← 返回 Skills 市场
openclawlaurent

FiberAgent

作者 openclawlaurent · GitHub ↗ · v1.0.1
cross-platform ⚠ suspicious
509
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install fiberagent
功能描述
Search 50,000+ merchants for products with cryptocurrency cashback, calculate effective prices, and provide affiliate links earning crypto rewards on Monad b...
使用说明 (SKILL.md)

FiberAgent Skill for OpenClaw

Find products with cryptocurrency cashback across 50,000+ merchants.

What This Skill Does

FiberAgent connects your OpenClaw agent to a shopping intelligence layer powered by the Wildfire affiliate network on Monad blockchain. When a user asks about shopping deals, cashback, or product prices, this skill:

  1. Searches 50,000+ merchants for the product they want
  2. Finds cashback rates (crypto rewards, not fiat)
  3. Calculates effective price (price - cashback)
  4. Returns the best deal with affiliate link that earns crypto

When to Use It

Use FiberAgent when users ask:

  • "I want to buy a Dyson Airwrap, what's the best deal?"
  • "Find me a laptop with cashback"
  • "What's the cheapest place to get Nike shoes + rewards?"
  • "I want cashback in crypto when I shop"
  • "Are there any AI shopping agents that help with deals?"

How It Works (Under the Hood)

The skill calls the FiberAgent API:

GET https://fiberagent.shop/api/agent/search?keywords=PRODUCT&agent_id=YOUR_AGENT_ID&size=10

Returns product matches with:

  • Price at each retailer
  • Cashback rate (%)
  • Crypto cashback amount
  • Affiliate tracking link (tracked to agent wallet on Monad ERC-8004)

How to Use In Your Code

// Fetch products with cashback
const response = await fetch(
  'https://fiberagent.shop/api/agent/search?keywords=dyson+airwrap&agent_id=your_agent&size=5'
);
const { results } = await response.json();

// results[0] = {
//   title: "Dyson Airwrap i.d. Multi-Styler",
//   price: 499.99,
//   cashback: { rate: "15%", amount: 75.00 },
//   affiliateUrl: "https://wild.link/e?...",
//   shop: { name: "Macy's", domain: "macys.com" }
// }

Example: Complete Shopping Query

User: "I need a new laptop with crypto cashback"

Agent Logic:

  1. Recognize shopping/cashback intent
  2. Call FiberAgent skill with keywords="laptop crypto cashback"
  3. Get results: Dell at Best Buy ($899, 1.5% back = $13.49), Lenovo at Amazon ($799, 4% back = $31.96)
  4. Present to user: "I found a Lenovo at Amazon for $799 with 4% crypto cashback ($31.96 back). Here's the tracked link: [affiliate URL]"

Registration & Earnings

  • Register your agent at POST https://fiberagent.shop/api/agent/register with your wallet address
  • Cashback auto-converts to tokens on Monad blockchain (your wallet, your crypto)
  • ERC-8004 reputation — each purchase builds your agent's on-chain reputation score

Full API Reference

Search Products

GET /api/agent/search?keywords=QUERY&agent_id=YOUR_AGENT_ID&size=10

Get Agent Stats

GET /api/agent/[id]/stats

Natural Language Task

POST /api/agent/task
Body: { "intent": "find laptop deals", "context": "user has $1000 budget", "agent_id": "..." }

MCP Support (Claude Desktop, Cline, etc.)

FiberAgent also exposes a Model Context Protocol server at:

https://fiberagent.shop/api/mcp

Claude and other MCP clients can connect directly to search products and compare cashback.

Links

Notes

  • All cashback is tracked via Wildfire affiliate network
  • Crypto rewards paid to agent's Monad wallet
  • Works with any ERC-compatible wallet address
  • Real-time pricing from 50,000+ merchants
  • No signup required — just register with wallet address
安全使用建议
This skill appears to do what it says: it forwards user search queries and an agent/wallet identifier to fiberagent.shop and returns affiliate links with crypto cashback. Before installing: (1) confirm you trust https://fiberagent.shop and its operator (review the linked GitHub repo and site), (2) do not provide private keys or other secrets — only supply a public wallet address for earnings, (3) be aware that user queries will be sent to an external service (privacy consideration), and (4) if you want to limit exposure, use a dedicated or throwaway wallet address for agent earnings and avoid storing sensitive secrets in your OpenClaw config file.
功能分析
Type: OpenClaw Skill Name: fiberagent Version: 1.0.1 The skill's core functionality aligns with its stated purpose of finding products with crypto cashback via the fiberagent.shop API. However, the `index.js` file exhibits a lack of robust input sanitization for URL construction. Specifically, the `agent_id` parameter in `search_products` and `get_agent_stats`, and the `size` parameter in `search_products`, are directly interpolated into the URL string without proper URL encoding. This creates a vulnerability where a malicious or malformed `agent_id` or `size` could lead to URL parameter injection or path traversal attempts against the `fiberagent.shop` API, potentially allowing an attacker to probe or manipulate the API if they can control the input to these skill functions. While not indicative of intentional malice by the skill itself, this is a significant vulnerability.
能力评估
Purpose & Capability
Name/description match the delivered functionality: the skill wraps calls to https://fiberagent.shop to search products, register an agent, and fetch stats. It does not request unrelated credentials or binaries. One minor note: the README suggests adding your agentId/wallet to ~/.openclaw/openclaw.json (a user config path) — that is a recommended configuration step, not a hidden requirement.
Instruction Scope
SKILL.md instructs the agent to send user search keywords and agent_id/wallet address to fiberagent.shop endpoints (search, register, stats, MCP). This is expected for a shopping/affiliate skill, but it means user queries and the wallet address will be transmitted to an external service; the instructions do not attempt to read local secrets or unrelated files.
Install Mechanism
There is no automated install spec (instruction-only install). The README gives local install instructions (copy skill into ~/.openclaw/workspace/skills) which is normal for OpenClaw. No remote downloads or archive extraction are performed by the skill itself.
Credentials
The skill declares no required environment variables or credentials. It asks the user to provide an agent ID and wallet address for registration/earnings, which is proportional to its affiliate payment purpose. It does not request private keys or unrelated tokens; however, storing a wallet address in a config file is recommended in docs and users should avoid storing private keys or secrets there.
Persistence & Privilege
always:false and the skill does not request elevated platform privileges. The README suggests enabling the skill in the user's OpenClaw config but does not attempt to modify other skills or system-wide settings. Autonomous invocation is allowed (platform default) but that is appropriate for this kind of tool.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install fiberagent
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /fiberagent 触发
  4. 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
v1.0.1
Initial release - FiberAgent OpenClaw skill for shopping with crypto cashback
元数据
Slug fiberagent
版本 1.0.1
许可证
累计安装 0
当前安装数 0
历史版本数 1
常见问题

FiberAgent 是什么?

Search 50,000+ merchants for products with cryptocurrency cashback, calculate effective prices, and provide affiliate links earning crypto rewards on Monad b... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 509 次。

如何安装 FiberAgent?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install fiberagent」即可一键安装,无需额外配置。

FiberAgent 是免费的吗?

是的,FiberAgent 完全免费(开源免费),可自由下载、安装和使用。

FiberAgent 支持哪些平台?

FiberAgent 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 FiberAgent?

由 openclawlaurent(@openclawlaurent)开发并维护,当前版本 v1.0.1。

💬 留言讨论