Skip to main content
Glama
GYINT

QCM MCP Server

by GYINT

QCM · Quality Crisis Management

Skill for quality issue resolution guidance. v1.0.0 release · Encapsulates "quality crisis handling + system governance assessment + knowledge accumulation" into a reusable problem-solving pipeline.

Status Version Tests License


5-Second Overview

# MCP server(10 工具:问题解决核心 6 + Infoseek 协同 3 + 插件 1)
python scripts/mcp_server.py
输入场景 → 协议匹配(action-orders 14 章)→ 场景路由(意图×领域)→ 4 形态输出
案例应用 / 决策卡片 / 评估报告 / 快速响应(防御性标注 + 缺口协同)

Related MCP server: LogicMCP Server

🎉 v1.0.0 Release Highlights

Capability

Description

🏗️ 4-tier architecture

Strict separation of output/protocol/input/underlying layers + validation layer + collaboration layer (5 paradigms throughout)

📜 14-chapter protocol authority

action-orders.md as the single source of truth: AO cards / crisis management / trigger matrix / double zeroing / 4 governance layers

🧭 Scenario routing and consumption

Intent × domain → form × component dynamic composition (confidence gate + no-match fallback)

🧩 Component pool: three mechanisms

Normalized registration (capacity constraints) + popularity tiers (new/active/stable) + constraint mapping (intent × D × complexity)

🛡️ Defensive output

[unverified] annotation + data timeliness + boundary statements + no-rephrase rule

🤝 Infoseek collaboration

5-gap triggering + hybrid strategy with 3 phases + L0→L3 fallback chain (optional dependency)

Regression all green

38 suites PASS + validator 96/96 + guard 9 checks 0 issues


Quick Start

1. Install / Upgrade

pip install -r requirements.txt          # 核心依赖(PyYAML)
pip install -r requirements-optional.txt # 可选能力(LLM / GraphQL / OTel 等)
# 完整依赖说明见 DEPENDENCIES.md

2. MCP Integration

The project provides .mcp.json (dual-server configuration, loadable directly by clients such as Claude/Codex):

{
  "mcpServers": {
    "qcm-search": {
      "command": "${QCM_ROOT}/scripts/mcp_server.py",
      "args": ["--transport", "stdio"],
      "env": {
        "QCM_ROOT": "${QCM_ROOT}",
        "QCM_AUTH_TOKEN": "${QCM_AUTH_TOKEN}"
      }
    },
    "infoseek-search": {
      "command": "${INFOSEEK_ROOT}/scripts/infoseek_mcp_server.py",
      "args": ["--transport", "stdio"],
      "env": {
        "INFOSEEK_ROOT": "${INFOSEEK_ROOT}"
      }
    }
  }
}

💡 On Windows, set command to python3 and args: ["脚本路径", "--transport", "stdio"].

Tool list (10 in total):

  • Issue Resolution Core – 6: qcm_research / qcm_decide / qcm_solve_problem / qcm_score_source / qcm_audit / qcm_validate

  • Infoseek collaboration – 3: qcm_attribution / qcm_attribution_phase / qcm_gap_detect

  • Plugin extension – 1: qcm_plugin_echo (hot-reload example)

3. Environment Variables

Variable

Purpose

Default

QCM_ROOT

QCM installation root (path normalization)

Self-derived

INFOSEEK_ROOT

Infoseek installation root (cross-Skill collaboration)

Probe list

QCM_KEYWORDS

Overrides the keyword-library path

references/config/keyword.yaml

MCP_AUTH_TOKEN

MCP authentication Token


Documentation Navigation

Document

Purpose

SKILL.md

Full Skill definition (concepts / capabilities / trigger words)

version_history.md

External version history

DEPENDENCIES.md

External dependency list + purpose + degradation paths

API_KEYS.md

External API Key list + benefits (no real key values)

references/protocol/action-orders.md

Single authority for the 14-chapter protocol

references/contract/

Input contracts (mds-input + guidelines)

references/config/

Keyword library + routing + component configuration

docs/CHANGELOG.md

Version changelog

docs/INSTALL.md

Installation guide

docs/TROUBLESHOOTING.md

Troubleshooting

tests/

Test suites (38, aggregated by run_all.py)


Version Roadmap

Version

Status

Notes

v1.0.0

🟢 Current release

Version normalization · 4-tier architecture · 14-chapter protocol · scenario routing and consumption · component pool three mechanisms

0.0.0-pre

✅ Historical

Pre-release baseline (original internal V8.3.1 system; history archived in docs/CHANGELOG.md)

Future

🟡 Pending

Keyword hit-rate elimination · suggest_research wiring · industry knowledge expansion


Test Matrix

⚠️ The case counts are suite-reported (PASS counts). Entry point: python tests/run_all.py (script style, not pytest).

Suite Group

Suite Count

Category

engines (8)

qcm_all / cross / loop / combo / super / reverse / full / solution / giant

Engine regression

basic (20)

qcm_mcp v0.1→v0.9 series

MCP protocol + capability evolution

protocol (8)

v1.1→v1.6 series (GraphQL / OTel / BS / WS / dual-channel)

Advanced protocols

main (2)

qcm_v82_test (27 suites) / qcm_router_sequence (15 suites)

Scenario routing / route risk regression

Environment notes: v021 (requires LLM Key) and v071 (Infoseek ws removed) automatically SKIP when the environment is missing so as not to report false FAIL; v131/v151 require the OTel dependency (env_restore.sh --deps one-click restore).


Project Structure

QCM/
├── SKILL.md               # Skill 定义(yfm + 文档)
├── manifest.yaml          # 平台 manifest
├── skill_meta.json        # 元数据(版本/范式/维度/输出形态)
├── README.md              # 本文件
├── version_history.md     # 版本历史
├── DEPENDENCIES.md        # 依赖声明
├── API_KEYS.md            # Key 声明(不含真实值)
├── core/                  # 核心库(路由/组装/验证/组件扫描/缺口检测)
│   ├── router.py          # 场景路由(意图×领域 → 形态×组件)
│   ├── assembler.py       # 输出组装(母模板锚点提取)
│   ├── validator.py       # 4 形态验证器(96/96)
│   ├── component_scan.py  # 组件热度分级
│   └── gap_detector.py    # 5 维缺口检测
├── scripts/               # 适配层 + MCP server + 工具模块
│   ├── mcp_server.py      # MCP server 门面(10 工具)
│   ├── infoseek_bridge.py # QCM↔Infoseek 归因桥(L0→L3 降级)
│   ├── sync_manifest.py   # 双绑校验
│   ├── config_sync.py     # 守卫 9 检
│   └── ...
├── references/            # 协议 + 契约 + 词库 + 配置 + 索引
│   ├── protocol/action-orders.md  # 14 章协议权威
│   ├── contract/          # mds-input + 输入指引
│   ├── config/            # keyword/entities/components/constraint 等 10 配置
│   └── index/             # cases/knowledge-base/masters/tools 索引
├── components/            # 组件池(28 组件,索引自动生成勿手改)
├── outputs/               # 4 形态模板(唯一真源)
├── plugins/               # 插件扩展(热加载)
├── deploy/                # 部署资产(docker/k8s/monitoring/api)
├── docs/                  # CHANGELOG / INSTALL / TROUBLESHOOTING / eval
└── tests/                 # 38 套件(basic/protocol/engines + 主测试)

Contributing & Feedback

  • Bug report: include qcm --version (or server startup logs) + a minimal reproducible case

  • Protocol issues: include the relevant action-orders section number

  • Feature requests: include use cases + expected output form

  • Test failures: run python tests/run_all.py and provide the full output


v1.0.0 | 4-tier architecture · 14-chapter protocol · 4 forms · scenario routing and consumption · component pool three mechanisms | industry-wide quality crisis management | Apache-2.0 License

Related MCP Connectors

Related MCP Servers

  • A
    license
    Not graded
    quality
    F
    maintenance
    Enterprise-grade MCP framework for building production-ready AI tools with data quality, security, audit trails, and comprehensive testing.
    68
    Business Source 1.1
  • F
    license
    Not graded
    quality
    B
    maintenance
    Enables querying quality procedures in natural language, retrieving relevant passages via RAG, and calculating supplier KPIs such as PPM and non-conformity rates through MCP-exposed tools.
    -
  • F
    license
    Not graded
    quality
    C
    maintenance
    Enables DORA-compliant ICT contract management through 10 MCP tools, including mandatory clause checking, critical/important function analysis, subcontracting chain assessment, exit readiness evaluation, and risk scoring.
    -