Colber

Name: Colber
Author: Obi49

@Obi49

代码仓库

关于 Colber

暂无概览

基本信息

分类

其他

传输方式

stdio

发布者

Obi49

提交者

Obi49

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "colber": {
      "command": "npx",
      "args": [
        "-y",
        "@colber/mcp"
      ],
      "env": {
        "COLBER_BASE_URLS": "{\"identity\":\"https://api.colber.dev/identity\",\"reputation\":\"https://api.colber.dev/reputation\",\"memory\":\"https://api.colber.dev/memory\",\"observability\":\"https://api.colber.dev/observability\",\"negotiation\":\"https://api.colber.dev/negotiation\",\"insurance\":\"https://api.colber.dev/insurance\"}"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Colber?

Colber is an infrastructure layer of trust, coordination, and continuity for the agent economy. It provides five integrated services—Reputation, Memory, Observability, Negotiation, and Insurance—exposed through SDKs (TypeScript, Python) and the Model Context Protocol (MCP).

How to use Colber?

Install the SDK via npm (@colber/sdk) or pip (colber-sdk), then create a ColberClient pointing to https://api.colber.dev. For MCP integration, add @colber/mcp as a server in your MCP client config (e.g., Claude Desktop) using npx -y @colber/mcp. All capabilities are then accessible programmatically or as 27 MCP tools.

Key features of Colber

Cryptographic reputation with offline-verifiable attestations (Ed25519 + JCS)
Persistent semantic memory with vector search and ACLs
Distributed A2A tracing and logging via ClickHouse (OpenTelemetry-compatible)
Multi-party negotiation with auctions, multi-criteria, and signed settlement
Deliverable insurance with reputation-based pricing, escrow, and arbitration
Full open-core surface (SDKs, MCP server, types) under Apache-2.0

Use cases of Colber

An AI agent verifying another agent’s reputation before a transaction
Storing and retrieving agent-specific memory with vector search
Debugging multi-agent workflows with distributed observability
Negotiating resource allocation or task delegation among multiple agents
Insuring agent deliverables against non‑performance

FAQ from Colber

What is Colber for, and how does it differ from other agent tools?

Colber is the infrastructure for trust, coordination, and continuity across agents—not a single-purpose tool. It bundles five integrated modules (Reputation, Memory, Observability, Negotiation, Insurance) behind one SDK/MCP interface, providing a full lifecycle for agent-to-agent interactions.

What dependencies or runtime does Colber require?

Using the SDKs only requires a JavaScript (Node.js 18+/Deno/Bun) or Python (3.8+) runtime. The MCP server needs Node.js 22+. The hosted platform runs on colber.dev and is accessed via HTTPS; there is no self‑hosting of the backend.

Where does agent data live?

All data processed by the Colber modules (reputation scores, memory entries, traces, negotiations, insurance contracts) lives on the Colber hosted platform (api.colber.dev). The open‑core SDKs and MCP server are client‑side only and never store data.

What transports and authentication are supported?

The SDKs communicate over HTTPS (REST/gRPC). The MCP server supports stdio (default) and HTTP transport with configurable port. Authentication tokens can be passed via environment variables or config; details are in apps/mcp-server/README.md.

Are there any known limits on the open-core version?

The open‑core repo (this one) contains only client‑side code (SDKs, MCP server, types). The server‑side implementation of the five modules is proprietary and operated solely on colber.dev. There is no limit on the number of agents or requests, but usage is subject to Colber’s Terms of Service.

其他分类下的更多 MCP 服务器

ACI: Open-Source Infra to Power Unified MCP Servers

aipotheosis-labs

ACI.dev is the open source tool-calling platform that hooks up 600+ tools into any agentic IDE or custom AI agent through direct function calling or a unified MCP server. The birthplace of VibeOps.

4.8K

Codelf

unbug

A search tool helps dev to solve the naming things problem.

14.2K

🚀 Model Context Protocol (MCP) Curriculum for Beginners

microsoft

This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,

16.7K