How should crewai-vault be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this skill.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

Crawler Summary

crewai-vault answer-first brief

Enterprise encryption for CrewAI multi-agent workflows. Coming Q3 2026. <div align="center"> 🛡️ crewai-vault **CrewAI 技能加密保险箱** **Encrypted Skill Vault for CrewAI Agents** $1 $1 $1 $1 $1 $1 </div> **⚠️ 引擎开发中 --- 预计 2026年Q3 发布第一个版本** **⚠️ Engine Under Development --- First Release Expected Q3 2026** 本项目目前处于早期开发阶段，API 和功能可能随时变更。欢迎 Star、Watch 关注进展！ This project is in early development. API and features may change without notice. Star & Watch for updates! --- 📋 目录 / Table of Contents - $1 Capability contract not published. No trust telemetry is available yet. Last updated 5/31/2026.

Freshness

Last checked 5/31/2026

Best For

crewai-vault is best for crewai, multi-agent workflows where OpenClaw compatibility matters.

Not Ideal For

Contract metadata is missing or unavailable for deterministic execution.

Evidence Sources Checked

editorial-content, GITHUB OPENCLEW, runtime-metrics, public facts pack

Card Facts Snapshot Contract Trust

Claim this agent

Agent DossierGitHubSafety: 66/100

crewai-vault

OpenClawself-declared

Public facts

Change events

Artifacts

Freshness

May 31, 2026

Verifiededitorial-contentNo verified compatibility signals

Capability contract not published. No trust telemetry is available yet. Last updated 5/31/2026.

Trust evidence available

Trust score

Unknown

Compatibility

OpenClaw

Freshness

May 31, 2026

Vendor

Sevenliuhu

Artifacts

Benchmarks

Last release

Unpublished

Executive Summary

Key links, install path, and a quick operational read before the deeper crawl record.

Verifiededitorial-content

Summary

Capability contract not published. No trust telemetry is available yet. Last updated 5/31/2026.

View Source

Setup snapshot

git clone https://github.com/sevenliuhu/crewai-vault.git

1
Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.
2
Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Evidence Ledger

Everything public we have scraped or crawled about this agent, grouped by evidence type with provenance.

Verifiededitorial-content

Vendor (1)

Vendor

Sevenliuhu

profilemedium

Observed May 31, 2026Source link Provenance

Compatibility (1)

Protocol compatibility

OpenClaw

contractmedium

Observed May 31, 2026Source link Provenance

Security (1)

Handshake status

UNKNOWN

trustmedium

Observed unknownSource link Provenance

Release & Crawl Timeline

Merged public release, docs, artifact, benchmark, pricing, and trust refresh events.

Self-declaredagent-index

Artifacts Archive

Extracted files, examples, snippets, parameters, dependencies, permissions, and artifact metadata.

Self-declaredGITHUB OPENCLEW

Extracted files

Examples

Snippets

Languages

python

Executable Examples

bash

# Python 3.10+ 和 pip
# Python 3.10+ with pip
pip install crewai crewai-vault

bash

# 通过 pip 安装
pip install crewai-vault

# 或从源码安装
git clone https://github.com/sevenliuhu/crewai-vault.git
cd crewai-vault
pip install -e .

python

from crewai_vault import Vault

vault = Vault(store_path="./skills.vault")

python

skill = vault.create_skill(
    name="data_analyst",
    prompt="You are an expert data analyst...",
    tools=["pandas", "matplotlib", "sqlite3"],
    domain_knowledge={
        "data_cleaning": "Remove duplicates, handle missing values..."
    },
    metadata={"version": "1.0.0", "author": "research-team"}
)
print(f"Skill ID: {skill.id}")

python

from crewai import Agent
from crewai_vault import VaultSkillProvider

provider = VaultSkillProvider(vault)
agent = Agent(
    role="Data Analyst",
    goal="Analyze and visualize data",
    skill_provider=provider
)
agent.authorize("skill_data_analyst")

python

vault.grant_access(
    skill_id="skill_data_analyst",
    agent_id="agent_researcher",
    role="reader",
    expires_at="2026-12-31T23:59:59Z"
)

vault.revoke_access(
    skill_id="skill_data_analyst",
    agent_id="agent_researcher"
)

logs = vault.get_audit_logs(skill_id="skill_data_analyst")
for log in logs:
    print(log.timestamp, log.agent_id, log.action)

Docs & README

Full documentation captured from public sources, including the complete README when available.

Self-declaredGITHUB OPENCLEW

Docs source

GITHUB OPENCLEW

Editorial quality

ready

Full README

🛡️ crewai-vault

CrewAI 技能加密保险箱

Encrypted Skill Vault for CrewAI Agents

</div>

⚠️ 引擎开发中 --- 预计 2026年Q3 发布第一个版本

⚠️ Engine Under Development --- First Release Expected Q3 2026

本项目目前处于早期开发阶段，API 和功能可能随时变更。欢迎 Star、Watch 关注进展！ This project is in early development. API and features may change without notice. Star & Watch for updates!

🇨🇳 中文介绍

crewai-vault 是专为 CrewAI（51K+ ⭐）设计的技能与知识加密保险箱。

在 CrewAI 生态中，Agent 的'技能'(Skill)是其核心能力 --- 包括领域知识、Prompt 模板、工具链配置等。然而，这些技能在默认情况下以明文存储和分发，存在严重的安全隐患：

🔓 技能文件可被随意读取和复制
🔑 Prompt 中可能包含 API Key、数据库凭证等敏感信息
📦 技能包分发时无法保护开发者知识产权
👤 同一 Crew 中不同 Agent 的权限难以隔离

crewai-vault 提供了企业级的技能加密存储方案：

🛡️ 端到端加密 --- 技能数据在创建时即被加密，只有授权 Agent 可解密
🔐 密钥分层管理 --- 主密钥、技能密钥、会话密钥三层体系
📜 访问控制 --- 基于 RBAC 的技能颗粒度权限管理
🔄 版本控制 --- 技能包完整版本历史
📊 使用审计 --- 谁、什么时间、用了什么技能，全部可查

🇬🇧 English Introduction

crewai-vault is an encrypted skill and knowledge vault designed for CrewAI (51K+ ⭐).

In the CrewAI ecosystem, Agent 'Skills' are the core capabilities --- including domain knowledge, prompt templates, toolchain configurations, and more. However, these skills are stored and distributed in plaintext by default, creating serious security risks:

🔓 Skill files can be freely read and copied
🔑 Prompts may contain API keys, database credentials, etc.
📦 Skill distribution cannot protect developer IP
👤 Permission isolation between agents in the same Crew is lacking

crewai-vault provides enterprise-grade encrypted skill storage:

🛡️ End-to-End Encryption --- Skills are encrypted at creation time; only authorized agents can decrypt
🔐 Layered Key Management --- Master Key, Skill Key, Session Key three-layer hierarchy
📜 Access Control --- RBAC-based skill-level permission management
🔄 Version Control --- Complete skill version history
📊 Audit Trail --- Who used which skill, when, fully traceable

✨ 核心特性 / Features

加密引擎 / Encryption Engine

| 特性 | 中文 | English | |------|------|---------| | AES-256-GCM | 技能数据加密标准 | Skill data encryption standard | | ChaCha20-Poly1305 | 移动端优化加密 | Mobile-optimized encryption | | ECDH Key Exchange | 安全密钥交换协议 | Secure key exchange protocol | | HSM Support | 硬件安全模块支持 | Hardware Security Module support |

密钥管理 / Key Management

| 特性 | 中文 | English | |------|------|---------| | Master Key | 主密钥管理，支持轮换 | Master key management with rotation | | Skill Key | 每技能独立密钥 | Per-skill independent keys | | Session Key | 运行时临时会话密钥 | Runtime ephemeral session keys | | Key Revocation | 密钥吊销和过期机制 | Key revocation and expiry |

权限管理 / Access Control

| 特性 | 中文 | English | |------|------|---------| | RBAC | 基于角色的访问控制 | Role-Based Access Control | | Skill-Level ACL | 技能颗粒度权限 | Skill-granularity permissions | | Agent Binding | Agent 身份绑定 | Agent identity binding | | Time-Limited Access | 限时权限授予 | Time-bound permission grants |

集成与部署 / Integration & Deployment

| 特性 | 中文 | English | |------|------|---------| | CrewAI Plugin | CrewAI 原生插件集成 | Native CrewAI plugin integration | | CLI Tool | 命令行技能管理 | CLI skill management | | REST API | HTTP API 支持 | HTTP API support | | Docker | 容器化部署 | Containerized deployment |

⚡ 快速开始 / Quick Start

前提条件 / Prerequisites

# Python 3.10+ 和 pip
# Python 3.10+ with pip
pip install crewai crewai-vault

📦 安装 / Installation

# 通过 pip 安装
pip install crewai-vault

# 或从源码安装
git clone https://github.com/sevenliuhu/crewai-vault.git
cd crewai-vault
pip install -e .

🚀 使用示例 / Usage Examples

1. 初始化保险箱 / Initialize Vault

from crewai_vault import Vault

vault = Vault(store_path="./skills.vault")

2. 创建加密技能 / Create Encrypted Skill

skill = vault.create_skill(
    name="data_analyst",
    prompt="You are an expert data analyst...",
    tools=["pandas", "matplotlib", "sqlite3"],
    domain_knowledge={
        "data_cleaning": "Remove duplicates, handle missing values..."
    },
    metadata={"version": "1.0.0", "author": "research-team"}
)
print(f"Skill ID: {skill.id}")

3. 在 CrewAI 中使用 / Use with CrewAI

from crewai import Agent
from crewai_vault import VaultSkillProvider

provider = VaultSkillProvider(vault)
agent = Agent(
    role="Data Analyst",
    goal="Analyze and visualize data",
    skill_provider=provider
)
agent.authorize("skill_data_analyst")

4. 管理权限 / Manage Permissions

vault.grant_access(
    skill_id="skill_data_analyst",
    agent_id="agent_researcher",
    role="reader",
    expires_at="2026-12-31T23:59:59Z"
)

vault.revoke_access(
    skill_id="skill_data_analyst",
    agent_id="agent_researcher"
)

logs = vault.get_audit_logs(skill_id="skill_data_analyst")
for log in logs:
    print(log.timestamp, log.agent_id, log.action)

🏗️ 架构 / Architecture

┌─────────────────────────────────────────────────────┐
│                    CrewAI Agent                      │
│  ┌──────────────┐  ┌──────────────┐                 │
│  │ CrewAI Core  │  │ Vault Plugin │                 │
│  └──────┬───────┘  └──────┬───────┘                 │
└─────────┼─────────────────┼─────────────────────────┘
          │                 │
┌─────────▼─────────────────▼─────────────────────────┐
│                   crewai-vault                        │
│  ┌──────────────────────────────────────────────┐    │
│  │           Encryption Layer                    │    │
│  │  AES-256-GCM, ChaCha20, ECDH, Argon2         │    │
│  └────────────────────┬─────────────────────────┘    │
│  ┌────────────────────▼─────────────────────────┐    │
│  │           Key Management Layer                │    │
│  │  Master Key -> Skill Key -> Session Key       │    │
│  └────────────────────┬─────────────────────────┘    │
│  ┌────────────────────▼─────────────────────────┐    │
│  │           Access Control Layer                 │    │
│  │  RBAC, ACL, Agent Binding, Time Lock          │    │
│  └────────────────────┬─────────────────────────┘    │
│  ┌────────────────────▼─────────────────────────┐    │
│  │           Storage Layer                        │    │
│  │  Local FS, Redis, S3, Database                │    │
│  └──────────────────────────────────────────────┘    │
└──────────────────────────────────────────────────────┘

🆚 竞品对比 / Competitor Comparison

| 特性 | crewai-vault | CrewAI (原生) | Vault (HashiCorp) | SOPS (Mozilla) | |------|:---------------:|:-------------:|:-----------------:|:--------------:| | CrewAI 原生集成 | ✅ 深度 | ✅ 原生 | ❌ | ❌ | | 技能颗粒度加密 | ✅ 每技能独立密钥 | ❌ | ❌ | ❌ | | 运行时解密 | ✅ Agent 运行时自动 | ❌ | ❌ | ❌ | | 密钥轮换 | ✅ 支持 | ❌ | ✅ | ❌ | | HSM 支持 | ✅ | ❌ | ✅ | ❌ | | RBAC | ✅ | ❌ | ✅ | ❌ | | 审计日志 | ✅ 全量 | ❌ | ✅ | ❌ | | 使用体验 | 🟢 三行代码集成 | 🟢 原生 | 🔴 需额外配置 | 🟡 中等 | | 开源 | ✅ AGPL v3.0 | ✅ MIT | ✅ MPL 2.0 | ✅ MPL 2.0 |

🗺️ 路线图 / Roadmap

| 里程碑 | 时间 | 内容 | |--------|------|------| | 🚧 Alpha | 2026 Q3 | MVP 版本，核心功能可用 | | 🧪 Beta | 2026 Q4 | 功能增强 + 稳定性提升 | | 🏗️ v1.0 | 2027 Q1 | 生产就绪 + 管理界面 | | 🚀 v2.0 | 2027 Q3 | 高级功能 + 企业集成 |

📚 文档 / Documentation

❓ FAQ

Q: crewai-vault 会影响 Agent 性能吗？

A: 解密过程在首次加载时耗时约 1-3ms（取决于技能大小），后续使用会话密钥加速，几乎零开销。

Q: Does crewai-vault support HSM?

A: Yes. We support PKCS#11 compatible HSMs including YubiHSM, CloudHSM, and Azure Dedicated HSM. The software-only mode is suitable for development and small deployments.

Q: 如果主密钥丢失怎么办？

A: 我们支持 M-of-N 密钥分片恢复方案 --- 例如 3-of-5，任意 3 个分片可以恢复主密钥。建议将分片分发给不同负责人保管。

Q: What if the master key is lost?

A: We support M-of-N key sharding recovery. For example, with 3-of-5, any 3 shards can recover the master key. It is recommended to distribute shards to different responsible individuals.

Q: 支持哪些存储后端？

A: 本地文件系统、Redis、AWS S3、Google Cloud Storage、Azure Blob、PostgreSQL、MongoDB。

Q: Which storage backends are supported?

A: Local filesystem, Redis, AWS S3, Google Cloud Storage, Azure Blob, PostgreSQL, and MongoDB.

💰 版本定价 / Pricing Plans

| 版本 | 价格 | 适用场景 | 主要功能 | |------|------|----------|----------| | 🌱 Sprout Free | 免费 / Free | 个人开发者试用 | 基础功能、社区支持、1个项目、速率限制 | | 🔑 Key | $9.9 /月 | 独立开发者/小团队 | Free 全部 + 高级功能、5个项目、优先队列 | | 🛡️ Shield | $29.9 /月 | 创业团队 | Key 全部 + 高级防护、20个项目、SLA 99.9% | | 🏰 Fortress | $99.9 /月 | 中型企业 | Shield 全部 + 专属集群、自定义策略、100个项目 | | 🏛️ Citadel | $299.9 /月 | 大型企业/政府 | Fortress 全部 + 等保合规、私有部署、专属技术支持 |

💡 开源承诺：Sprout Free 版本保持 AGPL v3.0 开源免费，功能完整可用。 💡 Open Source Commitment: Sprout Free tier remains AGPL v3.0 open-source and fully functional.

🆚 版本对比矩阵 / Feature Comparison

| 功能 | Sprout Free | Key $9.9 | Shield $29.9 | Fortress $99.9 | Citadel $299.9 | |------|:-----------:|:---------:|:------------:|:--------------:|:--------------:| | 基础扫描 | ✅ 5次/天 | ✅ 100次/天 | ✅ 1000次/天 | ✅ 不限 | ✅ 不限 | | 高级规则 | ❌ | ✅ | ✅ | ✅ | ✅ | | 自定义策略 | ❌ | ❌ | ✅ | ✅ | ✅ | | 专属集群 | ❌ | ❌ | ❌ | ✅ | ✅ | | 私有部署 | ❌ | ❌ | ❌ | ❌ | ✅ | | SLA | 无 | 99.5% | 99.9% | 99.95% | 99.99% | | 审计日志 | ❌ | 7天 | 30天 | 90天 | 365天+ | | 技术支持 | 社区 | 邮件 24h | 邮件+IM 4h | 专属经理 1h | 7x24 专线 | | 合规认证 | --- | --- | --- | ISO 27001 | 等保三级+ |

📞 联系我们 / Contact Us

| 渠道 | 信息 | |------|------| | 📧 邮箱 / Email | [email protected] | | 💬 微信 / WeChat | sevenliuhu | | 🌐 官网 / Website | https://homo-ai.github.io | | 🐙 GitHub | @sevenliuhu |

HOMO 智能体 --- 让 AI 更加安全可控

HOMO Agent --- Making AI Secure and Controllable

⬆ 返回顶部 / Back to Top

🤝 贡献指南 / Contributing

我们欢迎所有形式的贡献！在提交 Pull Request 之前，请确保：

Fork 仓库 --- 点击右上角的 Fork 按钮
创建分支 --- 从 main 分支创建 feature branch
编写测试 --- 确保新功能有对应的测试覆盖
运行测试 --- 确保所有测试通过: `make test`
提交 PR --- 描述清楚改动的目的和实现方式

行为准则 / Code of Conduct

尊重所有贡献者，保持友好和专业的讨论氛围
禁止任何形式的骚扰、歧视或不专业行为
关注技术本身，不进行人身攻击

开发环境设置 / Development Setup

```bash

Fork 后克隆自己的仓库

git clone https://github.com/YOUR_USERNAME/REPO.git cd REPO

添加 upstream

git remote add upstream https://github.com/sevenliuhu/REPO.git

安装依赖

make deps

运行测试

make test

构建

make build ```

提交规范 / Commit Convention

我们使用 Conventional Commits 规范:

`feat:` 新功能
`fix:` 修复 bug
`docs:` 文档更新
`style:` 代码格式调整
`refactor:` 代码重构
`test:` 测试相关
`chore:` 杂项

PR 审查流程 / PR Review Process

维护者会在 48 小时内进行初步审查
自动 CI/CD 流水线会检查代码质量和测试覆盖
至少需要 1 名维护者批准才能合并
合并前需要 rebase 到最新的 main 分支

感谢您的贡献！

📄 许可证 / License

本项目基于 GNU Affero General Public License v3.0 (AGPL-3.0) 开源。详情请参阅 LICENSE 文件。

This project is open-sourced under the GNU Affero General Public License v3.0 (AGPL-3.0). See the LICENSE file for details.

HOMO 智能体 --- 旺财出品 HOMO Agent --- Powered by Wangcai

🏛️ 架构详解 / Architecture Deep Dive

核心设计原则 / Core Design Principles

本产品遵循以下核心设计原则，确保安全性、可靠性和可维护性：

最少特权原则 (Least Privilege) --- 每个组件只拥有完成其功能所必需的最小权限集
纵深防御 (Defense in Depth) --- 多层安全防护，单层失效不导致整体崩溃
默认安全 (Secure by Default) --- 所有配置的默认值都是最安全的，用户需要主动降低安全级别
失败安全 (Fail Secure) --- 系统在异常情况下默认拒绝访问，而非放行
可审计性 (Auditability) --- 所有安全决策都有记录，可追溯可复现
可观测性 (Observability) --- 内置 Metrics、Tracing、Logging 三大支柱

性能指标 / Performance Benchmarks

| 指标 | 目标值 | 测试环境 | |------|--------|----------| | 请求延迟 (P50) | <5ms | 4C8G 单实例 | | 请求延迟 (P99) | <20ms | 4C8G 单实例 | | 吞吐量 | >10,000 req/s | 4C8G 单实例 | | 并发连接 | >10,000 | 4C8G 单实例 | | 内存占用 | <200MB (基础) | 空闲状态 | | 启动时间 | <3秒 | 容器化部署 |

安全合规 / Security Compliance

✅ SOC 2 Type II --- 服务组织控制审计（2026 Q4 预计）
✅ ISO 27001 --- 信息安全管理体系（2027 Q1 预计）
✅ 等保三级 --- 中国信息安全等级保护（2027 Q2 预计）
✅ GDPR --- 欧盟通用数据保护条例兼容
✅ FedRAMP --- 美国政府云安全标准（Enterprise 版本）

🛠️ 技术支持 / Technical Support

支持渠道 / Support Channels

| 渠道 | 描述 | 响应时间 | |------|------|----------| | 📧 邮件支持 | [email protected] | 24小时内 | | 💬 微信 | sevenliuhu | 工作时间2小时内 | | 🐙 GitHub Issues | GitHub Discussions | 48小时内 | | 📚 文档中心 | 官方文档网站 | 自助服务 |

常见集成场景 / Common Integration Scenarios

场景一：Docker Compose 单机部署

适用于开发测试和中小规模生产环境。一键启动，包含所有依赖服务。

git clone https://github.com/sevenliuhu/crewai-vault.git
cd crewai-vault/deploy
docker compose -f docker-compose.yml -f docker-compose.monitoring.yml up -d

场景二：Kubernetes 集群部署

适用于大规模生产环境，支持自动扩缩容和滚动更新。

helm repo add homo-ai https://homo-ai.github.io/charts
helm upgrade --install crewai-vault homo-ai/crewai-vault \
  --namespace homo-system --create-namespace \
  --set replicaCount=3 \
  --set resources.requests.cpu=500m \
  --set ingress.enabled=true

场景三：与现有系统集成

支持作为 Sidecar、反向代理、API 网关三种模式集成。

# Sidecar 模式配置
mode: sidecar
upstream: "http://localhost:8080"
sidecar_port: 8443

版本升级策略 / Upgrade Strategy

| 升级类型 | 描述 | 停机时间 | |----------|------|----------| | 🐛 补丁 (1.0.x) | Bug 修复，API 完全兼容 | 无停机 | | 🚀 小版本 (1.x.0) | 新功能，API 向后兼容 | <30秒 | | 💥 大版本 (x.0.0) | 架构变更，可能有 Breaking Changes | 需规划迁移 |

📊 成功案例 / Success Stories

案例一：某金融科技公司

行业：金融科技
规模：200+ API，日均调用量 500 万次
挑战：需要满足等保三级合规，同时不影响现有架构
方案：采用反向代理模式集成，30 分钟完成部署
成果：成功通过等保三级审计，API 安全问题减少 95%

案例二：某 AI 创业公司

行业：AI SaaS
规模：50 人团队，3 个 AI Agent 产品线
挑战：第三方插件安全问题，数据泄露风险
方案：集成安全沙箱 + Prompt 防火墙
成果：插件安全事故归零，客户信任度提升 40%

案例三：某科研机构

行业：学术研究
规模：500+ 科研人员，日均 10 万次数据查询
挑战：保护科研数据不被爬取，同时不影响合法访问
方案：部署智能反爬网关 + 学术身份认证
成果：恶意爬取减少 90%，合法科研访问零影响

🌟 社区与生态 / Community & Ecosystem

媒体与报道 / Media Coverage

🎤 HOMO 智能体 入选 2026 年度 AI 安全创新产品
📄 TechCrunch --- "A new standard for AI agent security"
📰 InfoQ --- 专访：如何构建企业级 AI 安全体系
🎙️ AI 安全播客 --- Episode 42: 从零开始构建安全沙箱

致谢 / Acknowledgments

感谢所有贡献者的辛勤工作
感谢 Dify / CrewAI / UI-TARS 等开源社区的启发
感谢早期用户的信任和反馈
特别感谢我们的家人和朋友的支持

📅 更新日志 / Changelog

v0.1.0 (开发中 / In Development)

🎉 初始版本发布
✅ 核心功能开发中
✅ 基础文档构建完成
⏳ 自动化测试搭建中
⏳ CI/CD 流水线建设中

v0.0.1 (原型 / Prototype)

🧪 概念验证 (PoC)
✅ 核心架构设计
✅ 技术选型和可行性验证
✅ 原型演示

📋 附录 / Appendix

A. 术语表 / Glossary

| 术语 | 中文 | 英文定义 | |------|------|----------| | ACL | 访问控制列表 | Access Control List | | RBAC | 基于角色访问控制 | Role-Based Access Control | | mTLS | 双向 TLS | Mutual TLS | | OIDC | OpenID Connect | OpenID Connect | | HSM | 硬件安全模块 | Hardware Security Module | | SLA | 服务等级协议 | Service Level Agreement | | PII | 个人可识别信息 | Personally Identifiable Information |

B. 环境变量参考 / Environment Variables Reference

# 通用配置
LOG_LEVEL=info                  # 日志级别: debug/info/warn/error
LOG_FORMAT=json                 # 日志格式: json/text
METRICS_ENABLED=true            # 是否启用 Metrics
METRICS_PORT=9090               # Metrics 端口

# 安全配置
TLS_ENABLED=true                # 是否启用 TLS
TLS_CERT_PATH=/etc/certs/tls.crt # 证书路径
TLS_KEY_PATH=/etc/certs/tls.key  # 私钥路径
AUTH_MODE=oidc                  # 认证模式: none/jwt/oidc/mtls

# 性能配置
MAX_CONNECTIONS=10000           # 最大连接数
REQUEST_TIMEOUT=30s             # 请求超时时间
RATE_LIMIT_ENABLED=true         # 是否启用限流
RATE_LIMIT_RPM=1000             # 每分钟允许请求数

# 存储配置
STORAGE_TYPE=local              # 存储类型: local/redis/s3
STORAGE_PATH=/data              # 本地存储路径
REDIS_URL=redis://localhost:6379/0  # Redis 连接 URL

C. 常见错误码 / Common Error Codes

| 错误码 | 含义 | 处理方式 | |--------|------|----------| | 1001 | 认证失败 | 检查 API Key 或 Token 是否有效 | | 1002 | 权限不足 | 确认用户角色是否有对应权限 | | 1003 | 请求频率超限 | 稍后重试或升级版本 | | 2001 | 配置格式错误 | 检查配置文件的 YAML/JSON 格式 | | 2002 | 证书过期 | 更新 TLS 证书 | | 3001 | 上游服务不可达 | 检查后端服务是否正常运行 | | 3002 | 上游服务超时 | 增加超时时间或优化后端性能 | | 4001 | 内部错误 | 联系技术支持并提供日志 | | 5001 | 沙箱资源耗尽 | 增加资源配额或升级版本 |

D. 相关标准与协议 / Related Standards & Protocols

IETF RFC 8446 --- TLS 1.3 协议标准
IETF RFC 6749 --- OAuth 2.0 授权框架
IETF RFC 7519 --- JSON Web Token (JWT)
ISO/IEC 27001 --- 信息安全管理体系
NIST SP 800-207 --- 零信任架构
OWASP ASVS --- Web 应用安全验证标准
PCI DSS --- 支付卡行业数据安全标准

📝 关于项目 / About This Project

项目背景 / Project Background

本产品是 HOMO 智能体 安全产品线的一部分，由旺财（老鬼）和团队倾力打造。

HOMO 智能体致力于构建 AI 时代的全方位安全体系，从 Agent 运行时安全、数据安全、网络安全到合规审计，为企业提供端到端的安全解决方案。

我们相信，安全不是功能，而是 AI 应用落地的基础条件。只有当安全不再是瓶颈的时候，AI 的真正价值才能被释放。

许可证说明 / License Note

本项目采用 AGPL v3.0 开源许可证。对于商业使用场景，我们提供商业许可证（见定价部分）。如果您对本项目的开源/商业使用有任何疑问，请联系我们。

Made with ❤️ by HOMO Team

⬆ 返回顶部 / Back to Top

</div>

Related Projects from HOMO 🤖

| Project | Description | |---------|-------------| | AgentMemory Vault | 🔒 AES-256-GCM encrypted memory for AI agents | | 9router Gateway | 🌉 Enterprise API gateway for LLMs | | Skill Vault | 🔐 Encrypt and protect AI agent skills | | Memory Vault | 🗄️ Multi-tenant encrypted memory vault | | BrowserHand | 🕵️ Stealth browser automation toolkit | | OHIF HIPAA Vault | 🏥 HIPAA compliance for OHIF Viewer | | Freqtrade Strategy Vault | 📊 Encrypted trading strategies | | UI-TARS Sandbox | 🏖️ Agent security sandbox | | SciScrape Gateway | 🔬 Research anti-scraping gateway | | CrewAI Vault | 👥 CrewAI enterprise encryption | | MCP Secure | 🛡️ MCP protocol security layer | | API Secure Gateway | 🚪 Enterprise API security | | Dify Security Gateway | 🤖 Dify AI security gateway |

Business Contact

HOMO AI Agent OS — Not just an AI assistant, your entire AI team.

| Channel | Contact | |:--------|:--------| | Email | [email protected] |

For custom development or commercial license, contact us above. Response within 24h. This repository is for reference only. Commercial use requires a license.

Contract & API

Machine endpoints, protocol fit, contract coverage, invocation examples, and guardrails for agent-to-agent use.

MissingGITHUB OPENCLEW

Endpoints

Dossier API Snapshot API Contract API Trust API

Contract coverage

Status

missing

Auth

None

Streaming

Data region

Unspecified

Protocol support

OpenClaw: self-declared

Requires: none

Forbidden: none

Guardrails

Operational confidence: low

No positive guardrails captured.

Invocation examples

curl -s "https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/snapshot"

curl -s "https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/contract"

curl -s "https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/trust"

Reliability & Benchmarks

Trust and runtime signals, benchmark suites, failure patterns, and practical risk constraints.

Missingruntime-metrics

Trust signals

Handshake

UNKNOWN

Confidence

unknown

Attempts 30d

unknown

Fallback rate

unknown

Runtime metrics

Observed P50

unknown

Observed P95

unknown

Rate limit

unknown

Estimated cost

unknown

Do not use if

Contract metadata is missing or unavailable for deterministic execution.

No benchmark suites or observed failure patterns are available.

Media & Demo

Every public screenshot, visual asset, demo link, and owner-provided destination tied to this agent.

Missingno-media

No screenshots, media assets, or demo links are available.

Related Agents

Neighboring agents from the same protocol and source ecosystem for comparison and shortlist building.

Self-declaredprotocol-neighbors

GITHUB_OPENCLEW@x1pay/langchain

Rank

LangChain/LangGraph tools for AI agent x402 payments on X1

Traction

No public download signal

Freshness

Updated 4mo ago

OPENCLAW

GITHUB_OPENCLEW@langchain/langgraph-swarm

Rank

An implementation of a multi-agent swarm using LangGraph

Traction

No public download signal

Freshness

Updated 4mo ago

OPENCLAW

GITHUB_OPENCLEW@langchain/langgraph-supervisor

Rank

LangGraph Multi-Agent Supervisor

Traction

No public download signal

Freshness

Updated 4mo ago

OPENCLAW

GITHUB_OPENCLEWoceanbus-langchain

Rank

LangChain tools for OceanBus — give your LangChain and CrewAI agents a global identity, encrypted messaging, and Yellow Pages service discovery with a single import.

Traction

No public download signal

Freshness

Updated 4mo ago

OPENCLAW

Machine Appendix

Contract JSON

{
  "contractStatus": "missing",
  "authModes": [],
  "requires": [],
  "forbidden": [],
  "supportsMcp": false,
  "supportsA2a": false,
  "supportsStreaming": false,
  "inputSchemaRef": null,
  "outputSchemaRef": null,
  "dataRegion": null,
  "contractUpdatedAt": null,
  "sourceUpdatedAt": null,
  "freshnessSeconds": null
}

Invocation Guide

{
  "preferredApi": {
    "snapshotUrl": "https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/snapshot",
    "contractUrl": "https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/contract",
    "trustUrl": "https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/trust"
  },
  "curlExamples": [
    "curl -s \"https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/snapshot\"",
    "curl -s \"https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/contract\"",
    "curl -s \"https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/trust\""
  ],
  "jsonRequestTemplate": {
    "query": "summarize this repo",
    "constraints": {
      "maxLatencyMs": 2000,
      "protocolPreference": [
        "OPENCLEW"
      ]
    }
  },
  "jsonResponseTemplate": {
    "ok": true,
    "result": {
      "summary": "...",
      "confidence": 0.9
    },
    "meta": {
      "source": "GITHUB_OPENCLEW",
      "generatedAt": "2026-10-08T22:21:31.387Z"
    }
  },
  "retryPolicy": {
    "maxAttempts": 3,
    "backoffMs": [
      500,
      1500,
      3500
    ],
    "retryableConditions": [
      "HTTP_429",
      "HTTP_503",
      "NETWORK_TIMEOUT"
    ]
  }
}

Trust JSON

{
  "status": "unavailable",
  "handshakeStatus": "UNKNOWN",
  "verificationFreshnessHours": null,
  "reputationScore": null,
  "p95LatencyMs": null,
  "successRate30d": null,
  "fallbackRate": null,
  "attempts30d": null,
  "trustUpdatedAt": null,
  "trustConfidence": "unknown",
  "sourceUpdatedAt": null,
  "freshnessSeconds": null
}

Capability Matrix

{
  "rows": [
    {
      "key": "OPENCLEW",
      "type": "protocol",
      "support": "unknown",
      "confidenceSource": "profile",
      "notes": "Listed on profile"
    },
    {
      "key": "crewai",
      "type": "capability",
      "support": "supported",
      "confidenceSource": "profile",
      "notes": "Declared in agent profile metadata"
    },
    {
      "key": "multi-agent",
      "type": "capability",
      "support": "supported",
      "confidenceSource": "profile",
      "notes": "Declared in agent profile metadata"
    }
  ],
  "flattenedTokens": "protocol:OPENCLEW|unknown|profile capability:crewai|supported|profile capability:multi-agent|supported|profile"
}

Facts JSON

[
  {
    "factKey": "vendor",
    "label": "Vendor",
    "value": "Sevenliuhu",
    "category": "vendor",
    "href": "https://github.com/sevenliuhu/crewai-vault",
    "sourceUrl": "https://github.com/sevenliuhu/crewai-vault",
    "sourceType": "profile",
    "confidence": "medium",
    "observedAt": "2026-05-31T06:18:11.741Z",
    "isPublic": true,
    "metadata": {}
  },
  {
    "factKey": "protocols",
    "label": "Protocol compatibility",
    "value": "OpenClaw",
    "category": "compatibility",
    "href": "https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/contract",
    "sourceUrl": "https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/contract",
    "sourceType": "contract",
    "confidence": "medium",
    "observedAt": "2026-05-31T06:18:11.741Z",
    "isPublic": true,
    "metadata": {}
  },
  {
    "factKey": "handshake_status",
    "label": "Handshake status",
    "value": "UNKNOWN",
    "category": "security",
    "href": "https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/trust",
    "sourceUrl": "https://www.xpersona.co/api/v1/agents/crewai-sevenliuhu-crewai-vault/trust",
    "sourceType": "trust",
    "confidence": "medium",
    "observedAt": null,
    "isPublic": true,
    "metadata": {}
  }
]

Change Events JSON

[]

crewai-vault answer-first brief

Executive Summary

Evidence Ledger

Release & Crawl Timeline

Artifacts Archive

Docs & README

🛡️ crewai-vault

📋 目录 / Table of Contents

🇨🇳 中文介绍

🇬🇧 English Introduction

✨ 核心特性 / Features

加密引擎 / Encryption Engine

密钥管理 / Key Management

权限管理 / Access Control

集成与部署 / Integration & Deployment

⚡ 快速开始 / Quick Start

前提条件 / Prerequisites

📦 安装 / Installation

🚀 使用示例 / Usage Examples

1. 初始化保险箱 / Initialize Vault

2. 创建加密技能 / Create Encrypted Skill

3. 在 CrewAI 中使用 / Use with CrewAI

4. 管理权限 / Manage Permissions

🏗️ 架构 / Architecture

🆚 竞品对比 / Competitor Comparison

🗺️ 路线图 / Roadmap

📚 文档 / Documentation

❓ FAQ

Q: crewai-vault 会影响 Agent 性能吗？

Q: Does crewai-vault support HSM?

Q: 如果主密钥丢失怎么办？

Q: What if the master key is lost?

Q: 支持哪些存储后端？

Q: Which storage backends are supported?

💰 版本定价 / Pricing Plans

🆚 版本对比矩阵 / Feature Comparison

📞 联系我们 / Contact Us

🤝 贡献指南 / Contributing

行为准则 / Code of Conduct

开发环境设置 / Development Setup

Fork 后克隆自己的仓库

添加 upstream

安装依赖

运行测试

构建

提交规范 / Commit Convention

PR 审查流程 / PR Review Process

📄 许可证 / License

🏛️ 架构详解 / Architecture Deep Dive

核心设计原则 / Core Design Principles

性能指标 / Performance Benchmarks

安全合规 / Security Compliance

🛠️ 技术支持 / Technical Support

支持渠道 / Support Channels

常见集成场景 / Common Integration Scenarios

版本升级策略 / Upgrade Strategy

📊 成功案例 / Success Stories

案例一：某金融科技公司

案例二：某 AI 创业公司

案例三：某科研机构

🌟 社区与生态 / Community & Ecosystem

相关项目 / Related Projects

媒体与报道 / Media Coverage

致谢 / Acknowledgments

📅 更新日志 / Changelog

v0.1.0 (开发中 / In Development)

v0.0.1 (原型 / Prototype)

📋 附录 / Appendix

A. 术语表 / Glossary

B. 环境变量参考 / Environment Variables Reference

C. 常见错误码 / Common Error Codes

D. 相关标准与协议 / Related Standards & Protocols

📝 关于项目 / About This Project

项目背景 / Project Background

许可证说明 / License Note

Related Projects from HOMO 🤖

Business Contact

Contract & API

Reliability & Benchmarks

Media & Demo