Poco MCP Service
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Poco MCP Serviceshow me the daily usage analysis for my company"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Poco MCP Service
独立 MCP 服务模块(目录内自洽),用于承载 Poco 的多类型 MCP 工具。
目录结构
app/core:配置与基础设施app/clients:后端 API 客户端app/schemas:结构化输出模型app/services:工具编排与分析逻辑app/tools:MCP tools(按领域拆分)app/main.py:MCP 入口,挂载多子服务skills:面向 AI 的调用与分析技能文档
Related MCP server: hiveconsciousness
快速启动
cd mcp_service
uv sync
uv run python -m app.main默认使用 http 传输(0.0.0.0:8090)。如需 stdio:
MCP_TRANSPORT=stdio uv run python -m app.main认证建议:
推荐通过 MCP headers 透传用户自己的 API Key(可同时传 admin/global)。
BACKEND_ADMIN_API_KEY作为无 HTTP 请求上下文时的可选回退(例如本地直连测试)。可选配置
BACKEND_GLOBAL_API_KEY(global-scope 工具或回退场景使用)。
可用工具(当前)
analysis_get_company_daily_usageanalysis_get_user_daily_usageanalysis_get_usage_scenariosanalysis_get_daily_analysis_briefanalysis_get_daily_content_taxonomyanalysis_get_daily_user_proficiencyanalysis_get_daily_user_personasanalysis_get_daily_report_datasetmanagement_pingmanagement_describe_service
MCP 服务配置示例
{
"mcpServers": {
"poco_analytics": {
"type": "http",
"url": "${MCP_ANALYTICS_HTTP_URL:-http://host.docker.internal:8090/mcp}",
"headers": {
"BACKEND_ADMIN_API_KEY": "${MCP_ANALYTICS_ADMIN_API_KEY}",
"BACKEND_GLOBAL_API_KEY": "${MCP_ANALYTICS_GLOBAL_API_KEY:-}"
}
}
}
}说明:
"/workspace/mcp_service"是 容器内路径,仅在stdio + command启动模式需要。当前推荐
http方式接入,不再需要在 MCP Server 配置里写command/args。MCP_ANALYTICS_ADMIN_API_KEY/MCP_ANALYTICS_GLOBAL_API_KEY建议配置为用户级环境变量。mcp_service会优先读取来路 headers 中的BACKEND_ADMIN_API_KEY、BACKEND_GLOBAL_API_KEY;同时兼容Authorization。仅在无 HTTP 请求上下文时,才回退
BACKEND_ADMIN_API_KEY/BACKEND_GLOBAL_API_KEY(便于本地直连测试)。
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/RipperTs/poco-service-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server