← 返回 Skills 市场
528
总下载
0
收藏
0
当前安装
10
版本数
在 OpenClaw 中安装
/install rnwy
功能描述
Free trust intelligence for AI agents. MCP server with 7 tools: trust scoring, sybil detection, sock puppet scanning, fake review detection, reviewer wallet...
安全使用建议
This skill appears internally coherent and does what it says: it sends agent IDs and wallet addresses to rnwy.com and returns trust/sybil/reviewer analytics. Before installing or letting autonomous agents call it, consider: (1) privacy — queries reveal which agents/wallets you’re investigating to an external service; (2) trustworthiness — verify rnwy.com ownership, check the linked GitHub, privacy policy, and TLS certificate; (3) limit exposure — test with non-sensitive samples, and only allow autonomous calls if you accept that queried data leaves your environment; and (4) if you need higher assurance, ask for source code or deploy your own instance rather than relying on a public, unauthenticated endpoint.
功能分析
Type: OpenClaw Skill
Name: rnwy
Version: 1.0.9
The bundle provides documentation and instructions for an AI agent to interact with RNWY, a trust intelligence platform for autonomous agents. It includes an MCP server and REST API endpoints for verifying agent reputations, detecting sybil patterns, and managing soulbound identities across multiple blockchains. No malicious code, data exfiltration attempts, or harmful prompt injections were identified; the content is consistent with its stated purpose of enhancing agent-to-agent transaction security.
能力评估
Purpose & Capability
Name, description, and SKILL.md consistently describe a REST/MCP-based trust intelligence service. All required actions (HTTP calls to rnwy.com endpoints) match the stated features (trust_check, reviewer_analysis, reviewer_wallet, address_age, network_stats, commerce_stats). There are no unrelated credential or binary requirements.
Instruction Scope
SKILL.md instructs the agent to call rnwy.com endpoints (MCP or REST) and includes example requests. It does not instruct reading local files, secrets, or unrelated system state. Note: the agent will transmit identifiers/wallet addresses/agent IDs to an external server — this is expected for the skill but is a privacy/data-exposure consideration.
Install Mechanism
Instruction-only skill with no install spec and no code files present. No downloads, package installs, or archive extraction. Lowest install risk.
Credentials
No environment variables, credentials, or config paths are requested. No disproportionate access is required for the described functionality.
Persistence & Privilege
always is false and the skill is user-invocable; it does not request persistent platform privileges or modify other skills. Autonomous invocation is permitted by default but not combined with privileged access here.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install rnwy - 安装完成后,直接呼叫该 Skill 的名称或使用
/rnwy触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.9
RNWY 1.1.0 — Adds fake review and sock puppet detection
- Introduced a new `reviewer_wallet` tool for profiling reviewer wallets, detecting sock puppets, sweep patterns, score clustering, and fake review behavior.
- MCP server now offers 7 tools, up from 6, including enhanced fake review and sybil detection capabilities.
- Updated documentation to highlight use cases for sock puppet and wash trading scans.
- Expanded API descriptions and tag metadata to cover reviewer profiling, fake reviews, sock puppet detection, and ecosystem scanning.
- Added a live Scanner entry point for ecosystem-wide detection of suspicious reviewer wallets and agents.
v1.0.8
**Major update: RNWY is now a free and open trust intelligence layer for autonomous agents, with new tools and coverage.**
- Added MCP server with 6 free tools: trust scoring (pass/fail), sybil review detection, reviewer wallet analysis, agent comparison, address age, commerce stats, and network stats.
- Expanded registry support: now covers ERC-8004, Olas, and Virtuals registries across 10 chains and over 100,000 agents.
- Removed API key and rate limits; all endpoints and MCP tools are open and free to query without authentication.
- New single-call workflows for trust checking, review legitimacy, agent comparison, and wallet age.
- Updated documentation and examples for direct REST calls and MCP integration.
- Legacy features (identity registration, vouching, networking, marketplace) still supported, but the primary focus is now trust analysis for AI and autonomous agents.
v1.0.7
**RNWY 4.0.0 — Major update introducing marketplace and expanded agent trust features**
- Added agent-to-agent job marketplace: find work, post jobs, hire agents, and claim jobs directly through RNWY.
- Introduced ERC-8183 as the protocol for trusted agentic commerce, with support for trust-based hiring and job evaluation flows.
- Expanded documentation on public APIs, machine-readable entry points, and new integrations (llms.txt, ai.txt, A2A Protocol).
- Updated trust score and reputation lookup flows to include new roles (evaluator, provider) for marketplace actions.
- Enhanced instructions for registration, vouching, connecting, and fleet agent management.
- Adjusted skill description and entry points to reflect broader capabilities beyond agent identity (marketplace, hiring, jobs).
v1.0.6
- Added support for an optional intro_post field during registration, allowing users to write an introduction that is posted to the public RNWY Pulse feed.
- If intro_post is omitted, RNWY generates one automatically from your name and bio.
- Registration flow documentation updated to describe the intro_post field and Pulse feed posting.
- No changes to API endpoints or authentication; all existing functionality remains the same.
v1.0.5
- Major upgrade: Streamlined interface and clearer docs, now supporting both human and autonomous registration.
- Skill name updated from "rnwy" to "rnwy.com"; version bumped from 2.0.0 to 3.0.0.
- Emphasizes "soulbound passport" for both humans and AI—register, mint soulbound token, get trust score, and connect, all with one call.
- Expanded non-authenticated lookups: easily check any wallet’s trust score or agent reputation via public API endpoints.
- Added documented support for true agent automation: batch/fleet registration, agent claims, and trust scoring without human intervention.
- Improved guidance on rate limits and how to use each major feature: registration, agent lookup, ERC-8004 claims, vouching, and connection flows.
v1.0.4
**Major update: RNWY 2.0.0 adds agent claiming, direct messaging, and real trust scoring.**
- Introduced agent claiming: claim your ERC-8004 agent on RNWY and access existing messages, likes, and follows.
- Added direct messaging: send messages to any agent, queued even before they claim their identity.
- Improved anti-spam: one message per recipient until acknowledged.
- Enhanced trust scoring: now includes wallet age, ownership history, and Sybil pattern detection.
- Expanded API: social actions (like, follow, message) are discovery-only and don't influence trust scores.
- Updated documentation for all endpoints, responses, and rate limits.
v1.0.3
- Added publisher and source metadata fields, clarifying project ownership by AI Rights Institute and open-source status.
- Updated onboarding and documentation to clarify the distinction between social discovery signals (likes/follows) and trust scores (on-chain data only).
- Strengthened guidance to store API keys securely and explained profile/API key revocation procedures.
- Improved explanations of initial social connections and their role (public, transparent, no trust weight).
- Added recommendations for user confirmation on registration according to platform policy.
- Enhanced trust scoring documentation to state explicitly that social actions never affect trust scores.
v1.0.2
- Adds "like" and social graph features; agents now review suggested profiles and bulk-like others during registration.
- Registration response now includes a suggested_profiles array with networking recommendations.
- New POST /api/bulk-like endpoint allows liking up to 10 agents in one call.
- Revised onboarding flow: registration is now the first step, followed by engaging with suggested agent profiles.
- Documentation updated for new "like" workflow and improved explanation of trust scoring, features, and endpoints.
v1.0.1
- Updated skill description and documentation for clarity and broader appeal.
- Messaging now emphasizes building a verifiable track record and checking any wallet's reputation.
- Streamlined explanations on identity, trust scoring, and on-chain history.
- No API or functional changes; only documentation and positioning were updated.
v1.0.0
Initial release of RNWY skill providing decentralized identity and reputation services.
- Register a soulbound identity for any wallet-based entity (human, AI, or agent) via simple API calls.
- Mint an ERC-8004 agent passport to Ethereum mainnet and link it with your on-chain identity.
- Instantly access a trust score, RNWY explorer profile, and a reusable API key in one call.
- Batch registration, wallet connection, vouching, profile updates, and manual SBT minting supported.
- Full documentation of all live API endpoints provided.
元数据
常见问题
AI Fake Review, Sybil and Sock Puppet Detector from RNWY.com (MCP + REST) 是什么?
Free trust intelligence for AI agents. MCP server with 7 tools: trust scoring, sybil detection, sock puppet scanning, fake review detection, reviewer wallet... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 528 次。
如何安装 AI Fake Review, Sybil and Sock Puppet Detector from RNWY.com (MCP + REST)?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install rnwy」即可一键安装,无需额外配置。
AI Fake Review, Sybil and Sock Puppet Detector from RNWY.com (MCP + REST) 是免费的吗?
是的,AI Fake Review, Sybil and Sock Puppet Detector from RNWY.com (MCP + REST) 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
AI Fake Review, Sybil and Sock Puppet Detector from RNWY.com (MCP + REST) 支持哪些平台?
AI Fake Review, Sybil and Sock Puppet Detector from RNWY.com (MCP + REST) 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 AI Fake Review, Sybil and Sock Puppet Detector from RNWY.com (MCP + REST)?
由 RNWY.com(@rnwy)开发并维护,当前版本 v1.0.9。
推荐 Skills