Salesforce-to-Glider MCP Server
Provides governed MCP tools for a controlled, AI-assisted asset-change pipeline, featuring natural-language requests, mandatory approvals, execution-time revalidation, and complete audit.
README
Salesforce-to-Glider Nexus Prototype
A contract-based mock prototype demonstrating a controlled, AI-assisted asset-change pipeline:
Natural-language request
-> controlled asset discovery
-> user review
-> immutable request manifest
-> deterministic manager routing
-> mandatory approval
-> execution-time revalidation
-> mock authoritative update
-> complete reconciliation and audit
This prototype uses assumed contracts and synthetic data only. It does
not represent a working integration with real Salesforce or Glider Nexus
systems. See docs/scope.md and
docs/assumptions.md before relying on anything
here.
Current status: Phase 0 (skeleton)
Only the repository skeleton, shared contracts, and service health checks
exist so far. See docs/scope.md for what's in and out of scope at this
phase, and the build-phase plan in the source spec for what comes next.
Local setup
python3.11 -m venv .venv
source .venv/bin/activate
pip install -e ".[dev]"
cp .env.example .env
Running tests
pytest
Running a service locally (Phase 0: health checks only)
uvicorn mock_glider.app.main:app --reload --port 8001
uvicorn mock_salesforce.app.main:app --reload --port 8002
uvicorn integration.app.main:app --reload --port 8003
Then check, e.g.:
curl http://localhost:8001/health
Docker
Docker Compose orchestration is added once individual services have real functionality (from Phase 1 onward). Not yet available in Phase 0.
Repository layout
See docs/scope.md for the full phase plan. Top-level directories:
shared/- contracts (enums, schemas, errors, correlation) used by all servicesmock_glider/- mock authoritative asset systemmock_salesforce/- mock request/approval systemintegration/- deterministic workflow, validation, approval routing, executionassistant/- conversational parser and optional LLM adapterfrontend/- Operations/manager/status UImcp_server/- governed MCP tools (optional, added last)config/- approval rules, responsibility mappings, status transitionsscripts/- seed data, demo, reset, performance test scriptsdocs/- scope, assumptions, architecture, contracts, test plan, demo script
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。