Find crypto-agent tooling by name, chain, standard or what it does
onchain_agent_search_resourcesUSE WHEN you need to look up crypto-agent tooling (frameworks, MCP servers, wallets, payment rails, data tools) by name, chain, standard or what it does, or check that a specific project exists and is maintained. A daily-rebuilt directory, each entry scored from public evidence and linking to a citable page.
NOT FOR a build goal in plain words such as "a trading agent on Base" (use onchain_agent_recommend_stack instead), a RUNNING agent to hire or pay (use onchain_agent_search_agents), a downloadable agent package (use onchain_agent_search_listings), what an agent skill touches (use onchain_agent_search_skills), or one record by slug (use onchain_agent_get_resource). search is the one-box version for ChatGPT deep research.
RETURNS (json): { total, count, offset, has_more, next_offset?, resources: [...] }; each resource has chains, status, liveness, github_stars, verification_status and the marketplace fields (is_agent/is_hirable/is_licensable). Read-only. With NO argument it is a compact overview (rule "defaulted": complete rows, fewer, plus directory counts); detail: "full" is the full first page.
FILTERS: query (free text, AND-matched terms), chain, status, liveness, featured, and the taxonomy facets (docs/taxonomy.md) entity_class, resource_type, use_case, standard and iface — closed vocabularies listed in the input schema: anything else is refused with the list, never answered with an empty result. Legacy flags still work: category, is_agent, is_skill, is_harness. Sort by priority (default), newest_release, stars, or name. Paginates via limit/offset. Deprecated resources are never returned.
STARTING POINTS: a query asking for somewhere to start — "tutorial", "starter", "beginner", "template", "example", "docs" — returns starting points labelled as such: listings named as a starter or template first, then listings whose install Sato Hub reproduced, plus the wiki guides. "starter" / "template" / "x402 starter" also carry templates: a pointer to onchain_agent_create_agent and the templates catalogue (never a result), unless the query names another chain or runtime. Only when the catalogue has no template for Base is it starter instead: a TypeScript template built by Sato Hub (labelled ours, not audited, never a result). The directory lists tools, not tutorials, and says so.
Examples:
"Active hirable agents on Base" -> { chain: "Base", is_agent: true, liveness: "Active" }
"newest releases" -> { sort: "newest_release", limit: 10 }
"wallet tooling" -> { query: "wallet" }
"MCP servers I can use from Claude Code" -> { integration: "claude code", iface: "mcp" }
"things I can run in Docker" -> { deploys_as: "docker" }
"MCP servers that actually answer" -> { mcp_answering: true } (the MCP endpoint answered our last handshake)
"...with a sustained record" -> { mcp_answering: true, min_observed_success: 95 }
"only installs Sato Hub has reproduced" -> { verified_only: true }
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| sort | No | Sort order: priority (default), newest_release, stars, or name. | priority |
| chain | No | Filter to resources supporting this chain. | |
| iface | No | Filter by how it is accessed/integrated. One of: mcp, sdk, rest-api, plugin, cli, ui, contract. | |
| limit | No | Max results to return (1-50, default 20). | |
| query | No | Free-text search across name, description, tags, chains, and agent type. Space-separated terms are AND-matched. | |
| detail | No | 'full' returns the full first page (20 rows by default) even with no filter. Omitted with no other argument, the answer is a compact overview: the first rows in the default order, what the directory holds by category and chain, and the exact calls for the rest. Any query, filter, sort, limit or offset answers as before. | |
| offset | No | Results to skip, for pagination (default 0). | |
| status | No | Filter by lifecycle status. | |
| creator | No | Filter by who built it (prefix match on creator name), e.g. 'Coinbase', 'Privy'. | |
| category | No | Filter to a resource category. | |
| featured | No | True = only editorially featured resources. | |
| is_agent | No | True = only resources that are themselves onchain agents. | |
| is_skill | No | True = only agent-skill resources (skill repos/marketplaces). | |
| liveness | No | Code activity (commit or release, not posts): Active ≤30d, Recent ≤90d, Quiet ≤1y, Dormant >1y. | |
| standard | No | Filter to resources implementing a standard. One of: x402, erc-8004, erc-8183, mcp, a2a. | |
| use_case | No | Filter by use case. One of: trading, payments, wallets, data, identity, privacy, launch, security, build, airdrops. | |
| deploys_as | No | Filter by deployment shape: npm, pip, docker, mcp server, hosted, self-hosted, cli, sdk, claude code plugin. | |
| is_harness | No | True = only agent frameworks/harnesses (OpenClaw, Codex, Claude Code…). | |
| integration | No | Filter to resources that integrate with a client or framework, e.g. 'claude code', 'cursor', 'langchain', 'openclaw'. Case-insensitive exact match. | |
| entity_class | No | Filter by class: 'resource' = things you build/deploy WITH (frameworks, tools, infra, venues, standards); 'agent' = curated, live onchain agents; 'reference' = editorial. | |
| mcp_answering | No | True = only listings whose MCP ENDPOINT answered Sato Hub's most recent MCP handshake (one check of ours, not uptime). Use this for 'which MCP servers answer'; min_observed_success alone is mostly a check of a listing's website. | |
| resource_type | No | Filter resources by role. Venue = DEX/launchpad/marketplace; Network = an L1/L2. | |
| verified_only | No | True = only listings whose documented install was reproduced in a container by Sato Hub (deploy_status verified). Says nothing about runtime safety. | |
| response_format | No | Text format; structuredContent is JSON either way. | markdown |
| min_observed_days | No | Minimum days of daily observation behind the success rate, counted in the rate window (default 14 when min_observed_success is set). | |
| min_observed_success | No | Only listings whose share of Sato Hub's daily checks that succeeded is at least this percent, over at least min_observed_days (default 14). The rate and the days both count only checks in the rate window: from 2026-09-01 (when the probe stopped counting a bot-block as a failure) if an earlier check failed, else from the first check. This is the share of OUR checks, not uptime. |