nobulex-mcp-server
Nobulex
自主 AI 代理的行为证明协议。
每个 AI 代理都会做出承诺——“我不会转账超过 500 美元”、“我只会访问已批准的 API”、“我不会触碰生产数据”。但目前,没有任何方法可以证明代理遵守了这些承诺。日志是由被审计的同一软件编写的。合规性只是被声明,从未被证明。
Nobulex 改变了这一点。定义行为规则。在执行前强制执行。用密码学而非信任来证明合规性。
什么是行为证明?
你无法审计神经网络。但你可以根据既定的承诺审计其行为。
verify(covenant, actionLog) → { compliant: boolean, violations: Violation[] }这是始终可判定的、确定性的且高效的。没有机器学习,没有启发式方法——只有数学证明。
行为证明意味着每个自主代理的行为都是:
已声明 — 在部署前以形式化语言定义行为规则
已强制执行 — 在运行时、执行前拦截违规行为
已证明 — 每个行为都通过哈希链链接到防篡改的审计追踪中,第三方可以独立验证
Related MCP server: Agent Receipts
快速入门
npm install @nobulex/sdkimport { createDID } from '@nobulex/identity';
import { parseSource } from '@nobulex/covenant-lang';
import { EnforcementMiddleware } from '@nobulex/middleware';
import { verify } from '@nobulex/verification';
// 1. Create an agent identity
const agent = await createDID();
// 2. Write behavioral rules
const spec = parseSource(`
covenant SafeTrader {
permit read;
permit transfer (amount <= 500);
forbid transfer (amount > 500);
forbid delete;
}
`);
// 3. Enforce at runtime
const mw = new EnforcementMiddleware({ agentDid: agent.did, spec });
// $300 transfer — allowed
await mw.execute(
{ action: 'transfer', params: { amount: 300 } },
async () => ({ success: true }),
);
// $600 transfer — BLOCKED before execution
await mw.execute(
{ action: 'transfer', params: { amount: 600 } },
async () => ({ success: true }), // never runs
);
// 4. Prove compliance
const result = verify(spec, mw.getLog());
console.log(result.compliant); // true
console.log(result.violations); // []跨代理验证握手
在两个代理进行交易之前,它们会验证彼此的行为证明。没有证明,就没有交易。
import { generateProof, verifyCounterparty } from '@nobulex/sdk';
// Agent A generates its proof-of-behavior
const proof = await generateProof({
identity: agentA,
covenant: spec,
actionLog: middleware.getLog(),
});
// Agent B verifies Agent A before transacting
const result = await verifyCounterparty(proof);
if (!result.trusted) {
console.log('Refusing transaction:', result.reason);
return; // No proof, no transaction
}
// Safe to transact — Agent A is verified
await executeTransaction(proof.agentDid, amount);握手按顺序检查六项内容:契约签名、证明签名、日志完整性、合规性、历史记录最小值以及必需的契约。如果任何检查失败,交易将被拒绝。
为什么行为证明很重要
现有技术 | 缺失内容 |
护栏过滤提示词和输出 | 无法证明代理在行为层遵循了规则 |
监控在事后观察代理的行为 | 无法在执行前进行强制执行 |
身份验证验证代理的身份 | 无法验证代理做了什么 |
治理平台提供仪表板和策略 | 缺乏第三方可独立验证的加密证据 |
行为证明填补了这一空白:声明 → 强制执行 → 证明。
契约 DSL
covenant SafeTrader {
permit read;
permit transfer (amount <= 500);
forbid transfer (amount > 500);
forbid delete;
require counterparty.compliance_score >= 0.8;
}禁止优先。 如果任何 forbid 匹配,无论是否允许,该行为都会立即被拦截。对于未匹配的行为,默认拒绝。条件支持对数值、字符串和布尔字段使用 >, <, >=, <=, ==, !=。
三个关键字。没有配置文件。没有 YAML。没有 JSON 模式。只有规则。
架构
┌─────────────────────────────────────────────────────────────┐
│ Platform │
│ cli · sdk · mcp-server │
├─────────────────────────────────────────────────────────────┤
│ Proof-of-Behavior Stack │
│ │
│ ┌──────────┐ ┌──────────────┐ ┌────────────┐ │
│ │ identity │ │ covenant-lang│ │ action-log │ │
│ │ (DID) │ │ (DSL) │ │(hash-chain)│ │
│ └──────────┘ └──────────────┘ └────────────┘ │
│ │
│ ┌────────────┐ ┌──────────────┐ ┌───────────────┐ │
│ │ middleware │ │ verification │ │ composability │ │
│ │(pre-exec) │ │ (post-hoc) │ │(trust graph) │ │
│ └────────────┘ └──────────────┘ └───────────────┘ │
├─────────────────────────────────────────────────────────────┤
│ Foundation │
│ core-types · crypto · types │
└─────────────────────────────────────────────────────────────┘核心包
包 | 功能 |
使用 Ed25519 密钥创建 W3C DID | |
受 Cedar 启发的 DSL:词法分析器、解析器、编译器 | |
带有 Merkle 证明的 SHA-256 哈希链式防篡改日志 | |
执行前强制执行 — 在运行前拦截违规行为 | |
确定性合规性验证 | |
结合所有原语的统一 API | |
适用于任何兼容 MCP 的代理的 MCP 合规服务器 | |
命令行工具: | |
LangChain 中间件集成 (PyPI) |
集成
npm —
npm install @nobulex/sdkPyPI —
pip install langchain-nobulexMCP —
npx @nobulex/mcp-server(适用于 Claude Desktop, Cursor, VS Code)LangChain — 即插即用的合规中间件
ElizaOS — 用于行为、评估器和提供程序的插件
概念对比
比特币 | 以太坊 | Nobulex | |
验证内容 | 货币转账 | 合约执行 | 代理行为 |
机制 | 工作量证明 | 权益证明 | 行为证明 |
证明内容 | 交易有效性 | 状态转换 | 行为合规性 |
保证 | 无需信任的货币 | 无需信任的合约 | 无需信任的代理 |
在线演示
npx tsx demo/covenant-demo.ts创建一个脚本,定义两个代理,设定行为规则,在运行时强制执行,拦截违规转账,并以加密方式验证合规性。
开发
git clone https://github.com/arian-gogani/nobulex.git
cd nobulex
npm install
npx vitest run # 4,237 tests, 80 files, 0 failures文档
行为证明规范 — 正式标准规范 (CC-BY-4.0)
白皮书 — 正式协议规范
入门指南 — 开发者指南
NIST RFI 回复 — 对 NIST AI 代理标准倡议的正式评论
链接
网站: nobulex.com
npm: @nobulex
PyPI: langchain-nobulex
NIST: Docket NIST-2025-0035 (已提交公开评论)
许可证
MIT — 随意使用。
Available Tools
4 toolscheck_actionA
Check whether an action is allowed or blocked by the current covenant rules.
Inputs: action (required), params.
get_audit_logB
Returns the full hash-chained audit trail of all compliance checks.
Inputs: no named parameters.
set_rulesA
Set covenant rules using permit/forbid/require syntax. Each rule is a string like 'forbid delete_user' or 'permit read_data safe to read'.
Inputs: rules (required).
verify_logA
Independently verify the integrity of the hash-chained audit log. Detects any tampering.
Inputs: no named parameters.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 4 tools
Each tool has a clearly distinct purpose: checking actions, retrieving audit logs, setting rules, and verifying log integrity. No overlap or ambiguity.
All tool names follow a consistent verb_noun pattern in snake_case (check_action, get_audit_log, set_rules, verify_log).
Four tools is an appropriate scope for a compliance/auditing server; each tool serves a necessary function without redundancy.
Covers core operations: rule setting, action checking, audit log retrieval, and log integrity verification. Minor gap: no explicit rule deletion or modification beyond full replacement, but this is acceptable for the domain.
Maintenance
Related MCP Connectors
Pre-execution governance for AI agents. Deterministic PASS/FAIL/REVIEW verdicts, replayable proof.
Policy-aware AI execution, verification, cost, and trajectory evidence for agents.
Check, limit, record and prove AI agent actions with 2-of-3 validators and onchain proofs.
Checks AI agent actions before they run. Signed, RFC 3161 timestamped evidence for audits.
Related MCP Servers
- AlicenseAqualityCmaintenanceProvides covenant rule enforcement, hash-chained audit logs, and integrity verification for MCP-compatible agents. It enables users to define granular permission rules and maintain a tamper-evident audit trail of all actions.41MIT
- AlicenseAqualityBmaintenanceCryptographic accountability for AI agents. Ed25519-signed receipts for every MCP tool call. Constraints, chains, AI judgment, invoicing, and local dashboard included.241MIT
- AlicenseNot gradedqualityDmaintenanceProvides tamper-proof audit logging for AI agents using SHA-256 hash chains, integrity verification, and compliance reporting for the EU AI Act.27 PyPI1MIT
- AlicenseNot gradedqualityCmaintenanceProvides permission gates and tamper-evident audit logging for AI agent tool executions, with declarative policies, consent ladders, and hash-chained verification.MIT