Echo Vault
Echo Vault — 認証情報ブローカー MCP
エージェントは認証情報を「使用」できます。エージェントは認証情報を「読み取る」ことはできません。
不透明な vault:// ハンドル。保存時のAES-GCM。フェイルクローズ。get_secret ツールなし。stdout/MCP の秘匿化が有効。
リポジトリ: echoomegaprime/echo-credential-broker
Grok、GPT、Claude へのインストール
完全な手順 → docs/INSTALL.md
クライアント | クイックパス |
Grok | GitHubアプリをインストール · コネクタURL |
GPT | Codexコネクタをインストール · |
Claude | Claudeアプリをインストール · デスクトップ: |
git clone https://github.com/echoomegaprime/echo-credential-broker.git
cd echo-credential-broker
npm install
cp .env.example .env
npm run dev # console + HTTP MCP
npm run mcp # stdio MCP for Claude Desktop / Claude Code / GrokRelated MCP server: Credential Vault MCP
ツール(シークレットを返さない)
vault_status · vault_list · vault_metadata · credential_health · credential_use · provider_request · github_get_user · github_get_repository · github_get_file · github_list_repositories · github_create_branch · github_create_or_update_file · cloudflare_zone_get · openai_account_status · stripe_balance · xai_account_status · audit_list · credential_rotate · credential_revoke
許可リストに登録されたホストのみ: api.github.com、api.cloudflare.com、api.openai.com、api.stripe.com、api.x.ai。
スタック
TanStack Start · Vite · AES-256-GCM ボールト · MCP JSON-RPC(/api/mcp + stdio)。
Echo Prime Technologies.
This server cannot be deployed
Maintenance
Related MCP Connectors
Secrets for developers and agents—secure context and workflows without exposing secret values.
- FullmaktOAuthai.fullmakt
Credential broker for AI agents: scoped, revocable API access with policy enforcement and audit.
Issue, rotate and revoke scoped API-key passes for 25+ providers — the agent never sees a real key
Encrypted secret store and rotation for autonomous agent credentials
Related MCP Servers
- AlicenseAqualityBmaintenanceSecure credential management for AI agents — encrypted storage, auto-refresh, and rate limiting via a single MCP call. Supports OpenAI, Anthropic, Stripe, GitHub, Brave Search, Kalshi, Coinbase, plus generic API key and OAuth2 for any service.444 npmMIT
- AlicenseNot gradedqualityCmaintenanceEnables secure credential storage for AI agents by encrypting secrets and providing agent-invisible references, ensuring sensitive data never leaks to the model.MIT
- AlicenseNot gradedqualityDmaintenanceCredential resilience for AI agents, providing AES-256 encrypted storage, automatic health checks, and failover to backup credentials for providers like OpenAI, Anthropic, GitHub, and Stripe.MIT
- AlicenseAqualityBmaintenanceEnables AI agents to make authenticated API calls and run commands with secrets injected, while keeping credentials completely hidden from the model, with policy enforcement, grants, and audit logging.6MIT