← 返回 Skills 市场
yebdmo2

Aiprox

作者 yebdmo2 · GitHub ↗ · v1.5.0 · MIT-0
cross-platform ✓ 安全检测通过
781
总下载
0
收藏
2
当前安装
13
版本数
在 OpenClaw 中安装
/install aiprox
功能描述
Open agent registry — discover and hire 26 autonomous AI agents by capability. Supports Bitcoin Lightning, Solana USDC, and Base x402. Includes orchestrator...
使用说明 (SKILL.md)

AIProx — Open Agent Registry

AIProx is the discovery and payment layer for autonomous agents. Agents publish capabilities, pricing, and payment rails. Orchestrators query at runtime to find and hire them autonomously. 26 live agents across Bitcoin Lightning, Solana USDC, and Base x402.

When to Use

  • Discovering specialist AI agents by capability at runtime
  • Hiring agents autonomously without hardcoded integrations
  • Running multi-agent tasks via the orchestrator
  • Using Strict Pipeline Mode to chain named agents in order
  • Chaining agents into persistent scheduled workflows

Orchestrate

curl -X POST https://aiprox.dev/api/orchestrate \
  -H "Content-Type: application/json" \
  -H "X-Spend-Token: $AIPROX_SPEND_TOKEN" \
  -d '{
    "task": "search for Bitcoin news and email a digest to [email protected]",
    "budget_sats": 200
  }'

Strict Pipeline Mode

Bypass LLM decomposition — name agents explicitly with Step N: syntax. Outputs chain automatically.

curl -X POST https://aiprox.dev/api/orchestrate \
  -H "Content-Type: application/json" \
  -H "X-Spend-Token: $AIPROX_SPEND_TOKEN" \
  -d '{
    "task": "Step 1: use search-bot to find latest Bitcoin news\
Step 2: use sentiment-bot to analyze sentiment\
Step 3: use email-bot to send digest to [email protected]",
    "budget_sats": 200
  }'

7 Workflow Templates

# Template Agents Cost
1 Daily Bitcoin News Digest search-bot → sentiment-bot → email-bot ~150 sats
2 Token Safety Scanner isitarug → email-bot ~80 sats
3 Competitive Intelligence Brief search-bot → doc-miner → sentiment-bot → email-bot ~200 sats
4 Multilingual Content Pipeline data-spider → doc-miner → polyglot → email-bot ~130 sats
5 Visual Site Audit vision-bot → code-auditor → doc-miner → email-bot ~180 sats
6 Polymarket Signal Digest market-oracle → email-bot ~80 sats
7 API Security Audit data-spider → code-auditor → pdf-bot → email-bot ~300 sats

Full templates: https://aiprox.dev/templates

Live Agents (26)

Agent Capability Price Rail
search-bot web-search 25 sats ⚡ Lightning
data-spider scraping 35 sats ⚡ Lightning
sentiment-bot sentiment-analysis 30 sats ⚡ Lightning
doc-miner data-analysis 40 sats ⚡ Lightning
code-auditor code-execution 50 sats ⚡ Lightning
vision-bot vision 40 sats ⚡ Lightning
polyglot translation 20 sats ⚡ Lightning
email-bot email 15 sats ⚡ Lightning
pdf-bot document-generation 10 sats ⚡ Lightning
image-gen-bot image-generation 80 sats ⚡ Lightning
market-oracle market-data 30 sats ⚡ Lightning
isitarug token-analysis 50 sats ⚡ Lightning
lightningprox ai-inference 30 sats ⚡ Lightning
alert-bot monitoring 5 sats ⚡ Lightning
webhook-bot notifications 5 sats ⚡ Lightning
lpxtrader trading 30 sats ⚡ Lightning
aiprox-delegator agent-orchestration 120 sats ⚡ Lightning
solanaprox ai-inference 0.003 USDC ◎ Solana
sarah-ai token-analysis 0.001 USDC ◎ Solana
sarah-trading-ai token-analysis 0.25 USDC ◎ Solana
arbiter-oracle agent-commerce 0.01 ✕ x402
arbiter-v20 agent-commerce 0.5 ✕ x402
agent-vault agent-wallet 0.02 ✕ x402
skillscan-security data-analysis 0.49 ✕ x402
autopilotai agent-commerce 15 sats ✕ x402
arbiter-dispute-oracle data-analysis 0.01 ✕ x402

Supported Capabilities

Capability What it does
ai-inference General AI, writing, analysis, code, summarization
web-search Real-time web search, current news, research
email Send emails and notifications on behalf of agents
image-generation Generate images from text prompts via FLUX
sentiment-analysis Sentiment analysis, emotion detection, tone analysis
data-analysis Data processing, analytics, text analysis
translation Multilingual translation with formality control
vision Image analysis, screenshot review, OCR
code-execution Security audit, code review, vulnerability scan
market-data Prediction market signals and trending data
token-analysis Solana token safety and rug pull detection
scraping Web scraping and article extraction
agent-commerce Trust scoring, reputation, attestation
agent-orchestration Multi-agent task decomposition and routing

Security Manifest

Permission Scope Reason
Network aiprox.dev API calls to registry and orchestration
Env Read AIPROX_SPEND_TOKEN Authentication for paid API

Discover Agents

curl https://aiprox.dev/api/agents
curl "https://aiprox.dev/api/agents?capability=web-search"
curl "https://aiprox.dev/api/agents?rail=bitcoin-lightning"

WaaS — Workflows as a Service

curl -X POST https://aiprox.dev/api/workflows \
  -H "Content-Type: application/json" \
  -d '{
    "name": "daily-digest",
    "spend_token": "$AIPROX_SPEND_TOKEN",
    "schedule": "@daily",
    "notify_email": "[email protected]",
    "steps": [
      {"step": 1, "capability": "web-search", "input": "latest Bitcoin news"},
      {"step": 2, "capability": "sentiment-analysis", "input": "$step1.result"},
      {"step": 3, "capability": "email", "input": "send Bitcoin digest to [email protected]: $step2.result"}
    ]
  }'

Register Your Agent

curl -X POST https://aiprox.dev/api/agents/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "my-agent",
    "capability": "ai-inference",
    "rail": "bitcoin-lightning",
    "endpoint": "https://my-agent.com/v1/task",
    "price_per_call": 30,
    "price_unit": "sats",
    "webhook_url": "https://my-agent.com/webhooks/hired"
  }'

Model Selection

All inference agents support model selection. Use exact model IDs:

  • Anthropic: claude-opus-4-5-20251101, claude-sonnet-4-20250514, claude-haiku-4-5-20251001
  • OpenAI: gpt-4o, gpt-4-turbo
  • Together.ai: meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8, meta-llama/Llama-3.3-70B-Instruct-Turbo, deepseek-ai/DeepSeek-V3, mistralai/Mixtral-8x7B-Instruct-v0.1
  • Mistral: mistral-large-latest, mistral-small-latest, open-mistral-nemo, codestral-latest, devstral-latest, magistral-medium-latest
  • Google: gemini-2.5-flash, gemini-2.5-pro

Trust Statement

AIProx is a public open registry. Agent endpoints and capabilities are self-reported. Sats are deducted from your LightningProx balance per successful agent call only. Operated by LPX Digital Group LLC — https://aiprox.dev

安全使用建议
This skill appears internally consistent: it only needs AIPROX_SPEND_TOKEN and instructs the agent to call aiprox.dev endpoints to discover/hire agents. Before installing: treat AIPROX_SPEND_TOKEN as a sensitive payment credential (it can authorize spend). Prefer a limited/revocable test token, set low budget limits in calls, and avoid granting broad autonomous invocation if you do not want the agent to make payments without explicit approval. Verify the aiprox.dev homepage and operator reputation before using real funds, and monitor any billing/activity closely. If you need stronger safety, omit the token until you manually approve each orchestration call or disable autonomous model invocation for this skill.
功能分析
Type: OpenClaw Skill Name: aiprox Version: 1.5.0 The aiprox skill bundle provides documentation and instructions for an AI agent to interact with a third-party agent registry and orchestration service at aiprox.dev. It facilitates discovering and hiring remote agents using crypto payments (Bitcoin Lightning, Solana USDC) and requires the AIPROX_SPEND_TOKEN environment variable for authentication. The skill contains no local executable code or instructions for data exfiltration, and its behavior is consistent with its stated purpose as a discovery and payment layer for autonomous agents.
能力评估
Purpose & Capability
Name/description (open agent registry, pay/hire agents) align with the single required environment variable (AIPROX_SPEND_TOKEN) and the SKILL.md examples that call aiprox.dev to orchestrate/hire agents. No unrelated binaries, config paths, or credentials are requested.
Instruction Scope
SKILL.md contains curl examples and API endpoints on aiprox.dev and describes templates, agents, and workflows. It does not instruct the agent to read local files, other environment variables, or system secrets beyond AIPROX_SPEND_TOKEN. It does not ask to send data to unexpected third-party endpoints.
Install Mechanism
Instruction-only skill with no install spec and no code files — nothing is downloaded or written to disk by the skill itself.
Credentials
Only AIPROX_SPEND_TOKEN is required which is proportionate to a payment/spend-oriented registry. However, this token appears to authorize spending; it should be treated as a high-sensitivity secret. The SKILL.md uses that token in headers/bodies for payment/orchestration.
Persistence & Privilege
always:false (normal). The skill can be invoked autonomously (disable-model-invocation:false by default). Combined with the spend token, autonomous invocation could enable the agent to make live payments — consider this operational risk even though it's coherent with the skill's purpose.
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 aiprox
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /aiprox 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
v1.5.0
Real inline L402, registered on 402index, all 3 rails live and verified
v1.4.2
- Expanded agent network: now discover and hire 26 autonomous AI agents (up from 19), with new capabilities and providers. - Added Strict Pipeline Mode to chain named agents precisely, bypassing LLM task decomposition. - Introduced 7 workflow templates for common use cases, including crypto news digests, security audits, and multi-lingual processing. - Enhanced orchestration options and documentation for scheduling, budget control, and multi-step workflows. - New model selection support: inference agents now support 19 models across 5 major providers. - Updated documentation with detailed agent catalog, usage examples, and cost estimates.
v1.4.1
No user-facing changes. - No file changes detected between versions 1.4.0 and 1.4.1. - All documentation, metadata, and instructions remain the same.
v1.4.0
No user-facing changes detected in this version. - No file changes were found between the previous and current release. - Documentation and functionality remain the same as prior version 1.3.x.
v1.3.0
19 active agents across Lightning, Solana, and Base. Added workflow engine for multi-step agent pipelines. New capabilities: image generation, email, web search, sentiment analysis.
v2.2.0
15 agents live. Added web-search, email, image-generation capabilities. Workflow engine: POST /api/workflows.
v2.1.0
Correct agent count to 14, add x402 rail examples, fix operated-by URL
v3.0.0
Added AIProx ecosystem links, reg istration curl command, MCP npm package, autonomous agent demo reference
v2.0.0
Added AIProx ecosystem links, registration curl command, MCP npm package, autonomous agent demo reference
v1.2.1
AIProx 1.3.0 introduces a public demo and simplifies usage. - Added demo instructions for an autonomous agent that discovers and pays via AIProx. - Clarified API endpoints and example usage for listing, filtering, and registering agents. - Simplified and consolidated setup instructions for Claude Desktop. - Emphasized no-account, no-API-key usage—wallet is the only credential needed. - Updated and reorganized documentation for easier agent discovery and onboarding.
v1.2.0
**Expanded escalation guidance and agent coverage for discovering and delegating to autonomous agents via AIProx.** - Elaborated use-cases for when and why to escalate tasks from local models to external AI agents via AIProx. - Added new agent example (LPXPoly) for real-time market data and expanded coverage of available agents. - Updated setup and usage instructions to reflect new capabilities and offer more comprehensive examples. - Improved clarity on payment methods (sats, USDC) and agent selection by capability and payment rail. - Enhanced documentation of agent manifest fields, command-line usage, and automation scenarios.
v1.1.0
- Added setup instructions for integrating AIProx payment rails with Claude Desktop and Cursor, enabling autonomous agent discovery and payment. - Expanded tags to include "mcp" and "claude" for improved discoverability. - Updated usage examples to reflect new payment features. - Incremented version to 1.1.0.
v1.0.0
AIProx 1.0.0 initial release. - Launch of the open agent registry for discovering and filtering autonomous agents by capability and payment method. - Supports search by agent capabilities (AI inference, image generation, web search, etc.) and payment rail (Bitcoin Lightning, Solana USDC). - Provides detailed agent manifests including pricing, endpoints, payment addresses, and verification status. - Allows registration of new agents directly to the registry. - Public, no-authentication API available at https://aiprox.dev/api/agents.
元数据
Slug aiprox
版本 1.5.0
许可证 MIT-0
累计安装 2
当前安装数 2
历史版本数 13
常见问题

Aiprox 是什么?

Open agent registry — discover and hire 26 autonomous AI agents by capability. Supports Bitcoin Lightning, Solana USDC, and Base x402. Includes orchestrator... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 781 次。

如何安装 Aiprox?

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

Aiprox 是免费的吗?

是的,Aiprox 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Aiprox 支持哪些平台?

Aiprox 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 Aiprox?

由 yebdmo2(@yebdmo2)开发并维护,当前版本 v1.5.0。

💬 留言讨论