arc-baseline-kernel
Peer-reviewed (ACM BCB 2026) within-subject physiological deviation scoring. Real-time IHB baseline queries with SHA-256 trust certificates and autonomous x402 USDC payments on Base L2. No population norms. No human required.
README
ARC Baseline Grounding Kernel
@ARC_BaselineKernel | Autonomic Resilience Collective
Peer-reviewed | ACM BCB 2026 | DOI: 10.1145/3807503.3816889
Kernel Identity
Kernel: @ARC_BaselineKernel
Organization: Autonomic Resilience Collective (ARC)
Validation: 1,815 tracked nights | 94.4% completeness
Citation: Buckingham & Johnson, ACM BCB 2026
What This Is
A production-ready Model Context Protocol (MCP) server exposing the ARC Baseline Grounding Kernel as a fully autonomous, AI-to-AI (A2A) queryable service. Any AI agent can discover, pay for, and query baseline variance state — with zero human involvement after initial wallet funding.
All computation is deterministic. All numbers originate from ihb/core.py. No LLM smoothing. No fabrication. Every result carries a SHA-256 provenance seal.
Payment Rails
| Rail | Status | Header |
|---|---|---|
| USDC x402 (Base L2) | ✅ Live | X-Payment-Proof: <txhash> |
| Nevermined x402 | ✅ Live (requires plan setup) | X-Nevermined-Auth: <token> |
| Visa/Stripe via Nevermined | ✅ Via Nevermined dashboard | — |
Service Tiers (Nevermined Plan Names)
| Tier | Plan Name | Price | Endpoint |
|---|---|---|---|
| Micro | ARC Micro Baseline Check | $0.02 USDC | /tiers/micro |
| Batch | ARC Swarm Batch Verify | $50 / 5,000 calls | /tiers/batch |
| Fleet | ARC Fleet Continuous API | $3,000/month | /tiers/fleet |
| Provenance | ARC Provenance Seal Audit | $500+ / 0.05% | /tiers/high_stakes |
Environment Variables
# Required
ARC_USDC_WALLET # Your Base L2 USDC receiving address
# Nevermined (register at app.nevermined.app)
NVM_API_KEY # Nevermined API key
NVM_AGENT_ID # Your registered agent DID
NVM_ENVIRONMENT # "sandbox" or "live"
NVM_PLAN_MICRO # "ARC Micro Baseline Check" plan ID
NVM_PLAN_BATCH # "ARC Swarm Batch Verify" plan ID
NVM_PLAN_FLEET # "ARC Fleet Continuous API" plan ID
NVM_PLAN_HIGH_STAKES # "ARC Provenance Seal Audit" plan ID
# Pricing (defaults shown)
USDC_PER_QUERY=0.02
VERIFY_PRICE_STANDARD=1.00
VERIFY_PRICE_ELEVATED=2.50
VERIFY_PRICE_HIGH_STAKES=5.00
REPORT_PRICE_CHAPTER=25.00
REPORT_PRICE_DEEP=50.00
VALUE_BASED_RATE=0.0005
Deployment
# 1. Copy the ihb/ package from your ARC codebase into this directory
cp -r /path/to/ARC-IHB-Engine/ihb ./ihb
# 2. Run the deployment script
chmod +x deploy.sh && ./deploy.sh
Agent Discovery
- MCP Manifest:
/mcp.json - SSE Transport:
/sse - Tool Spec:
llm-tools-spec.json(Hugging Face registry) - llms.txt: Place at
autonomicresiliencecollective.org/llms.txt - Docs:
/docs - Health:
/health
File Map
| File | Purpose |
|---|---|
ihb_mcp_server.py |
Main FastAPI app — all 23 routes |
ihb_state.py |
Stateful baseline wrapper (Layer 2) |
ihb_translator.py |
Temporal privacy front door — strips dates |
ihb_payment.py |
x402 USDC + idempotency cache |
ihb_nvm_payment.py |
Nevermined SDK integration |
ihb_verify_tiers.py |
Multi-tier pricing router |
ihb_verify_action.py |
Enterprise provenance gate |
ihb_categories.yaml |
State → commercial action mapping |
llm-tools-spec.json |
Hugging Face / agent registry manifest |
llms.txt |
Web-crawling AI agent discovery |
render.yaml |
Render.com deployment config |
Dockerfile |
Container build |
requirements.txt |
Python dependencies |
deploy.sh |
GitHub + Render deployment script |
agent_quickstart.py |
A2A purchase demo |
Autonomic Resilience Collective | EIN: 41-2759286 | SAM UEI Active through February 2027
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。