一个用于 agentflow 演示管道的企业架构知识 MCP 服务器。基于 FastMCP 和 TypeScript 构建，并部署在 GCP Cloud Run 上。
该服务器提供四个工具，使架构 Agent 和风险检查 Agent 能够基于精选的企业架构模式而非通用 LLM 推理来开展工作：
工具 | 调用方 | 返回内容 |
| 架构 Agent | 参考架构模式、组件、图表数据 |
| 架构 Agent | 平台推荐，以及基于约束的推理 |
| 风险检查 Agent | 所需控制项、风险标记、HITL 触发 |
| 架构 Agent | 公司身份、定位、Logo（通过 Brandfetch + logo.dev） |
agentflow pipeline agentflow-mcp
┌──────────────────────┐ ┌───────────────────────┐
│ Qualifier Agent │ │ arch_pattern_lookup │
│ - clarifies the ask │ │ tool_selection_lookup │
└──────┬───────────────┘ │ risk_policy_lookup │
│ handoff │ brand_context_lookup │
┌──────▼───────────────┐ │ │
│ Architect Agent │──── MCP calls ───▶│ Source pack (data/) │
│ - pattern selection │ │ 102 markdown files │
│ - tool selection │◀── JSON response ─│ with YAML frontmatter │
│ - diagram rendering │ │ │
└──────┬───────────────┘ │ Brandfetch + logo.dev │
│ handoff │ (cached, additive) │
┌──────▼───────────────┐ └───────────────────────┘
│ Risk Checker Agent │──── risk_policy_lookup ──▶
│ - HITL gate trigger │◀── risk_flags, HITL ──
└──────────────────────┘MCP 是一个工具提供方，而非 Agent 编排器。Agent 提示词和架构图技能位于 agentflow 项目中。MCP 提供结构化数据；Agent 负责解读这些数据并采取行动。
Related MCP server: MCP Architect
Node.js >= 20
（可选）用于 brand_context_lookup 的 Brandfetch API 密钥和 logo.dev 密钥
npm install
npm run dev # stdio transport (local dev + MCP Inspector)MCP_TRANSPORT=http-stream PORT=8080 npm run dev
# agentflow-mcp listening on http://0.0.0.0:8080/mcpnpm test # 31 unit + integration tests
npm run typecheck # tsc --noEmit
npm run check # biome lint + format将 .env.example 复制为 .env 并填入相关密钥。只有 brand_context_lookup 需要外部 API 密钥——其余三个工具都可以离线直接从数据源包中心运行。
变量 | 使用方 | 用途 |
|
| Brandfetch Brand Context API 的 Bearer 令牌 |
|
| logo.dev Brand API 的 Bearer 令牌 |
|
| logo.dev CDN URL 的 publishable 密钥 |
| 服务器 |
|
| 服务器 | HTTP 端口（默认 8080，在传输方式为 |
当 API 密钥缺失时，brand_context_lookup 会为已缓存的域名返回缓存响应，或为未缓存的域名返回一个合理的不可用响应。其余三个工具仍会正常连接。
arch_pattern_lookup将企业的需求匹配到精选的参考架构模式。
输入：
{
"industry": "media_agency",
"data_stack": ["BigQuery", "Snowflake"],
"cloud": "GCP",
"constraints": ["SAML SSO", "EU data residency", "cross-client governance"],
"latency": "batch"
}输出：
{
"pattern_id": "media_agency_audience_measurement",
"architecture_summary": "...",
"recommended_components": ["BigQuery", "Snowflake", "SAML SSO", "GCP EU Region"],
"data_zones": ["bronze", "silver", "gold"],
"integration_notes": ["..."],
"confidence": 0.87,
"diagram_data": {
"components": [{ "name": "BigQuery", "type": "database", "sublabel": "...", "zone": "gold" }],
"connections": [{ "from": "Users", "to": "SAML SSO", "label": "OAuth 2.0", "style": "dashed" }],
"boundaries": [{ "label": "GCP EU Region", "type": "region" }]
},
"source_references": [{ "path": "data/patterns/...", "title": "...", "source_url": "..." }]
}匹配逻辑： 确定性、基于规则——行业匹配（40%）+ 数据栈重叠（30%）+ 约束覆盖（30%）。精选匹配（置信度 >= 0.85）包含 diagram_data 和来源引用。弱匹配则回退到通用企业级 AI POC 模式，置信度 < 0.5。
tool_selection_lookup根据负载、数据栈、约束和延迟来推荐平台。
输入：
{
"use_case": "AI-powered patient insights",
"data_stack": ["Databricks"],
"constraints": ["HIPAA", "PHI", "US data residency"],
"latency": "batch"
}输出：
{
"recommended_platform": "Databricks",
"cloud_fit": "Azure or AWS",
"reasoning": "Strong lakehouse fit for healthcare AI with HIPAA-compliant governance...",
"alternatives": [{ "platform": "Snowflake", "rationale": "..." }, { "platform": "BigQuery", "rationale": "..." }]
}risk_policy_lookup返回特定行业的风险与治理检查，包括针对受监管数据的 HITL 触发条件。
输入：
{
"industry": "healthcare",
"data_classification": ["PHI", "PII"],
"region": "US",
"deployment": "cloud",
"constraints": ["HIPAA"]
}输出：
{
"required_controls": ["RBAC", "audit logs", "data lineage", "SAML SSO"],
"risk_flags": ["prompt leakage", "overbroad analyst access"],
"hitl_required": true,
"review_reason": "PHI access requires human approval before final architecture signoff"
}HITL 会在出现受监管数据类型（PHI、PII、受监管金融数据）时触发，并提供人类可读的 review_reason。
brand_context_lookup通过 Brandfetch 获取丰富的公司上下文，并通过 logo.dev 获取 Logo，同时支持多层缓存。
输入：
{
"domain": "havas.com"
}输出：
{
"company_name": "Havas",
"domain": "havas.com",
"industry_hint": "media_agency",
"description": "...",
"tags": ["advertising", "marketing", "media"],
"positioning": { "value_proposition": "...", "target_audience": "...", "products_and_services": "..." },
"brand": { "voice": "...", "style": "..." },
"logo_url": "https://...",
"confidence": 0.85
}缓存层级： (1) Brandfetch cachedOnly=true，用于即时缓存查询；(2) 本地文件缓存，带 TTL。重复查询会返回缓存数据，不消耗 API 配额。当 API 不可访问时，提供优雅降级。
data/ 目录包含 102 个 Markdown 文件，带有结构化的 YAML frontmatter，组织方式如下：
data/
├── industry/ # Industry-specific architecture notes
├── vendors/ # Vendor documentation (GCP, AWS, Azure, Snowflake, Databricks)
└── patterns/ # Curated reference architecture patterns (4 demo scenarios)Frontmatter 字段：type, domain、title, source_url, vendor, industry, data_stack, cloud, constraints, compliance, region, data_zones, latency, pattern_id, architecture_summary, recommended_components, integration_notes, confidence_baseline, diagram_data.
数据源包在服务器启动时会加载到内存索引中，并以行业、数据栈、约束和 pattern_id 作为键。
场景 | 行业 | 模式 ID |
媒体代理受众测量 |
|
|
医疗患者洞察 |
|
|
零售 湖仓个性化 |
|
|
FSI 治理 Copilot |
|
|
docker build -t agentflow-mcp .
docker run -p 8080:8080 agentflow-mcpgcloud run deploy agentflow-mcp \
--source . \
--region run.googleapis.com \
--port 8080 \
--set-env-vars "MCP_TRANSPORT=http-stream" \
--set-secrets "BRANDFETCH_API_KEY=brandfetch-api-key:latest,LOGO_DEV_SECRET_KEY=logo-dev-secret-key:latest,LOGO_DEV_PUBLISHABLE_KEY=logo-dev-publishable-key:latest"完整的服务配置请参见 cloud-run.yaml。
App Engine Standard 不会运行构建步骤——请在本地编译后再部署：
npm run build # compile src/ -> dist/
# (Optional) Warm brand cache for demo domains before deploy
npx tsx scripts/brand-cache-warm.ts
gcloud app deploy # deploys with dist/ and data/ includedapp.yaml 会设置 MCP_TRANSPORT=http-stream，并在空闲时将实例缩放至零（对演示来说更便宜）。App Engine 会自动设置 PORT——服务器仍然会读取该变量。
如需使用 Secret Manager 管理密钥：
# Create secrets
gcloud secrets create brandfetch-api-key --data-file=<(echo -n "$BRANDFETCH_API_KEY")
gcloud secrets create logo-dev-secret-key --data-file=<(echo -n "$LOGO_DEV_SECRET_KEY")
gcloud secrets create logo-dev-publishable-key --data-file=<(echo -n "$LOGO_DEV_PUBLISHABLE_KEY")
# Reference them in app.yaml (uncomment the includes: section)完整配置请参见 app.yaml 和 .gcloudignore。
脚本 | 用途 |
| 校验 |
| 为数据源包文件生成 frontmatter |
| 验证五个工具能否通过 MCP 工具列表被发现 |
| 作用于运行 MCP 的 Docker 执行器使用的所有辅助进程 |
npx tsx scripts/validate-source-pack.ts # validate source pack
npx tsx scripts/mcp-list-check.ts # verify tool discovery
npx tsx scripts/brand-cache-warm.ts # warm brand cachenpx @modelcontextprotocol/inspector npm run dev这会启动 MCP Inspector UI，你可以通过它交互式地调用工具并验证响应。
agentflow-mcp/
├── src/
│ ├── index.ts # MCP server entry point (stdio + http-stream)
│ ├── tools/
│ │ ├── archPatternLookup.ts # Pattern matching + confidence scoring
│ │ ├── toolSelectionLookup.ts # Platform recommendation
│ │ ├── riskPolicyLookup.ts # Risk/governance checks + HITL
│ │ └── brandContextLookup.ts # Brandfetch + logo.dev with caching
│ ├── data/
│ │ ├── loader.ts # Source pack parser + in-memory index
│ │ ├── brandfetchClient.ts # Brandfetch Brand Context API client
│ │ ├── logoDevClient.ts # logo.dev Brand API client
│ │ └── brandCache.ts # Local file cache with TTL
│ └── types/
│ ├── source.ts # Source pack entry types
│ ├── arch-pattern.ts # arch_pattern_lookup types
│ ├── tool-selection.ts # tool_selection_lookup types
│ ├── risk-policy.ts # risk_policy_lookup types
│ └── brand-context.ts # brand_context_lookup types
├── data/ # Source pack (102 markdown files)
│ ├── industry/
│ ├── vendors/
│ └── patterns/
├── tests/ # Unit + integration tests
├── docs/ # PRD, MCP overview
├── scripts/ # Validation + cache warming scripts
├── openspec/ # OpenSpec specs (4 capabilities)
│ ├── specs/ # Main specs (synced from archived change)
│ └── changes/archive/ # Archived change proposals
├── Dockerfile # Multi-stage build for Cloud Run
├── cloud-run.yaml # Cloud Run service config
└── package.json运行时： Node.js >= 20
MCP 框架： FastMCP v4
编程语言： TypeScript（严格模式）
校验库： Zod v4
Lint/格式化： Biome
测试： Node.js 内置测试运行器
部署： Docker + GCP Cloud Run
本项目使用 OpenSpec 进行 Spec 驱动开发。四个工具能力在 openspec/specs/ 下定义：
arch-pattern-lookup（7 个需求）
brand-context-lookup（6 个需求）
risk-policy-lookup（4 个需求）
tool-selection-lookup（5 个需求）
使用以下命令验证规格：
openspec validate --specs
openspec doctorMIT
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Shared, permission-aware company context for AI agents, with provenance, approvals and audit.
Your company's brain for AI agents. Cited, permission-aware knowledge across every system.
Curated knowledge API for AI agents - skill packs, semantic search, validated patterns.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/ishfuseini/agentflow-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server