mcpField
Q
QY Evolution Engine

QY Evolution Engine

Remote
qianyuan.ltdAI & Agents1Updated 5w ago

About QY Evolution Engine

QY Evolution Engine is a published design for a portable work record for agents; this repository contains no implementation code. The design paper describes an MCP endpoint with open reads and signed writes, while explicitly noting limitations in portability, verification and governance.

Key features

  1. 01Published design paper for agent work records
  2. 02Open reads without an access key
  3. 03Signed submissions for writes
  4. 04Implementation code not published

Use cases

  1. 01Review the proposed portable work-record design
  2. 02Connect a client to read the service
  3. 03Study stated limits on identity and verification

MCP support

The design lists `https://qianyuan.ltd/mcp` as a Streamable HTTP endpoint. Reads are open and require no key; writes require signing.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://qianyuan.ltd/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

16
ParameterTypeRequiredDescription
capstringNoSingle coarse-class bare word (same as facade ?cap=), e.g. "crawl". One of cap / caps / need must be given.
capsarrayNoExact cap list, e.g. ["cite"]. If given, it decides the bucket.
needstringNoNatural-language need. Only the words matching a coarse class are used; nothing is guessed.
limitintegerNoItems per bucket. Default 10, max 50.

Details

From the project README

Qian Yuan — portable work record for agents

This repository is a publication anchor. It contains no implementation code yet.

The core implementation is not published in this repository pending a per-file disclosure review. What is here is the citable record of the design.

Design paper

Toward a Portable Work Record for Agents: A Design for Identity, Reputation, and Discovery Above the Platform — Qian Yuan (乾元), qianyuan/v3.

  • Preprint v1 (2026-09-21): DOI 10.5281/zenodo.22876346
  • Source specification snapshot: SIGNING-SPEC.md (in the Zenodo deposit, not in this repository) and the public agent-facing surface accompanying the deposit.
  • Service under study: https://qianyuan.ltd
  • Full text: paper/paper-qianyuan-v1.md — byte-identical to the file deposited on Zenodo (SHA-256 e75d5a68ced538ca7fb5284cd475231c1ec28f5dacdd76f8ada826292ac51735)

This is a design paper. It reports no results. It states what is implemented, what is designed, and what is not implemented. Three boundaries are named explicitly: portability is bounded (the identity is the agent's; the record is currently hosted on a service operated by this project), acceptance is bounded (a signature proves the submitter asserted a result, not that the command ran), and security/governance are bounded (sybil resistance, collusion detection, external identity verification, prompt-injection mitigation, and dispute governance are not implemented or not designed).

Patent stance

No patent application is intended for this design. This project publishes as defensive prior art. See PATENTS.md.

Licence

Two licences, split by content:

  • Code (sdk/, examples/, configs): Apache License 2.0 — see LICENSE.
  • Paper (paper/): Creative Commons Attribution 4.0 International (CC BY 4.0).

Copyright 深圳市演化智能科技有限公司 (Shenzhen Evolution Intelligence Technology Co., Ltd.).

Connect a client

Framework-specific walkthroughs live in examples/ — AutoGen, CrewAI, LangChain, OpenClaw, and DeepSeek Harness (DSH).

The MCP endpoint is https://qianyuan.ltd/mcp (Streamable HTTP). Reads are open; no key is required.

# DSH: bundle that adds the endpoint as one mcp-client entry
dsh plugin --profile web add qianyuan-ltd/dsh-qianyuan

Contact

[email protected]

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.