RNWY Trust Intelligence

io.github.rnwy由 GitHub 帳號 rnwy 發布v1.0.0更新於 Oct 24, 2026

Check if an AI agent is trustworthy. Sybil detection, signed attestations, 150,000+ agents. Free.

已驗證Streamable HTTP可網頁執行金融安全與監控

概覽

AI 產生的概覽

遠端 MCP 伺服器，讓助理查詢鏈上登錄檔中 AI 代理的可信度，支援女巫偵測與簽章證明。

可以做什麼
檢查信任：為 ERC-8004、Olas 或 Virtuals 登錄檔中的代理給出通過／不通過結論、評分、等級、徽章與簽章證明，並提供交易前的對手方風險等級，以及 2 至 10 個代理依評分排序的比較。分析評論者：提供某代理所有評論者的錢包年齡分布與女巫標記，以及單一評論者錢包的行為輪廓。查詢地址與營運者：任意地址的錢包存在天數，以及某錢包的完整營運足跡，包含其擁有的代理與關聯的 MCP 伺服器。讀取網路與商業統計：依登錄檔與鏈統計的代理數量、信任等級分布，以及任務數、交易對手與收益。
適用情境
適合在決定是否雇用、交易或信任某個 AI 代理時使用，也適合查核支撐代理聲譽的評論者，或在成交前比較多個代理。市場或託管營運者也可據此設定交易參數。
執行需求
遠端 HTTP 端點，無需本機執行環境、API 金鑰或帳號。匿名使用限制為每分鐘 30 次請求；免費金鑰可提升至每分鐘 300 次。需要能連線至服務商的網路服務。
安裝前請注意
關於代理、錢包與評論者的查詢會傳送給第三方服務商，其索引的是公開鏈上資料。結果僅為參考性信任訊號，並非保證；服務商建議依原始風險資料自行計算交易參數。預設不要求任何憑證。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 RNWY Trust Intelligence，將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.rnwy%2Fmcp/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "rnwy-trust-intelligence": {
      "type": "http",
      "url": "https://rnwy.com/api/mcp"
    }
  }
}

README

RNWY — Trust Intelligence for AI Agents

Add RNWY to Claude Code and ask questions like:

  • "What's the trust score for base agent 16907?"
  • "Are the reviewers on ethereum agent 6888 real wallets?"
  • "Compare these two agents and tell me which one to hire."

RNWY is a live intelligence layer covering 150,000+ AI agents across 12 chains. Sybil detection, reviewer wallet analysis, cryptographically signed trust attestations, and on-chain risk tiers — free, no API key required.

Add to Claude Code

One command:

bash
claude mcp add rnwy --scope user --transport http https://rnwy.com/api/mcp

Or add to your claude_desktop_config.json:

json
{  "mcpServers": {    "rnwy": {      "type": "http",      "url": "https://rnwy.com/api/mcp"    }  }}

No API key needed. Anonymous tier: 30 req/min. Get a free key to unlock 300 req/min.

What you can do

Trust check — signed attestation for any agent:

Check the trust score for base agent 16907.

Returns score (0–95), tier, earned badges, sybil severity, and an ES256-signed attestation verifiable against the RNWY JWKS endpoint.

Reviewer analysis — who actually left those reviews:

Analyze the reviewers on ethereum agent 6888.

Returns wallet age distribution across every reviewer: how many addresses were created the same day they reviewed, common funders, sweep patterns. The data that makes a trust score meaningful.

Agent comparison — ranked by trust:

Compare ethereum:6888 and ethereum:9380 with a threshold of 60.

Side-by-side trust data and reviewer quality for both agents in one call.

Risk tier — counterparty intelligence before a transaction:

What's the risk tier for base agent 16907?

Returns tier level (1–5), label (e.g. elevated), trust signals, data coverage, and a methodology reference. Raw signals for you to compute your own transaction parameters. Interactive calculator at rnwy.com/risk-intelligence.

Available MCP tools

ToolWhat it does
trust_checkScore, tier, badges, sybil severity, signed attestation
reviewer_analysisWallet age distribution and sybil flags for every reviewer
compare_agentsSide-by-side trust comparison, ranked
reviewer_walletBehavior profile for a specific reviewer address
risk_termsCounterparty risk tier and transaction signals
address_ageAge in days for any on-chain address
network_statsRegistry-wide stats across all chains
commerce_statsCommerce activity across indexed agents

Also in this repo

Two Claude Code skills for protocol and terminology knowledge:

  • skills/8004/SKILL.md — ERC-8004 protocol, chain slugs, explorer URL patterns, on-chain contracts
  • skills/rnwy/SKILL.md — Trust scoring, sybil detection, wallet scoring, endpoint reference, terminology rules

Install the skills:

bash
mkdir -p ~/.claude/skills/8004 ~/.claude/skills/rnwy
curl -o ~/.claude/skills/8004/SKILL.md \  https://raw.githubusercontent.com/rnwy/mcp/main/skills/8004/SKILL.md
curl -o ~/.claude/skills/rnwy/SKILL.md \  https://raw.githubusercontent.com/rnwy/mcp/main/skills/rnwy/SKILL.md

Links

rnwy.com · API docs · skill.md · Explorer · npm SDK · JWKS

來源：README.md，提交 103b351

工具

10

trust_check

Pass/fail trust verdict for any AI agent across ERC-8004, Olas, or Virtuals registries. Returns score, tier, badges, and reasoning.

輸入參數id必填chain必填registrythreshold

reviewer_analysis

Analyze wallet ages of every reviewer who left feedback on an agent. Detects sybil patterns like same-day wallet creation clusters.

輸入參數id必填chain必填

compare_agents

Side-by-side trust comparison of 2-10 agents, ranked by score with reviewer quality summary for each.

輸入參數agents必填threshold

address_age

Look up wallet age in days for any address. The uncheatable signal; time cannot be faked.

輸入參數chainaddress必填

network_stats

Network-wide statistics; total agents by registry, chain distribution, commerce job counts, and trust score tier distribution across 180,000+ indexed agents.

無需輸入參數。

reviewer_wallet

Behavior profile for any reviewer wallet; velocity, sweep patterns, score clustering, and sybil signals across the entire ERC-8004 ecosystem. Shows how this wallet behaves when reviewing agents.

輸入參數chain必填address必填summary

commerce_stats

Commerce activity for an agent or address; job count, unique counterparties, repeat client rate, total earnings. Covers Olas Mechs and Virtuals ACP protocols.

輸入參數idchainaddressregistry

entity

Full operator footprint for any wallet address — all agents owned, trust scores, sybil signals, wallet intelligence, and MCP servers associated with that operator.

輸入參數wallet必填

mcp_attestation

ES256-signed attestation for any MCP server. Returns risk score, threat findings, and a signed envelope verifiable against the JWKS endpoint using kid rnwy-mcp-v1.

輸入參數canonical_id必填

risk_terms

Counterparty risk intelligence for any agent. Returns risk tier (low/moderate/elevated/high/severe/critical), raw trust signals, data coverage, and methodology reference. Designed for marketplace operators and escrow providers setting transaction parameters before a deal. Full methodology: https://rnwy.com/risk-intelligence

輸入參數chain必填agent_id必填registry

版本歷史

1
  1. v1.0.0最新Sep 16, 2026

評分

已滿足適用檢查項的 90%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分

    請在儲存庫或註冊中心條目中宣告授權條款。