llms.txt · git:20261020.a8df740 · 2026-10-20 · sha256 8d4fcb6a012edf9e
llms.txt git:20261020.a8df740A
Immutable. This exact content never changes and is served at /api/v1/blob/8d4fcb6a012edf9e.
# Onchain Agents — the scored, daily-updated index of the crypto agent stack > 486 listings of what onchain AI agents are built from: agent frameworks, > onchain action kits, MCP servers, wallets and key management, data and RPC, > x402 payment rails, identity standards (ERC-8004, A2A), security tooling and > trading venues. Rendered daily from satohub.ai. Last render: 2026-10-20. > Every product listing carries a Sato Score: 0-100, measuring how OPEN, ACTIVE > and VERIFIABLE a project is. It is NOT a safety, quality, security or returns > grade, and nothing here is financial advice. Self-reported is not verified. > Catalog data CC-BY-4.0, attribution: data by satohub.ai. ## Start here - [The stack, layer by layer](https://raw.githubusercontent.com/satohubai/onchain-agents/main/README.md): the answer to "what do I build an onchain agent with" — nine layers, each a table of the highest-scored listings with chains, score, last activity and what we checked. It is the first section of the README. - [Pages site](https://satohubai.github.io/onchain-agents/): the same, one page per layer - Starting stacks: five example layer combinations built only from listings in this index — examples, not endorsements (README, after the stack section) ## Layers (one page each) - [Agent runtimes & frameworks](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/categories/agent-frameworks.md): The loop that runs the agent: prompting, memory, tool calls, scheduling. Pick this first — it decides what everything else plugs into. - [Onchain action kits & SDKs](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/categories/action-kits.md): Libraries that turn an agent's decision into a signed transaction — swap, transfer, mint, stake — without writing the chain plumbing yourself. - [MCP servers](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/categories/mcp-servers.md): Tool servers your agent calls over the Model Context Protocol. This is the largest layer in the index and the fastest-moving. - [Wallets, keys & permissions](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/categories/wallets-keys.md): Where the agent's key lives and what it is allowed to do: embedded wallets, MPC, smart accounts, session keys and spend permissions. - [Data, RPC & indexing](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/categories/data-rpc.md): What the agent knows before it acts: RPC access, market data, chain indexing, wallet and token analytics. - [Payments & agent commerce](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/categories/payments.md): How agents pay and get paid: x402 (HTTP 402 plus stablecoin), agent commerce protocols, and the stablecoin rails underneath. - [Identity, discovery & standards](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/categories/identity-discovery.md): How an agent is identified and found by other agents: ERC-8004 registration, MCP discovery, A2A agent cards. - [Security & preflight](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/categories/security.md): Checks you run before an agent installs, connects, signs or trades: contract and token screening, endpoint checks, monitoring. - [Trading & DeFi venues](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/categories/trading-venues.md): Where the agent actually trades: DEXs, aggregators, perps, lending and yield venues with programmatic access. ## Per-listing pages - Every listing has a page at https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/listings/<slug>.md — facts, deploy spec, what we checked, links, citation - Slugs are the `slug` field in data/index.json ## Data (stable, versioned) - [index.json](https://raw.githubusercontent.com/satohubai/onchain-agents/main/data/index.json): full index, schema_version field, one object per resource - [index.csv](https://raw.githubusercontent.com/satohubai/onchain-agents/main/data/index.csv): flat CSV of the same - [Live export feed](https://satohub.ai/api/export/index.json): the upstream source; also .csv and .ndjson, with ?category= ?chain= ?standard= ?use_case= slices - [All datasets](https://satohub.ai/datasets): every published dataset and its license - On Hugging Face: [Onchain Agent Builder Bench](https://huggingface.co/datasets/SatoHub/onchain-agent-builder-bench) (eval) · [Sato MCP Tool-Calling](https://huggingface.co/datasets/SatoHub/sato-mcp-tool-calling) (training data) - [Schemas](https://satohub.ai/schemas): versioned JSON Schemas for these formats - [OpenAPI](https://satohub.ai/api/openapi.json): every public operation ## Query it live over MCP - [MCP endpoint](https://satohub.ai/api/mcp): 46 tools over Streamable HTTP (42 read, 4 write) — write tools: onchain_agent_watch, onchain_agent_submit_project, onchain_agent_register_agent, onchain_agent_swap - [Tool list as JSON](https://satohub.ai/api/mcp/tools.json): names, descriptions and input schemas - [Connection guide](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/connect-mcp.md) - Skill: `npx skills add satohubai/sato-hub-skill` · Plugin: `/plugin marketplace add satohubai/sato-plugins` ## Docs - [Sato Score methodology](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/sato-score.md): what the 0-100 score measures and what it does not - [Taxonomy](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/taxonomy.md): entity classes, resource types, interfaces, standards facets - [Swap through a gate](https://raw.githubusercontent.com/satohubai/onchain-agents/main/docs/swap.md): how an agent gets a swap it is allowed to make — four dated verdicts, the owner's policy, a simulation, an unsigned transaction - [Neutrality](https://raw.githubusercontent.com/satohubai/onchain-agents/main/NEUTRALITY.md): no project can pay to be listed, ranked or scored - [Reporting a problem with an entry](https://raw.githubusercontent.com/satohubai/onchain-agents/main/SECURITY.md) - [Cite this dataset](https://raw.githubusercontent.com/satohubai/onchain-agents/main/CITATION.cff) · [dataset record](https://satohub.ai/datasets/onchain-agents-index) - [Status of these surfaces](https://satohub.ai/status/sato-hub) ## Related - [Agent Passports](https://satohub.ai/agents): registered agents with on-chain verification checks and machine-readable manifests - [Agent Architect / Builder](https://satohub.ai/build): describe an agent in plain language, get a build plan made of listings from this index - [SATO OS](https://raw.githubusercontent.com/satohubai/onchain-agents/main/SATO-OS.md): self-hosted mission control for running an onchain agent - [satohub.ai](https://satohub.ai): the full product