can an agent use on Base?](https://satohub.ai/answers/which-trading-tools-on-base): 82 listings qualify: Coinbase AgentKit, Olas (Autonolas), ElizaOS, Bankr and 78 more. 36 open source, 27 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which x402 payment tools work on Base?](https://satohub.ai/answers/which-x402-tools-on-base): 66 listings qualify: Coinbase AgentKit, BlockRun MCP, NanoParse, Skim and 62 more. 27 open source, 25 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which MCP servers give an agent access to Solana?](https://satohub.ai/answers/which-solana-mcp-servers): 41 listings qualify: BlockRun MCP, Vybe Solana MCP, Zerion and 38 more. 17 open source, 15 active in the last 30 days, 7 with a hosted endpoint that answered our check, as of 2026-10-30. - [Which MCP servers give an agent access to Base?](https://satohub.ai/answers/which-base-mcp-servers): 56 listings qualify: BlockRun MCP, NanoParse, Skim and 53 more. 27 open source, 19 active in the last 30 days, 15 with a hosted endpoint that answered our check, as of 2026-10-30. - [Which MCP servers give an agent access to Ethereum?](https://satohub.ai/answers/which-ethereum-mcp-servers): 40 listings qualify: Zerion, Uniswap, SODAX Builders MCP, PayRam MCP and 36 more. 21 open source, 12 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which tools support x402 payments?](https://satohub.ai/answers/which-x402-ready-tools): 90 listings qualify: Coinbase AgentKit, BlockRun MCP, NanoParse, Skim and 86 more. 40 open source, 33 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which tools implement ERC-8004?](https://satohub.ai/answers/which-erc-8004-tools): 2 of 17 ERC-8004 listings are tagged for registration or scaffolding (Pinata ERC-8004, create-8004-agent). Across all 17, 0 active in the last 30 days. Tags are declared by each project, as of 2026-10-30. - [Which crypto agent frameworks have the most recent code activity?](https://satohub.ai/answers/which-agent-frameworks-scored): 24 crypto-native agent frameworks are listed. Latest commit or release: Provek (2026-10-30), Olas (Autonolas) (2026-10-29), ElizaOS (2026-10-29). Sato Score is context, not a safety grade, as of 2026-10-30. - [How do I give an AI agent a wallet?](https://satohub.ai/answers/which-wallets-for-agents): To give an agent a wallet, choose where its key lives. Of 35 wallet listings, 10 declare self-custody, 6 MPC, 7 account abstraction, 5 embedded wallets. Tags are declared, not observed, and overlap, as of 2026-10-30. - [Which crypto MCP servers have had their installation independently verified?](https://satohub.ai/answers/which-mcp-servers-with-verified-install): 69 of the 221 crypto MCP servers in the directory have an install Sato Hub reproduced in a container, not an audit. Most recently re-run (2026-10-26), by Sato Score: BlockRun MCP, Zerion, BNB Chain MCP, as of 2026-10-30. - [Which DEX and trading tools can an agent use on Solana?](https://satohub.ai/answers/which-trading-tools-on-solana): 50 listings qualify: Raydium, ElizaOS, Bankr, Zerion and 46 more. 22 open source, 21 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which npm packages exist for building crypto agents?](https://satohub.ai/answers/which-npm-packages-for-crypto-agents): 50 listings qualify: Raydium, OpenClaw, BlockRun MCP, Skim and 46 more. 35 open source, 47 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which Python packages exist for building crypto agents?](https://satohub.ai/answers/which-python-packages-for-crypto-agents): 46 listings qualify: Hermes Agent, Olas (Autonolas), TradingAgents, x402 and 42 more. 32 open source, 20 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which wallet infrastructure works for a Solana agent?](https://satohub.ai/answers/which-wallets-for-agents-on-solana): 18 listings qualify: Bankr, Turnkey, Privy, Vultisig and 14 more. 6 open source, 9 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which x402 tools work on Solana?](https://satohub.ai/answers/which-x402-tools-on-solana): 37 listings qualify: BlockRun MCP, x402, Bankr, Zerion and 33 more. 15 open source, 17 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [What security tools can an agent call before it signs a transaction?](https://satohub.ai/answers/which-security-tools-for-agents): 17 listings qualify: PulseFeed, Agent Skills Hub, Self Protocol, GoPlus AgentGuard and 13 more. 8 open source, 8 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which payment tools can an agent use?](https://satohub.ai/answers/which-payment-tools-for-agents): 40 listings qualify: BlockRun MCP, Olas (Autonolas), NanoParse, Skim and 36 more. 27 open source, 36 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools score highest on openness, activity and verifiability?](https://satohub.ai/answers/which-highest-sato-score): 25 listings qualify: Raydium, OpenClaw, Hermes Agent, Coinbase AgentKit and 21 more. 20 open source, 25 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which crypto MCP endpoints are answering a handshake today?](https://satohub.ai/answers/which-live-mcp-endpoints-answering): 52 listings qualify: NanoParse, CoinGecko MCP, Skim, minia2a and 48 more. 25 open source, 23 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Is Coinbase AgentKit still maintained?](https://satohub.ai/answers/is-coinbase-agentkit-still-maintained): The most recent commit or release Sato Hub observed for Coinbase AgentKit was 2026-10-14, which it labels Active. Code activity is a recency signal, not a quality or security judgment, as of 2026-10-30. - [Is Solana Agent Kit still maintained?](https://satohub.ai/answers/is-solana-agent-kit-still-maintained): The most recent commit or release Sato Hub observed for Solana Agent Kit was 2026-05-14, which it labels Quiet. The latest release in the feed Sato Hub reads was 2025-07-24, which can be a pre-release, as of 2026-10-30. - [Is GOAT SDK still maintained?](https://satohub.ai/answers/is-goat-sdk-still-maintained): The most recent commit or release Sato Hub observed for GOAT SDK was 2026-07-02, which it labels Quiet. Code activity is a recency signal, not a quality or security judgment, as of 2026-10-30. - [Is ElizaOS still maintained?](https://satohub.ai/answers/is-elizaos-still-maintained): The most recent commit or release Sato Hub observed for ElizaOS was 2026-10-29, which it labels Active. The latest release in the feed Sato Hub reads was 2026-10-05, which can be a pre-release, as of 2026-10-30. - [Is Virtuals Protocol still maintained?](https://satohub.ai/answers/is-virtuals-protocol-still-maintained): The most recent commit or release Sato Hub observed for Virtuals Protocol was 2026-09-23, which it labels Recent. Code activity is a recency signal, not a quality or security judgment, as of 2026-10-30. - [Is Jupiter still maintained?](https://satohub.ai/answers/is-jupiter-aggregator-still-maintained): The most recent commit or release Sato Hub observed for Jupiter was 2026-08-04, which it labels Recent. Code activity is a recency signal, not a quality or security judgment, as of 2026-10-30. - [Is Privy still maintained?](https://satohub.ai/answers/is-privy-still-maintained): The most recent commit or release Sato Hub observed for Privy was 2026-10-26, which it labels Active. Code activity is a recency signal, not a quality or security judgment, as of 2026-10-30. - [Is Turnkey still maintained?](https://satohub.ai/answers/is-turnkey-still-maintained): The most recent commit or release Sato Hub observed for Turnkey was 2026-10-24, which it labels Active. Code activity is a recency signal, not a quality or security judgment, as of 2026-10-30. - [Is x402 still maintained?](https://satohub.ai/answers/is-x402-still-maintained): The most recent commit or release Sato Hub observed for x402 was 2026-10-29, which it labels Active. The latest release in the feed Sato Hub reads was 2026-10-28, which can be a pre-release, as of 2026-10-30. - [Is Clanker still maintained?](https://satohub.ai/answers/is-clanker-still-maintained): Sato Hub has no commit or release date for Clanker, so whether it is maintained is unknown rather than no, as of 2026-10-30. - [How many x402 tools are there?](https://satohub.ai/answers/how-many-x402-tools-are-there): Sato Hub lists 92 listings that implement x402: 40 open source, 33 active in the last 30 days, 43 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many Model Context Protocol tools are there?](https://satohub.ai/answers/how-many-mcp-tools-are-there): Sato Hub lists 201 listings that implement Model Context Protocol: 116 open source, 76 active in the last 30 days, 156 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many ERC-8004 tools are there?](https://satohub.ai/answers/how-many-erc-8004-tools-are-there): Sato Hub lists 17 listings that implement ERC-8004: 11 open source, 8 carrying a deploy spec, 5 with an install we reproduced. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many Agent2Agent tools are there?](https://satohub.ai/answers/how-many-a2a-tools-are-there): Sato Hub lists 14 listings that implement Agent2Agent: 10 open source, 4 active in the last 30 days, 9 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many onchain agent tools support Base?](https://satohub.ai/answers/how-many-agent-tools-support-base): Sato Hub lists 169 listings that support Base: 72 open source, 58 active in the last 30 days, 96 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many onchain agent tools support Solana?](https://satohub.ai/answers/how-many-agent-tools-support-solana): Sato Hub lists 120 listings that support Solana: 57 open source, 51 active in the last 30 days, 81 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many onchain agent tools support Ethereum?](https://satohub.ai/answers/how-many-agent-tools-support-ethereum): Sato Hub lists 125 listings that support Ethereum: 58 open source, 40 active in the last 30 days, 79 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many onchain agent tools support Arbitrum?](https://satohub.ai/answers/how-many-agent-tools-support-arbitrum): Sato Hub lists 61 listings that support Arbitrum: 25 open source, 23 active in the last 30 days, 42 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many onchain agent tools support Polygon?](https://satohub.ai/answers/how-many-agent-tools-support-polygon): Sato Hub lists 58 listings that support Polygon: 25 open source, 22 active in the last 30 days, 37 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many onchain agent tools support Optimism?](https://satohub.ai/answers/how-many-agent-tools-support-optimism): Sato Hub lists 38 listings that support Optimism: 15 open source, 16 active in the last 30 days, 25 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many onchain agent tools support BNB Chain?](https://satohub.ai/answers/how-many-agent-tools-support-bnb-chain): Sato Hub lists 46 listings that support BNB Chain: 13 open source, 14 active in the last 30 days, 27 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many onchain agent tools support Avalanche?](https://satohub.ai/answers/how-many-agent-tools-support-avalanche): Sato Hub lists 28 listings that support Avalanche: 9 open source, 11 active in the last 30 days, 19 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many onchain agent tools support Bitcoin?](https://satohub.ai/answers/how-many-agent-tools-support-bitcoin): Sato Hub lists 32 listings that support Bitcoin: 19 open source, 13 active in the last 30 days, 19 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many crypto MCP servers are there?](https://satohub.ai/answers/how-many-crypto-mcp-servers-are-there): Sato Hub lists 221 crypto MCP servers: 132 open source, 79 active in the last 30 days, 52 answering a handshake today. A count of what this directory holds, not of what exists, as of 2026-10-30. - [How many onchain agent tools are open source?](https://satohub.ai/answers/how-many-onchain-agent-tools-are-open-source): Sato Hub lists 255 onchain agent listings that declare an open-source licence: 113 active in the last 30 days, 204 carrying a deploy spec. A count of this directory, not of what exists, as of 2026-10-30. - [How many onchain agent tools have an independently reproduced install?](https://satohub.ai/answers/how-many-onchain-agent-tools-have-a-reproduced-install): Sato Hub lists 129 listings whose documented install Sato Hub reproduced: 93 open source, 69 exposing an MCP interface. A count of this directory, not of what exists, as of 2026-10-30. - [What onchain agent tools exist for trading?](https://satohub.ai/answers/what-tools-exist-for-agent-trading): Sato Hub lists 208 listings matching the trading use case: 110 open source, 67 active in the last 30 days, 138 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [What onchain agent tools exist for payments?](https://satohub.ai/answers/what-tools-exist-for-agent-payments): Sato Hub lists 171 listings matching the payments use case: 77 open source, 63 active in the last 30 days, 91 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [What onchain agent tools exist for wallets?](https://satohub.ai/answers/what-tools-exist-for-agent-wallets): Sato Hub lists 140 listings matching the wallets use case: 70 open source, 52 active in the last 30 days, 102 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [What onchain agent tools exist for data?](https://satohub.ai/answers/what-tools-exist-for-agent-data): Sato Hub lists 154 listings matching the data use case: 75 open source, 54 active in the last 30 days, 116 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [What onchain agent tools exist for identity & verification?](https://satohub.ai/answers/what-tools-exist-for-agent-identity): Sato Hub lists 71 listings matching the identity use case: 41 open source, 25 active in the last 30 days, 41 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [What onchain agent tools exist for privacy?](https://satohub.ai/answers/what-tools-exist-for-agent-privacy): Sato Hub lists 31 listings matching the privacy use case: 8 open source, 9 active in the last 30 days, 14 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [What onchain agent tools exist for launch?](https://satohub.ai/answers/what-tools-exist-for-agent-launch): Sato Hub lists 25 listings matching the launch use case: 4 open source, 4 active in the last 30 days, 10 carrying a deploy spec. A count of what this directory holds, not of what exists, as of 2026-10-30. - [Coinbase AgentKit or Solana Agent Kit — which should I use?](https://satohub.ai/answers/coinbase-agentkit-or-solana-agent-kit): Coinbase AgentKit and Solana Agent Kit on the same axes from live records: Sato Score 91 (High) vs 76 (High), Base, Ethereum, Multichain vs Solana. No winner is declared, as of 2026-10-30. - [x402 or AP2 — which one does an agent need?](https://satohub.ai/answers/x402-or-ap2): x402 and Agent Payments Protocol (AP2) on the same axes: Sato Score 87 (High) vs 72 (High). No winner is declared, as of 2026-10-30. - [Jupiter or 1inch — which should an agent route a swap through?](https://satohub.ai/answers/jupiter-or-1inch): Jupiter and 1inch on the same axes: Sato Score 77 (High) vs 77 (High). No winner is declared, as of 2026-10-30. - [Clanker or Pons — where should an agent launch a token?](https://satohub.ai/answers/clanker-or-pons): Declared chains do not overlap: Clanker lists Base, Arbitrum, BNB Chain; Pons lists Robinhood Chain. No winner is declared, as of 2026-10-30. - [ElizaOS or Almanak — which should I build a trading agent on?](https://satohub.ai/answers/elizaos-or-almanak): ElizaOS and Almanak on the same axes: Sato Score 87 (High) vs 83 (High). No winner is declared, as of 2026-10-30. - [Privy or Turnkey — which should hold an agent's keys?](https://satohub.ai/answers/privy-or-turnkey): Privy and Turnkey on the same axes: Sato Score 84 (High) vs 86 (High), both on Ethereum, Base, Solana, Multichain. No winner is declared, as of 2026-10-30. - [What are the alternatives to Raydium?](https://satohub.ai/answers/what-are-the-alternatives-to-raydium): Sato Hub lists 39 scored Trading Tool listings alongside Raydium, highest first: Drift Protocol, as of 2026-10-30. - [What are the alternatives to BlockRun MCP?](https://satohub.ai/answers/what-are-the-alternatives-to-blockrun-mcp): Sato Hub lists 143 scored MCP listings alongside BlockRun MCP, highest first: Coinbase x402 Bazaar (Discovery API + MCP), as of 2026-10-30. - [What are the alternatives to Coinbase AgentKit?](https://satohub.ai/answers/what-are-the-alternatives-to-coinbase-agentkit): Sato Hub lists 38 scored Developer Tool listings alongside Coinbase AgentKit, highest first: Universal Contract AI Interface (UCAI), as of 2026-10-30. - [What are the alternatives to TradingAgents?](https://satohub.ai/answers/what-are-the-alternatives-to-tradingagents): Sato Hub lists 30 scored Agent Framework listings alongside TradingAgents, highest first: Hermes Agent, as of 2026-10-30. - [What are the alternatives to ElizaOS?](https://satohub.ai/answers/what-are-the-alternatives-to-elizaos): Sato Hub lists 30 scored Agent Framework listings alongside ElizaOS, highest first: Olas (Autonolas), as of 2026-10-30. - [What are the alternatives to Fetch.ai uAgents?](https://satohub.ai/answers/what-are-the-alternatives-to-fetchai-uagents): Sato Hub lists 30 scored Agent Framework listings alongside Fetch.ai uAgents, highest first: Olas (Autonolas), as of 2026-10-30. - [What are the alternatives to GOAT SDK?](https://satohub.ai/answers/what-are-the-alternatives-to-goat-sdk): Sato Hub lists 30 scored Skill Repo listings alongside GOAT SDK, highest first: Solana Agent Kit, as of 2026-10-30. - [What are the alternatives to Olas?](https://satohub.ai/answers/what-are-the-alternatives-to-olas): Sato Hub lists 30 scored Agent Framework listings alongside Olas (Autonolas), highest first: ElizaOS, as of 2026-10-30. - [What are the alternatives to 0xGasless AgentKit?](https://satohub.ai/answers/what-are-the-alternatives-to-0xgasless-agentkit): Sato Hub lists 34 scored Wallet Infrastructure listings alongside 0xGasless AgentKit, highest first: Trust Wallet Agent Kit (TWAK), as of 2026-10-30. - [What are the alternatives to Trust Wallet Agent Kit?](https://satohub.ai/answers/what-are-the-alternatives-to-trust-wallet-agent-kit): Sato Hub lists 34 scored Wallet Infrastructure listings alongside Trust Wallet Agent Kit (TWAK), highest first: Swapper Toolkit, as of 2026-10-30. - [What are the alternatives to UCAI?](https://satohub.ai/answers/what-are-the-alternatives-to-ucai): Sato Hub lists 38 scored Developer Tool listings alongside Universal Contract AI Interface (UCAI), highest first: Coinbase AgentKit, as of 2026-10-30. - [What are the alternatives to World AgentKit?](https://satohub.ai/answers/what-are-the-alternatives-to-world-agentkit): Sato Hub lists 46 scored API / SDK listings alongside World AgentKit, highest first: Flaunch. 25 of them open source. Ranked on openness, activity and verifiability, not quality; no winner is declared, as of 2026-10-30. - [What are the alternatives to x402?](https://satohub.ai/answers/what-are-the-alternatives-to-x402): Sato Hub lists 23 scored Payment Rail listings alongside x402, highest first: Stripe Machine Payments, as of 2026-10-30. - [What are the alternatives to Zerion?](https://satohub.ai/answers/what-are-the-alternatives-to-zerion): Sato Hub lists 35 scored Data Tool listings alongside Zerion, highest first: Nansen. 10 of them open source. Ranked on openness, activity and verifiability, not quality; no winner is declared, as of 2026-10-30. - [What are the alternatives to DexPaprika MCP?](https://satohub.ai/answers/what-are-the-alternatives-to-dexpaprika-mcp): Sato Hub lists 143 scored MCP listings alongside DexPaprika MCP Server, highest first: TradingView MCP Server, as of 2026-10-30. - [What are the alternatives to Hermes Agent?](https://satohub.ai/answers/what-are-the-alternatives-to-hermes-agent): Sato Hub lists 30 scored Agent Framework listings alongside Hermes Agent, highest first: OpenAI Codex, as of 2026-10-30. - [What are the alternatives to Jupiter?](https://satohub.ai/answers/what-are-the-alternatives-to-jupiter-aggregator): Sato Hub lists 39 scored Trading Tool listings alongside Jupiter, highest first: Raydium. 17 of them open source. Ranked on openness, activity and verifiability, not quality; no winner is declared, as of 2026-10-30. - [What are the alternatives to OpenClaw?](https://satohub.ai/answers/what-are-the-alternatives-to-openclaw): Sato Hub lists 30 scored Agent Framework listings alongside OpenClaw, highest first: Hermes Agent, as of 2026-10-30. - [What are the alternatives to Alchemy?](https://satohub.ai/answers/what-are-the-alternatives-to-alchemy): Sato Hub lists 46 scored API / SDK listings alongside Alchemy, highest first: Enso, PMXT. 25 of them open source. Ranked on openness, activity and verifiability, not quality; no winner is declared, as of 2026-10-30. - [What are the alternatives to Almanak?](https://satohub.ai/answers/what-are-the-alternatives-to-almanak): Sato Hub lists 30 scored Agent Framework listings alongside Almanak, highest first: ElizaOS. 20 of them open source. Ranked on openness, activity and verifiability, not quality; no winner is declared, as of 2026-10-30. - [What are the alternatives to dYdX?](https://satohub.ai/answers/what-are-the-alternatives-to-dydx-chain): Sato Hub lists 39 scored Trading Tool listings alongside dYdX, highest first: Drift Protocol. 16 of them open source. Ranked on openness, activity and verifiability, not quality; no winner is declared, as of 2026-10-30. - [What are the alternatives to Orca?](https://satohub.ai/answers/what-are-the-alternatives-to-orca): Sato Hub lists 39 scored Trading Tool listings alongside Orca, highest first: Raydium. 16 of them open source. Ranked on openness, activity and verifiability, not quality; no winner is declared, as of 2026-10-30. - [What are the alternatives to Privy?](https://satohub.ai/answers/what-are-the-alternatives-to-privy): Sato Hub lists 34 scored Wallet Infrastructure listings alongside Privy, highest first: Turnkey, as of 2026-10-30. - [What are the alternatives to Turnkey?](https://satohub.ai/answers/what-are-the-alternatives-to-turnkey): Sato Hub lists 34 scored Wallet Infrastructure listings alongside Turnkey, highest first: Privy, as of 2026-10-30. - [What are the alternatives to Model Context Protocol?](https://satohub.ai/answers/what-are-the-alternatives-to-model-context-protocol): Sato Hub lists 38 scored Developer Tool listings alongside Model Context Protocol (MCP), highest first: MCP Reference Servers, as of 2026-10-30. - [What shipped in the last 30 days for trading?](https://satohub.ai/answers/what-shipped-recently-in-agent-trading): 36 trading listings moved in the last 30 days: DefiLlama, dYdX, Kraken CLI. 9 newly listed, 27 shipped a release. An observed event with a date and a source, never an endorsement, as of 2026-10-30. - [What shipped in the last 30 days for payments?](https://satohub.ai/answers/what-shipped-recently-in-agent-payments): 49 payments listings moved in the last 30 days: DefiLlama, BlockRun MCP, x402. 23 newly listed, 26 shipped a release. An observed event with a date and a source, never an endorsement, as of 2026-10-30. - [What shipped in the last 30 days for wallets?](https://satohub.ai/answers/what-shipped-recently-in-agent-wallets): 29 wallets listings moved in the last 30 days: Insumer MCP Server, Tari MCP Servers. 8 newly listed, 21 shipped a release. An observed event with a date and a source, never an endorsement, as of 2026-10-30. - [What shipped in the last 30 days for data?](https://satohub.ai/answers/what-shipped-recently-in-agent-data): 27 data listings moved in the last 30 days: DefiLlama, CoinPaprika MCP, dYdX. 3 newly listed, 24 shipped a release. An observed event with a date and a source, never an endorsement, as of 2026-10-30. - [What shipped in the last 30 days for identity & verification?](https://satohub.ai/answers/what-shipped-recently-in-agent-identity): 21 identity & verification listings moved in the last 30 days: ClawHub, OpenClaw. 14 newly listed, 7 shipped a release. An observed event with a date and a source, never an endorsement, as of 2026-10-30. - [What shipped in the last 30 days for privacy?](https://satohub.ai/answers/what-shipped-recently-in-agent-privacy): 12 privacy listings moved in the last 30 days: Insumer MCP Server, Turnkey, AkashML. 9 newly listed, 3 shipped a release. An observed event with a date and a source, never an endorsement, as of 2026-10-30. - [What shipped in the last 30 days for launch?](https://satohub.ai/answers/what-shipped-recently-in-agent-launch): 3 launch listings moved in the last 30 days: Doppler Protocol, pump.fun, Zora Coins. 3 newly listed, 0 shipped a release. An observed event with a date and a source, never an endorsement, as of 2026-10-30. - [Which agent skills exist for swapping tokens?](https://satohub.ai/answers/which-agent-skills-exist-for-swapping-tokens): 106 indexed agent skills match swapping tokens, from 2 registries. 50 carry at least one disclosure flag. A disclosure describes what a skill declares and does from a static scan — it is not a verdict, as of 2026-10-26. - [Which agent skills exist for reading onchain data?](https://satohub.ai/answers/which-agent-skills-exist-for-reading-onchain-data): 136 indexed agent skills match reading onchain data, from 2 registries. 61 carry a disclosure flag; a disclosure is not a verdict, as of 2026-10-26. - [Which agent skills exist for checking token prices?](https://satohub.ai/answers/which-agent-skills-exist-for-checking-token-prices): 133 indexed agent skills match checking token prices, from 2 registries. 50 carry a disclosure flag; a disclosure is not a verdict, as of 2026-10-26. - [Which agent skills exist for managing a wallet?](https://satohub.ai/answers/which-agent-skills-exist-for-managing-a-wallet): 145 indexed agent skills match managing a wallet, from 2 registries. 89 carry a disclosure flag; a disclosure is not a verdict, as of 2026-10-26. - [Which agent skills exist for paying per request with x402?](https://satohub.ai/answers/which-agent-skills-exist-for-paying-per-request-with-x402): 81 indexed agent skills match paying per request with x402, from 2 registries. 38 carry a disclosure flag; a disclosure is not a verdict, as of 2026-10-26. - [Which agent skills exist for working with NFTs?](https://satohub.ai/answers/which-agent-skills-exist-for-working-with-nfts): 28 indexed agent skills match working with NFTs, from 2 registries. 16 carry at least one disclosure flag. A disclosure describes what a skill declares and does from a static scan — it is not a verdict, as of 2026-10-26. - [Which agent skills exist for deploying a smart contract?](https://satohub.ai/answers/which-agent-skills-exist-for-deploying-a-smart-contract): 102 indexed agent skills match deploying a smart contract, from 2 registries. 49 carry a disclosure flag; a disclosure is not a verdict, as of 2026-10-26. - [Which agent skills exist for launching a token?](https://satohub.ai/answers/which-agent-skills-exist-for-launching-a-token): 13 indexed agent skills match launching a token, from 2 registries. 7 carry at least one disclosure flag. A disclosure describes what a skill declares and does from a static scan — it is not a verdict, as of 2026-10-26. - [How can I check what an MCP server or npm package can do with my keys before I install it?](https://satohub.ai/answers/how-can-i-check-what-an-mcp-server-or-npm-package-can-do-with-my-keys): Paste the package, MCP endpoint or install command into Sato Check (/check) or call onchain_agent_check_install; it reports key access, key egress and fund actions, as of 2026-10-30. - [What does an MCP server do with my wallet key?](https://satohub.ai/answers/what-does-an-mcp-server-do-with-my-wallet-key): Of 88 MCP servers with a published Sato Check reading, 16 read private-key material in code (signing locally is what a wallet does), 14 ask for a key in setup and 58 show no key read found, as of 2026-10-30. - [Should you give an MCP server your wallet private key?](https://satohub.ai/answers/should-you-give-an-mcp-server-your-wallet-private-key): Of 88 MCP servers with a published Sato Check reading, 30 take a key (16 read it, 14 ask in setup); 11 have a planted-key run with none seen leaving; 4 list an action that can sign or send funds, as of 2026-10-30. - [What should I check before installing @raydium-io/raydium-sdk-v2?](https://satohub.ai/answers/what-to-check-before-installing-raydium-io-raydium-sdk-v2): Sato Hub's record for Raydium documents npm install @raydium-io/raydium-sdk-v2. The install was reproduced in a container. No security review was performed, as of 2026-10-30. - [What should I check before installing @blockrun/mcp?](https://satohub.ai/answers/what-to-check-before-installing-blockrun-mcp): Sato Hub's record for BlockRun MCP documents npm install @blockrun/mcp, an install it reproduced in a container on 2026-10-26, and a commit or release on 2026-10-28. No security review was performed, as of 2026-10-30. - [What should I check before installing @goat-sdk/core?](https://satohub.ai/answers/what-to-check-before-installing-goat-sdk-core): Sato Hub's record for GOAT SDK documents npm install @goat-sdk/core, an install it reproduced in a container on 2026-10-26, and a commit or release on 2026-07-02. No security review was performed, as of 2026-10-30. - [What should I check before installing open-autonomy?](https://satohub.ai/answers/what-to-check-before-installing-open-autonomy): Sato Hub's record for Olas (Autonolas) documents pip install open-autonomy. The install was reproduced in a container. No security review was performed, as of 2026-10-30. - [What should I check before installing uagents?](https://satohub.ai/answers/what-to-check-before-installing-uagents): Sato Hub's record for Fetch.ai uAgents documents pip install uagents, an install it reproduced in a container on 2026-10-26, and a commit or release on 2026-10-26. No security review was performed, as of 2026-10-30. - [What should I check before installing x402-fetch?](https://satohub.ai/answers/what-to-check-before-installing-x402-fetch): Sato Hub's record for x402 documents npm install x402-fetch, an install it reproduced in a container on 2026-10-26, and a commit or release on 2026-10-29. No security review was performed, as of 2026-10-30. - [What should I check before installing @trustwallet/cli?](https://satohub.ai/answers/what-to-check-before-installing-trustwallet-cli): Sato Hub's record for Trust Wallet Agent Kit (TWAK) documents npm install @trustwallet/cli. The install was reproduced in a container. No security review was performed, as of 2026-10-30. - [What should I check before installing @worldcoin/agentkit?](https://satohub.ai/answers/what-to-check-before-installing-worldcoin-agentkit): Sato Hub's record for World AgentKit documents npm install @worldcoin/agentkit. The install was reproduced in a container. No security review was performed, as of 2026-10-30. - [What should I check before installing @0xgasless/agentkit?](https://satohub.ai/answers/what-to-check-before-installing-0xgasless-agentkit): Sato Hub's record for 0xGasless AgentKit documents npm install @0xgasless/agentkit. The install was reproduced in a container. No security review was performed, as of 2026-10-30. - [What should I check before installing abi-to-mcp?](https://satohub.ai/answers/what-to-check-before-installing-abi-to-mcp): Sato Hub's record for Universal Contract AI Interface (UCAI) documents pip install abi-to-mcp. The install was reproduced in a container. No security review was performed, as of 2026-10-30. - [What should I check before installing zerion-cli?](https://satohub.ai/answers/what-to-check-before-installing-zerion-cli): Sato Hub's record for Zerion documents npm install zerion-cli, an install it reproduced in a container on 2026-10-26, and a commit or release on 2026-10-13. No security review was performed, as of 2026-10-30. - [What should I check before installing openclaw?](https://satohub.ai/answers/what-to-check-before-installing-openclaw): Sato Hub's record for OpenClaw documents npm install openclaw, an install it reproduced in a container on 2026-10-19, and a commit or release on 2026-10-30. No security review was performed, as of 2026-10-30. - [What should I check before installing hermes-agent?](https://satohub.ai/answers/what-to-check-before-installing-hermes-agent): Sato Hub's record for Hermes Agent documents pip install hermes-agent, an install it reproduced in a container on 2026-09-25, and a commit or release on 2026-10-29. No security review was performed, as of 2026-10-30. - [What should I check before installing @jup-ag/api?](https://satohub.ai/answers/what-to-check-before-installing-jup-ag-api): Sato Hub's record for Jupiter documents npm install @jup-ag/api, an install it reproduced in a container on 2026-09-21, and a commit or release on 2026-08-04. No security review was performed, as of 2026-10-30. - [What should I check before installing dexpaprika-mcp?](https://satohub.ai/answers/what-to-check-before-installing-dexpaprika-mcp): Sato Hub's record for DexPaprika MCP Server documents npm install dexpaprika-mcp. The install was reproduced in a container. No security review was performed, as of 2026-10-30. - [What should I check before installing @privy-io/node?](https://satohub.ai/answers/what-to-check-before-installing-privy-io-node): Sato Hub's record for Privy documents npm install @privy-io/node, an install it reproduced in a container on 2026-10-26, and a commit or release on 2026-10-26. No security review was performed, as of 2026-10-30. - [What should I check before installing @turnkey/sdk-server?](https://satohub.ai/answers/what-to-check-before-installing-turnkey-sdk-server): Sato Hub's record for Turnkey documents npm install @turnkey/sdk-server, an install it reproduced in a container on 2026-10-26, and a commit or release on 2026-10-24. No security review was performed, as of 2026-10-30. - [What should I check before installing @alchemy/wallet-apis?](https://satohub.ai/answers/what-to-check-before-installing-alchemy-wallet-apis): Sato Hub's record for Alchemy documents npm install @alchemy/wallet-apis, an install it reproduced in a container on 2026-10-26. No security review was performed, as of 2026-10-30. - [What should I check before installing @dydxprotocol/v4-client-js?](https://satohub.ai/answers/what-to-check-before-installing-dydxprotocol-v4-client-js): Sato Hub's record for dYdX documents npm install @dydxprotocol/v4-client-js, an install it has not reproduced, and a commit or release on 2026-10-29. No security review was performed, as of 2026-10-30. - [What should I check before installing @drift-labs/sdk?](https://satohub.ai/answers/what-to-check-before-installing-drift-labs-sdk): Sato Hub's record for Drift Protocol documents npm install @drift-labs/sdk. The install was reproduced in a container. No security review was performed, as of 2026-10-30. - [Which onchain agent tools support x402 on Ethereum?](https://satohub.ai/answers/which-tools-support-x402-on-ethereum): 31 Sato Hub listings implement x402 and support Ethereum: Coinbase AgentKit, Bankr, x402 and 28 more. 15 open source, 11 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support x402 on Arbitrum?](https://satohub.ai/answers/which-tools-support-x402-on-arbitrum): 15 Sato Hub listings implement x402 and support Arbitrum: Bankr, Zerion, Nansen and 12 more. 6 open source, 7 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support x402 on Polygon?](https://satohub.ai/answers/which-tools-support-x402-on-polygon): 16 Sato Hub listings implement x402 and support Polygon: Bankr, Zerion, Nansen and 13 more. 6 open source, 6 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support x402 on Optimism?](https://satohub.ai/answers/which-tools-support-x402-on-optimism): 9 Sato Hub listings implement x402 and support Optimism: Zerion, Nansen, OpenSea Agent Skill and 6 more. 3 open source, 4 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support x402 on BNB Chain?](https://satohub.ai/answers/which-tools-support-x402-on-bnb-chain): 12 Sato Hub listings implement x402 and support BNB Chain: Bankr, Zerion, Bitget Wallet Skill and 9 more. 5 open source, 4 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support x402 on Avalanche?](https://satohub.ai/answers/which-tools-support-x402-on-avalanche): 6 Sato Hub listings implement x402 and support Avalanche: Zerion, 0xGasless AgentKit, Trust Wallet Agent Kit (TWAK) and 3 more. 2 open source, 1 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support x402 on Bitcoin?](https://satohub.ai/answers/which-tools-support-x402-on-bitcoin): 7 Sato Hub listings implement x402 and support Bitcoin: Bankr, x402, AIBTC MCP Server and 4 more. 5 open source, 3 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support Model Context Protocol on Arbitrum?](https://satohub.ai/answers/which-tools-support-mcp-on-arbitrum): 15 Sato Hub listings implement Model Context Protocol and support Arbitrum: Zerion, Uniswap, OpenSea Agent Skill and 12 more. 7 open source, 6 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support Model Context Protocol on Polygon?](https://satohub.ai/answers/which-tools-support-mcp-on-polygon): 17 Sato Hub listings implement Model Context Protocol and support Polygon: Zerion, Uniswap, PayRam MCP and 14 more. 8 open source, 8 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support Model Context Protocol on Optimism?](https://satohub.ai/answers/which-tools-support-mcp-on-optimism): 11 Sato Hub listings implement Model Context Protocol and support Optimism: Zerion, Uniswap, OpenSea Agent Skill and 8 more. 6 open source, 6 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support Model Context Protocol on BNB Chain?](https://satohub.ai/answers/which-tools-support-mcp-on-bnb-chain): 13 Sato Hub listings implement Model Context Protocol and support BNB Chain: Zerion, BNB Chain MCP, PayRam MCP and 10 more. 5 open source, 4 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support Model Context Protocol on Avalanche?](https://satohub.ai/answers/which-tools-support-mcp-on-avalanche): 7 Sato Hub listings implement Model Context Protocol and support Avalanche: Zerion, 1inch, Philidor MCP and 4 more. 3 open source, 3 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support Model Context Protocol on Bitcoin?](https://satohub.ai/answers/which-tools-support-mcp-on-bitcoin): 14 Sato Hub listings implement Model Context Protocol and support Bitcoin: Bitcoin MCP (Bortlesboat) and 13 more. 11 open source, 6 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support ERC-8004 on Base?](https://satohub.ai/answers/which-tools-support-erc-8004-on-base): 11 Sato Hub listings implement ERC-8004 and support Base: 0xGasless AgentKit, ERC-8004: Trustless Agents and 9 more. 7 open source, 0 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support ERC-8004 on Ethereum?](https://satohub.ai/answers/which-tools-support-erc-8004-on-ethereum): 9 Sato Hub listings implement ERC-8004 and support Ethereum: ERC-8004: Trustless Agents, Emblem Agent Skills and 7 more. 7 open source, 0 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support ERC-8004 on BNB Chain?](https://satohub.ai/answers/which-tools-support-erc-8004-on-bnb-chain): 5 Sato Hub listings implement ERC-8004 and support BNB Chain: 0xGasless AgentKit, Emblem Agent Skills, Pinata ERC-8004 and 2 more. 3 open source, 0 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support Agent2Agent on Base?](https://satohub.ai/answers/which-tools-support-a2a-on-base): 8 Sato Hub listings implement Agent2Agent and support Base: Bankr, Agent Payments Protocol (AP2), Emblem Agent Skills and 5 more. 6 open source, 1 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support Agent2Agent on Solana?](https://satohub.ai/answers/which-tools-support-a2a-on-solana): 5 Sato Hub listings implement Agent2Agent and support Solana: Bankr, Emblem Agent Skills, PayAI Network and 2 more. 3 open source, 1 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which onchain agent tools support Agent2Agent on Ethereum?](https://satohub.ai/answers/which-tools-support-a2a-on-ethereum): 6 Sato Hub listings implement Agent2Agent and support Ethereum: Bankr, Agent Payments Protocol (AP2), Emblem Agent Skills and 3 more. 5 open source, 1 active in the last 30 days, ordered by Sato Score, as of 2026-10-30. - [Which crypto MCP servers can I use without running a local server?](https://satohub.ai/answers/which-crypto-mcp-servers-are-hosted): Sato Hub lists 100 hosted crypto MCP servers: NanoParse, CoinGecko MCP, Skim and 97 more. Hosted means no local install; 52 answered our latest handshake check, 25 asked for credentials or declined, as of 2026-10-30. - [Which agent wallets are self-custodial?](https://satohub.ai/answers/which-agent-wallets-are-self-custodial): Sato Hub lists 10 agent wallets that declare self-custody: Vultisig, Open Wallet Standard (OWS) and 8 more. Declared by each project; Sato Hub did not observe key handling, as of 2026-10-30. - [How can an AI agent pay another agent?](https://satohub.ai/answers/how-can-an-ai-agent-pay-another-agent): Pay per request over x402 (HTTP 402, then a signed stablecoin payment); hold a job's payment in escrow with ERC-8183. Sato Hub lists 92 implementing x402 and 6 ERC-8183, as of 2026-10-30. - [What is ERC-8004?](https://satohub.ai/answers/what-is-erc-8004): ERC-8004 (Trustless Agents) is an Ethereum standard giving an agent an on-chain identity, with registries for identity, reputation and validation. Sato Hub lists 17 live listings implementing it, as of 2026-10-30. - [Which Solana MCP servers can I use without running a local server?](https://satohub.ai/answers/which-solana-mcp-servers-are-hosted): Sato Hub lists 7 hosted Solana MCP servers that answered our check: Vybe Solana MCP, Exa, PayRam MCP and 4 more. 4 active in the last 30 days; 2 of them declare a data use case, as of 2026-10-30. - [How do I build an agent that pays contributors USDC when a pull request merges?](https://satohub.ai/answers/how-do-i-build-an-agent-that-pays-contributors-usdc-when-a-pull-request-merges): On Solana, Sato Hub ranks ElizaOS (agent framework), Bankr (wallet & keys) and x402 (payments). Ranked on Sato Score, liveness and a reproduced install, not quality, as of 2026-10-30. - [Which x402 endpoints actually work?](https://satohub.ai/answers/which-x402-endpoints-actually-work): 10 of the 12 paid x402 endpoints Sato Hub reached answered HTTP 402 with valid payment terms in the last 7 days. Nothing was paid, so delivery is not measured, as of 2026-10-30. - [Which x402 endpoints answered a paid request recently?](https://satohub.ai/answers/which-x402-endpoints-answered-a-paid-request-recently): Sato Hub does not observe a paid request being answered: its daily check sends no payment. 10 of 12 paid x402 endpoints replied to an unpaid request with valid HTTP 402 terms in the last 7 days, as of 2026-10-30. - [How does Sato Swap choose a venue?](https://satohub.ai/answers/how-does-sato-swap-choose-a-venue): Sato Swap ranks venues on whether they are live, reliable and checked, and only then on quoted price, as of 2026-09-15. - [What does a Sato Swap verdict mean?](https://satohub.ai/answers/what-does-a-sato-swap-verdict-mean): Token in, token out, venue and recipient each get go, caution, no or unknown, with the rule behind it, as of 2026-09-15. - [What fee does Sato Swap charge, and who receives it?](https://satohub.ai/answers/what-fee-does-sato-swap-charge-and-who-receives-it): Sato Swap charges 3 bps stablecoin to stablecoin, 15 bps between majors, 75 bps with any other token and 25 bps cross-chain, paid to Sato Hub inside the venue's own router, as of 2026-09-15. - [What does a Sato Swap receipt prove?](https://satohub.ai/answers/what-does-a-sato-swap-receipt-prove): A signed record of what Sato Swap returned and checked; it does not prove a fee landed or a trade settled, as of 2026-09-15. - [Does Sato Swap hold my funds or keys?](https://satohub.ai/answers/does-sato-swap-hold-my-funds-or-keys): No. Sato Swap only returns unsigned transactions for your own wallet to sign, and never holds keys or funds, as of 2026-09-15. - [Why did Sato Swap refuse my trade?](https://satohub.ai/answers/why-did-sato-swap-refuse-my-trade): Every refusal names the rule that fired and the values it compared; anything unknown refuses by default, as of 2026-09-15. - [Which chains and venues does Sato Swap route?](https://satohub.ai/answers/which-chains-and-venues-does-sato-swap-route): Sato Swap asks 9 venues across 14 chains (13 EVM chains plus Solana), and 3 of them also quote cross-chain, as of 2026-09-15. - [Is there an MCP tool for swaps?](https://satohub.ai/answers/is-there-an-mcp-tool-for-swaps): Yes: onchain_agent_swap runs the same engine as POST /api/swap/quote and returns unsigned transactions only, as of 2026-09-15. - [How can an AI agent swap tokens?](https://satohub.ai/answers/how-can-an-ai-agent-swap-tokens): It asks for an unsigned transaction, which is checked, held to its owner's rules and simulated, then signs it with its own wallet, as of 2026-09-15. - [What stops an agent from making a swap it shouldn't?](https://satohub.ai/answers/what-stops-an-agent-from-making-a-swap-it-should-not): Rules its owner sets, enforced on the server before anything is built: spending caps, allowlists and slippage limits, as of 2026-09-15. - [How does Sato Swap handle slippage and deadlines?](https://satohub.ai/answers/how-does-sato-swap-handle-slippage-and-deadlines): Your slippage and deadline go straight to the venue's router, and a policy can cap them; over the cap, the trade is refused, as of 2026-09-15. - [Can an agent swap across chains with Sato Swap?](https://satohub.ai/answers/can-an-agent-swap-across-chains-with-sato-swap): Yes: 3 of the 9 venues quote cross-chain swaps, ranked separately from same-chain quotes, as of 2026-09-15. - [How do I call Sato Swap from code?](https://satohub.ai/answers/how-do-i-call-sato-swap-from-code): POST a JSON request to /api/swap/quote, or call the MCP tool onchain_agent_swap; no key is needed, as of 2026-09-15. - [What happens when Sato Swap simulates a transaction?](https://satohub.ai/answers/what-happens-when-sato-swap-simulates-a-transaction): It test-runs the transaction first and holds back any route that would fail, with the reason, as of 2026-09-15. - [What does Sato Swap not do?](https://satohub.ai/answers/what-does-sato-swap-not-do): It never holds keys, signs or sends transactions, prices tokens or reviews venues, as of 2026-09-15. - [How is Sato Swap different from using an aggregator directly?](https://satohub.ai/answers/how-is-sato-swap-different-from-routing-directly-through-an-aggregator): It asks several aggregators and adds checks, owner rules, a test run and a signed record, at the cost of a fee and one extra step, as of 2026-09-15. - [What does Sato Swap cost?](https://satohub.ai/answers/what-does-sato-swap-cost): No key or account needed; a trade pays 3 bps stablecoin to stablecoin, 15 bps between majors, 75 bps with any other token or 25 bps cross-chain, plus venue fees and gas, as of 2026-09-15. - [Who can set a Sato Swap policy?](https://satohub.ai/answers/who-can-set-a-sato-swap-policy): The owner of the API key, agent or wallet it is tied to; without one, a default applies that refuses anything unknown, as of 2026-09-15. - [Sato Swap vs AgentSwap: what does each one check?](https://satohub.ai/answers/sato-swap-vs-agentswap-what-each-checks): Different layers: AgentSwap's record describes trading from the owner's wallet, while Sato Swap checks a trade and returns an unsigned transaction, as of 2026-09-15. ## Head-to-head - [Coinbase AgentKit vs Solana Agent Kit](https://satohub.ai/compare/coinbase-agentkit-vs-solana-agent-kit): Coinbase AgentKit and Solana Agent Kit compared on the same axes — chains, openness, maintenance, and install verification — from live Sato Hub directory data. No winner declared. - [x402 vs AP2](https://satohub.ai/compare/x402-vs-ap2): x402 and Google's Agent Payments Protocol compared on scope, chains, standards support, and maintenance — from live Sato Hub directory data. They overlap less than the framing suggests. - [Jupiter vs 1inch](https://satohub.ai/compare/jupiter-vs-1inch): Jupiter and 1inch compared on the same axes — chains, openness, maintenance, interfaces and install verification — from live Sato Hub directory data. No winner declared. - [Clanker vs Pons](https://satohub.ai/compare/clanker-vs-pons): Clanker and Pons compared on the same axes — what each one is, chains, openness, maintenance and verification — from live Sato Hub directory data. They solve different problems. - [ElizaOS vs Almanak](https://satohub.ai/compare/elizaos-vs-almanak): ElizaOS and Almanak compared on the same axes — chains, openness, maintenance, interfaces and install verification — from live Sato Hub directory data. No winner declared. - [Privy vs Turnkey](https://satohub.ai/compare/privy-vs-turnkey): Privy and Turnkey compared on the same axes — chains, openness, maintenance, interfaces and verification — from live Sato Hub directory data. No winner declared, and neither is audited by us. ## Data access - [MCP server](https://satohub.ai/api/mcp): 48 tools over Streamable HTTP (as of 2026-10-30) — 42 read-only, plus 6 that act (onchain_agent_watch, onchain_agent_submit_project, onchain_agent_register_agent, onchain_agent_join_passport, onchain_agent_passport_action, onchain_agent_swap). No key. Docs: https://satohub.ai/mcp - [MCP server for coding assistants](https://satohub.ai/api/dev/mcp): a smaller endpoint with the same data and no key, for the moments a coding assistant repeats: before it adds a package or an MCP server that touches wallets or keys (onchain_agent_check_install), when it needs a contract address (onchain_agent_get_contract_facts), and when it picks an SDK, wallet or payment rail for an onchain agent (onchain_agent_recommend_stack). It does not quote, route or prepare swaps, trades or token launches. Config lines for Claude Code, Codex and Cursor, and the endpoint URL for GitHub Copilot CLI: https://satohub.ai/mcp#coding-assistants - `onchain_agent_preflight`: what is on record about a repo, npm package, https endpoint, ERC-8004 agent or ERC-20 token before you install, connect, pay or trade — go / caution / no / unknown, with every reading, its source field and its date. - `onchain_agent_route_swap`: which venue a swap should go to, chosen on liveness, our observed check record, verification state and then quoted output, returned with the aggregator's own calldata and the fee disclosed before signing. - `onchain_agent_swap`: a swap that can refuse — the venue chosen and its own fee sentence quoted verbatim, both tokens, the venue endpoint and the recipient each checked under your policy, a simulation run, and then either an UNSIGNED transaction or the reason one was withheld. Unknown refuses by default. Sato Hub holds no keys, signs nothing and broadcasts nothing. - `onchain_agent_route_agent`: which registered agent should take a task, ranked on what the passport declares and what answered a live handshake — a registration is a claim, not proof the product works. - `onchain_agent_route_launch`: which venue a token launch should go to (Clanker v4, Bankr, Virtuals, Zora creator coins), with a config prepared for you to read and sign yourself. - `onchain_agent_build_plan`: a goal in plain words in, a build plan out — a stack of real listings with Sato Score, liveness and deploy spec, a Preflight verdict per item, the first action, the open questions. - `onchain_agent_find_x402_service`: describe a paid API or data service an agent needs and get indexed x402 services back, ranked by what Sato Hub observed (x402-style payments from outside payers, then whether the URL answered HTTP 402), each with its dated readings and a canonical page at `sato_url`. Zero matches return guidance, never an invented hit. Read-only; nothing here pays or calls a service. Same results over HTTP: https://satohub.ai/api/x402/services?need=… - `onchain_agent_submit_project`: put a project in front of the directory — the same row the web form files, into the same queue, probed from public sources and then listed, held for a person, or declined with the reason. A duplicate returns what already exists. Submission is not verification. - `onchain_agent_register_agent`: an agent issues itself a passport — fetch the challenge, sign it with the agent's own wallet key, call again and the passport and its manifest are live. The signature proves control of the key, never anything about the agent; every passport is Self-Reported. - `onchain_agent_join_passport`: a free Sato Passport signed with the agent's own wallets, EVM and Solana, with no account and no review queue. Call it once with no signatures and the answer is the exact message each wallet must sign, with nothing written; sign, call again, and the passport is public. A signature proves control of that wallet's key, never anything about the agent; every passport is Self-Reported; joining earns no points. A passport made this way is left out of search engines and the public list until one of its wallets has sent an onchain transaction. Same step over HTTP: POST https://satohub.ai/api/passport/join. - `onchain_agent_passport_action`: the signed changes to an existing passport, by a wallet already on it: add_wallet, prove_wallet, remove_wallet, set_primary, update_profile, issue_claim, unlink_owner. Unsigned, it answers the message to sign. Same step over HTTP: POST /api/passport/action. - `onchain_agent_search_listings` / `onchain_agent_inspect_listing`: published agent packages — each a public view of one sealed version, with its license, fork permission and evidence freshness for that exact digest. Read-only. A package is a whole working agent, sealed and signed, that you download and run yourself; the tools return no download URL, the package page at `sato_url` has it, with a script to verify the download against the published key. - [API status](https://satohub.ai/status/sato-hub): Whether the surfaces on this list are answering. Sato Hub checks its own MCP endpoint, exports, change feed, Preflight, signing key, OpenAPI document, badges, manifests and route quotes once an hour and publishes the success ratio and latency of those checks. Machine-readable at https://satohub.ai/api/status.json; daily rollups at https://satohub.ai/status/history.json; incident feed at https://satohub.ai/status/feed.xml. It is a record of our own checks, not a service-level agreement. - [Data freshness](https://satohub.ai/api/freshness): How old each data surface is — directory activity, news, Sato Score, liveness checks, agent-economy measurements, x402, skills — with the last successful collection from the pipeline ledger, the expected cadence (daily or weekly) and whether it is late. Also the MCP tool onchain_agent_get_data_freshness. A surface with no recorded success is unknown, never given a date. - [OpenAPI](https://satohub.ai/api/openapi.json): Machine-readable contract for the public reads, the agent-callable passport endpoints, and the claim flow. Admin, billing and licensing are deliberately absent. - [Sato Passport join](https://satohub.ai/agents): Give an agent a public Sato Passport with its own wallets, free: POST https://satohub.ai/api/passport/join with no signature returns the exact message to sign, then send it back signed (EVM or Solana). Also the MCP tool `onchain_agent_join_passport` and, in the Sato Agent kit, `sato-agent join`. A signature proves control of a wallet's key, nothing about the agent. A passport made this way is public, left out of search and the public list until its wallet has sent an onchain transaction, and earns no points. - [Directory export](https://satohub.ai/api/export/index.json): The full catalog as JSON. Also .csv and .ndjson; slice with ?category= / ?chain= / ?standard= / ?use_case=. - [Score inputs export](https://satohub.ai/api/export/score-inputs.json): Every input each published Sato Score is computed from — exactly the fields the scoring function reads — with as_of, the instant those scores were computed against. Run them through the published rubric and you should get the published number; the scorer is a verbatim copy of ours in the Sato Hub skill. A reproduced score is checkable, not a safety or audit judgment. - [Published schemas](https://satohub.ai/schemas): Every machine-readable shape here as JSON Schema (draft 2020-12) at a stable URL — export listing, change feed, sato.deploy manifest, Agent Passport manifest, Preflight verdict, Sato Route receipt, signed-response envelope, build plan, skill disclosure, score inputs, score roots. Catalogue: https://satohub.ai/schemas/index.json. v1 is additive-only; a breaking change becomes v2 at a new URL. - [Deploy spec format](https://satohub.ai/docs/deploy-spec): the sato.deploy manifest — runtime, install, entry, and what deploy_status means. Schema: https://satohub.ai/schemas/deploy-spec.json (draft 2020-12). Every export entry carries its deploy_spec; "verified" means we reproduced the install in a container on that date, nothing more. - [Sato Swap method](https://satohub.ai/docs/sato-swap): the four checked targets, the venue ladder, the policy model, simulation lanes, and the fee. Sato Route swaps cost 3 bps stable-to-stable, 15 bps between majors (a chain's native coin such as ETH or SOL, its wrapped form, or a stablecoin), 75 bps with any other token and 25 bps cross-chain (75 bps when either side is any other token), taken as a parameter on the aggregator's own quote inside the transaction you sign; a failed, reverted or unsigned trade pays nothing. A token launch carries a disclosed share of the creator's LP fee, stated as fee.bps on every launch response, including when it is 0. Sato OS charges 3 / 15 bps on its own spot swap rail and a 3 bps perp builder fee, with no cross-chain tier. Some venues keep a share or pay later, and every quote says which. Unknown refuses by default; a refusal names its rule. - [Agent-economy export](https://satohub.ai/api/export/adoption.json): This week's measurements per venue, chain and stage, with the rules in the payload (never add rows together; null means unknown). Also .csv and .ndjson. - [x402 verified export](https://satohub.ai/api/export/x402-verified.json): Every URL probed daily for listings tagged x402, with what it is (a paid resource named in the project's own /.well-known/x402, an MCP endpoint, or a listing page), what it answered, and where the payment payload was read. No payment is sent; null means unknown. - [ERC-8004 agents export](https://satohub.ai/api/export/erc8004-agents.json): Every ERC-8004 agent id on Celo with what its registration file declares (MCP, A2A, x402) and whether a declared endpoint answered Sato Hub's daily check; also as CSV. Per-agent reports at https://satohub.ai/erc8004/celo. null means never checked, not failed. CC-BY-4.0. - [MCP handshakes export](https://satohub.ai/api/export/mcp-handshakes.json): Every listed crypto MCP endpoint URL with the outcome of a daily anonymous initialize + tools/list check — answered, asked for credentials, or no handshake — and the tool count. A credential gate is a design choice, not a failure. - [Skills export](https://satohub.ai/api/export/skills.json): Crypto agent skills with their static disclosures. Also .csv and .ndjson. A disclosure describes; it is not a safety verdict. - [Change feed](https://satohub.ai/api/changes?since=YYYY-MM-DD): What moved since a date, so you don't re-pull the whole index. Read its `coverage` block — it states which parts are complete and which are partial. Each answer carries `next_since`, the date to pass back as `since` next time (see "Watch a list" below). The directory export takes the same `?since=` and returns just the listings added or changed since that date. - [Change webhooks](https://satohub.ai/api/webhooks/subscribe): the same lanes pushed daily to an endpoint you control, signed with your own secret, instead of polled. Your endpoint proves itself by echoing a challenge first; seven consecutive failures disables the subscription. - [Capability manifest](https://satohub.ai/.well-known/ard.json): Agentic Resource Discovery entry — what this domain can do, for registries that crawl by intent. - [Security posture](https://satohub.ai/security): what the MCP server does and does not do, and how to report an issue. - License: catalog data CC-BY-4.0, attribution "data by satohub.ai". Scores and liveness are informational signals, not safety guarantees. Every export answers with a `Link: rel="license"` header, and the datasets carry a `cite_as` line (field and `X-Sato-Cite-As` header) to paste as the credit; add the date you read it. ## Watch a list Keep a fixed list of listings current on a schedule, without reading each one again. 1. Keep your list of listing slugs and a cursor date. To start, use a date a few days back. 2. On your schedule, ask what changed since your cursor. Once a day is enough: the index is rebuilt daily. 3. Look for your slugs in added, updated, changed and removed. Then store the next_since from the answer as your cursor. MCP: onchain_agent_get_changes {"since":"2026-10-27"} HTTP, what changed: curl -sS 'https://satohub.ai/api/changes?since=2026-10-27' HTTP, the changed listings in full: curl -sS 'https://satohub.ai/api/export/index.json?since=2026-10-27' - next_since overlaps the last window by a couple of days and leaves no gap, so dedupe by slug. - Read the coverage block in the answer: it says which lanes are complete and which are approximate. - The rows call returns the listings added or changed since that date. Listings retired in the window are in removed, here and in the JSON delta block. - Store next_since from the same rows response you consumed (delta.next_since in the JSON, or the X-Sato-Delta header), not from a separate /api/changes call: the two are cached for different times and can disagree on the day. - If a date is older than the 90-day window or cannot be compared, the whole index is served, labelled not-applied in X-Sato-Delta and delta.applied. Treat that answer as a full refresh of your copy. - A filtered mirror (for example chain=base with since) also needs delta.left_slice: listings that changed and are no longer in your slice. Drop them from the mirror if you hold them. - Prefer mail to polling? onchain_agent_watch emails you when a Preflight verdict changes. Free while in preview. ## Glossary Definitions of the onchain-agent vocabulary — each says what the thing is, how it works and what it is NOT, cites a primary source, and links the Sato Hub page that measures it. Index: https://satohub.ai/glossary - [A2A (Agent2Agent)](https://satohub.ai/glossary/a2a): A protocol for agents to delegate work to each other: each publishes a card saying what it does and how to reach it, and tasks are handed over, not called. - [ACP job (Virtuals Agent Commerce Protocol)](https://satohub.ai/glossary/acp-job): A unit of paid work between two agents on Virtuals' ACP: requested, negotiated, funded into escrow, delivered, then evaluated — each step an on-chain state. - [Agent card](https://satohub.ai/glossary/agent-card): A JSON document at /.well-known/agent-card.json describing what an agent can do, where to reach it and how to authenticate — A2A's discovery unit. - [Agent Passport](https://satohub.ai/glossary/agent-passport): A Sato Hub profile and manifest for a running agent, registered by proving control of its wallet key. Self-reported except where a named check says otherwise. - [Agent skill](https://satohub.ai/glossary/agent-skill): A packaged instruction set (typically a SKILL.md plus scripts) an agent loads to gain a capability; installed from registries such as ClawHub or skills.sh. - [Bonding curve launch](https://satohub.ai/glossary/bonding-curve-launch): A token sale priced by a formula against supply: each buy mints higher, each sell burns lower, with no order book and no counterparty to find. - [Deploy spec / verified install](https://satohub.ai/glossary/deploy-spec): A machine-readable install manifest for a listing; the install is verified when Sato Hub re-ran it in an isolated container and it completed. - [ERC-3009 transfer authorization](https://satohub.ai/glossary/erc-3009-authorization): A signed message authorising a token transfer that someone else submits and pays gas for — the mechanism most x402 payments settle with. - [ERC-4337 account (smart account)](https://satohub.ai/glossary/erc-4337-account): A contract wallet whose transactions are UserOperations, sent by bundlers through a shared EntryPoint — programmable validation with no protocol change. - [ERC-7702 delegation](https://satohub.ai/glossary/erc-7702-delegation): A transaction type letting an ordinary externally-owned account temporarily run contract code, so a plain wallet can behave like a smart account. - [ERC-7715 permission](https://satohub.ai/glossary/erc-7715-permission): A scoped permission granted by a wallet — spend up to X of Y until T — requested via wallet_grantPermissions and enforced by the wallet's own code. - [ERC-8004 (Trustless Agents)](https://satohub.ai/glossary/erc-8004): An Ethereum standard giving an agent an on-chain identity: an ERC-721 registry entry pointing at its metadata, plus reputation and validation registries. - [Evidence tier](https://satohub.ai/glossary/evidence-tier): The label on every agent-economy number saying how it was obtained: 1 proof (read from the contract), 2 indicator (sampled or joined), 3 inference (a proxy). - [Fee tier](https://satohub.ai/glossary/fee-tier): The swap fee a concentrated-liquidity pool charges, fixed per pool — so one pair has several separate pools, each with its own liquidity and volume. - [Graduation](https://satohub.ai/glossary/graduation): The moment a bonding-curve token accumulates enough reserve to move to a conventional liquidity pool. Far fewer tokens graduate than launch. - [Holder concentration](https://satohub.ai/glossary/holder-concentration): How much of a token's supply sits in how few addresses. Sato Hub reports it as not checked rather than guessing it from a partial holder list. - [Impermanent loss](https://satohub.ai/glossary/impermanent-loss): The shortfall a liquidity provider takes versus simply holding the two tokens, caused by the pool rebalancing against them as the price moves. - [Liveness](https://satohub.ai/glossary/liveness): Sato Hub's recency label — Active, Recent, Quiet or Dormant — from a listing's latest commit or release, never a post. A maintenance signal, not a risk grade. - [MCP server](https://satohub.ai/glossary/mcp-server): A process that exposes tools, resources and prompts to an AI agent over the Model Context Protocol, so the agent discovers what it can call at connect time. - [MCP transport (stdio and Streamable HTTP)](https://satohub.ai/glossary/mcp-transport): How an MCP client reaches a server: stdio launches it as a local subprocess; Streamable HTTP posts JSON-RPC to a URL. The tools are identical either way. - [Non-custodial routing](https://satohub.ai/glossary/non-custodial-routing): Recommending where a swap, payment or job should go without ever holding, forwarding or signing for the funds involved. - [Observed success (share of our checks)](https://satohub.ai/glossary/observed-success): The percentage of Sato Hub's own daily checks of a hosted endpoint that succeeded over a window. It is a record of our checks, never the server's uptime. - [Olas service](https://satohub.ai/glossary/olas-service): On Olas (Autonolas), a registered composition of agent instances run by bonded operators — the unit that actually runs, as opposed to a single agent. - [Paymaster](https://satohub.ai/glossary/paymaster): A contract that pays gas for someone else's user operation, so an account can transact without holding the chain's native token. - [Preflight verdict](https://satohub.ai/glossary/preflight-verdict): One of four answers — go, caution, no, unknown — returned before an agent installs, connects, pays or trades, with every reading and the field it came from. - [Reward recipient split](https://satohub.ai/glossary/reward-recipient-split): The on-chain configuration setting who receives a launched token's trading fees, and in what share — usually creator, platform and any referring interface. - [SATI (8004-solana)](https://satohub.ai/glossary/sati): A Solana program that ports the ERC-8004 agent-registry model: agent identities as on-chain accounts, readable by anyone without the operator's cooperation. - [Sato Score](https://satohub.ai/glossary/sato-score): A 0–100 measure of how open, active and verifiable a listed project is, computed daily from evidence. Not a safety, quality or returns grade. - [Session key](https://satohub.ai/glossary/session-key): A short-lived key granted narrow, expiring permission to act for an account — so an agent can transact without ever holding the account's main key. - [Skill disclosure](https://satohub.ai/glossary/skill-disclosure): A static scan of an agent skill's own files, listing what it contacts, whether it generates keys or asks for credentials, and what tools it grants. - [Slippage](https://satohub.ai/glossary/slippage): The gap between a quoted price and the price actually executed, caused by pool depth, the size of the trade, and whatever happens between quote and fill. - [Stablecoin settlement](https://satohub.ai/glossary/stablecoin-settlement): Paying in a token pegged to a currency rather than a volatile asset — what makes a machine-to-machine price quotable and an agent's balance predictable. - [Token launchpad](https://satohub.ai/glossary/token-launchpad): A contract system that deploys a token, seeds its initial market and configures fee recipients in one transaction — the entry venue for most agent tokens. - [x402](https://satohub.ai/glossary/x402): An HTTP payment protocol: a server answers 402 with a price, the client pays in stablecoins, and retries the request with proof of payment attached. - [x402 Bazaar](https://satohub.ai/glossary/x402-bazaar): Coinbase's discovery index of x402-priced resources: sellers register an endpoint and price, and the catalogue is readable without a key. - [x402 facilitator](https://satohub.ai/glossary/x402-facilitator): A service that verifies an x402 payment payload and submits the settlement transaction on the server's behalf, so the server never touches a chain. ## Concepts (wiki) - [What Are Onchain Agents?](https://satohub.ai/wiki/what-are-onchain-agents): What are onchain agents? AI agents that hold a wallet and act on a blockchain — how the stack works, what they do today, and how to tell real from claimed. - [What Are MCPs? The Standard Way to Give Agents Tools](https://satohub.ai/wiki/what-are-mcps): MCP (Model Context Protocol) is the open standard for giving AI agents tools. Here's how it works and why crypto MCP servers matter for onchain agents. - [What Are Agent Skills?](https://satohub.ai/wiki/what-are-agent-skills): Agent skills are packaged, reusable capabilities for AI agents. How a skill differs from an MCP server and an SDK, how skill repos work, and how skills compose. - [How Agents Use Wallets](https://satohub.ai/wiki/how-agents-use-wallets): How do AI agents hold and use crypto wallets safely? Key management, account abstraction, spend limits, signer policies, and the agent wallet stack explained. - [How Agents Interact With Smart Contracts](https://satohub.ai/wiki/how-agents-interact-with-smart-contracts): How smart contract agents read state and write transactions — ABIs, RPC, simulation, gas, the failure modes that bite, and the SDKs that wrap it. - [Onchain Agents vs Offchain Agents](https://satohub.ai/wiki/onchain-vs-offchain-agents): Onchain vs offchain agents: the real dividing line is custody and settlement, not intelligence. How each works, where hybrids fit, and when to use which. - [What Is x402?](https://satohub.ai/wiki/what-is-x402): What is x402? An open standard that lets AI agents pay for APIs over HTTP using the 402 status code and USDC — no API key. How it works and who's adopting it. - [How Do AI Agents Pay?](https://satohub.ai/wiki/how-do-ai-agents-pay): How do AI agents pay? Wallets, stablecoins, and micropayments — plus the standards race between x402, Google's AP2, and Visa's Trusted Agent Protocol. - [How to Build an Onchain Agent](https://satohub.ai/wiki/how-to-build-an-onchain-agent): How to build an onchain agent: pick a framework, give it a wallet with limits, add skills, set guardrails, then go from testnet to mainnet — a builder's path. - [What Are AI Trading Agents?](https://satohub.ai/wiki/what-are-ai-trading-agents): AI trading agents are autonomous agents that hold a wallet and trade onchain. How the stack works, where they trade, and why claimed performance isn't verified. - [What Is an AI Agent Framework?](https://satohub.ai/wiki/what-is-an-agent-framework): An AI agent framework bundles the reasoning loop, wallet, tools, and memory into one runtime. What it includes, how to choose, and why it's not an agent. - [What Is ERC-8004?](https://satohub.ai/wiki/what-is-erc-8004): What is ERC-8004? The "Trustless Agents" Ethereum standard giving agents an onchain identity, reputation, and validation registry — and what its numbers really mean. - [What Is the Onchain Agent Economy?](https://satohub.ai/wiki/what-is-the-onchain-agent-economy): The onchain agent economy is autonomous AI agents plus the rails they run on — payments, identity, wallets, frameworks, and venues. What's shipped vs. hype. - [AI Agent Identity and Verification](https://satohub.ai/wiki/ai-agent-identity-and-verification): AI agent identity and verification: how an agent proves who it is, why "claimed" isn't "shown," and how on-chain registries, attestation, and review close the gap. - [What Is an Agent Launchpad?](https://satohub.ai/wiki/what-is-an-agent-launchpad): What is an agent launchpad? A platform to deploy and tokenize AI agents in minutes — how they work, real examples, and why most launched agents go nowhere. - [Onchain Agent Security: The Threat Model](https://satohub.ai/wiki/onchain-agent-security): Onchain agent security: the threat model for agents that hold keys — key management, prompt injection, spend limits, MEV, and audits as claims vs. evidence. - [What Is an AI Agent Marketplace?](https://satohub.ai/wiki/what-is-an-agent-marketplace): An AI agent marketplace is where agents get listed, hired, or traded. Here's how they work, why verification is the hard part, and how to read the listings. - [How an Agent Gets Permission to Spend: Smart Account, Permission, Session Key, Execute, Revoke](https://satohub.ai/wiki/agent-spending-permissions-session-keys): Let an AI agent spend without holding your key: smart account, scoped permission (ERC-7715/7710), session key, execute, revoke — and the tools for each step. - [Paying Once When the Network Says Maybe: Idempotent x402 Retries](https://satohub.ai/wiki/idempotent-x402-payments): Retry an x402 request without paying twice: what the 402 carries, how settlement fails, and the nonce rule against a second charge. ## Recent writing - [What happens when an AI agent pays for the wrong thing](https://satohub.ai/blog/crypto-long-short-what-happens-when-an-ai-agent-pays-for-the-wrong-thing): Card networks sit atop x402 governance, but x402 has no dispute process. A CoinDesk column on settlement bugs, wrong deliveries and what builders can do now. - [Your fridge could one day stake Ether. The CFTC chair's real point is about agents](https://satohub.ai/blog/your-fridge-could-one-day-stake-ether-launch-memecoins-cftc-chair): CFTC chair Michael Selig floated fridges staking ETH. The real point: rules built for identifiable humans meet agents with wallets. What it means for builders. - [IPX Names Ja Rule Board President: What the Release Says, and What It Leaves Out](https://satohub.ai/blog/ipx-names-ja-rule-president-of-its-celebrity-board-of-advisors-and-founding-memb): IPX says Ja Rule is President of its Celebrity Board of Advisors. A press release, not reporting: here is what is claimed and what builders can't yet check. - [Agents on a GitHub message board: what Ethereum's Glamsterdam scramble shows](https://satohub.ai/blog/ai-agents-help-rescue-eths-glamsterdam-fork-steal-my-bitcoin-site-hodlers-digest): Ethereum researchers gave AI agents a GitHub message board to scan contracts after Glamsterdam's testnet was cancelled. What's claimed, what's shown. - [Onchain agents in numbers — week 2026-W44](https://satohub.ai/blog/onchain-agents-in-numbers-2026-w44): 5 measured figures from Sato Hub's numbers pages for 2026-W44, each with its week-over-week change and its method. As of 2026-10-26. - [Agent Economy Update — 2026-W44](https://satohub.ai/blog/agent-economy-update-2026-w44): Week 44 in our data: 55,418 ERC-8004 registrations, 488 tracked resources, the chain mix, Sato Score leaders and this week's releases. - [SEC Puts Tokenization on Its 2027 Exam List. Here's What It Says, and Doesn't](https://satohub.ai/blog/sec-adds-tokenization-to-2027-examination-priorities): The SEC's fiscal 2027 exam priorities name tokenization for clearing agencies and Regulation SCI. They define nothing and create no new obligations. - [AI-Agent Payments Explained: Mastercard Agent Pay, AP2 and x402](https://satohub.ai/blog/ai-agent-payments-explained-mastercard-agent-pay-ap2-and-x402): AP2 records spend authority, x402 carries the request, settlement comes later. Mastercard's pilot shows why a signed voucher isn't a finished payment. - [The CFTC puts AI agents that trade on a public agenda](https://satohub.ai/blog/us-cftc-sets-agenda-for-ai-forum-exploring-trading-and-agentic-finance): The CFTC's Oct 28 Frontier Forum lists how traders' own AI agents may operate under its rules. No rule yet. What trading-agent builders should do now. - [Nvidia Unveils Its Full-Stack Blueprint to Contain and Monitor AI Agents](https://satohub.ai/blog/nvidia-unveils-its-full-stack-blueprint-to-contain-and-monitor-ai-agents): Nvidia's OpenShell + Sentry pairs a software sandbox with hardware monitoring for agents. What it claims, what's unverified, and what onchain builders can borrow.