micro-app-mcp
by didengren
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DATA_DIR | No | 数据存储路径 | |
| HF_ENDPOINT | No | HuggingFace 镜像地址，如 https://hf-mirror.com | |
| GITHUB_TOKEN | No | GitHub API Token，强烈建议配置（否则更新知识库时易触发限流） | |
| EMBEDDING_MODEL | No | 向量化模型类型，目前只支持 local | local |
| FALLBACK_DATA_DIR | No | 当 DATA_DIR 不可写时自动回退目录 | /tmp/micro_app_mcp |
| GITHUB_RETRY_TOTAL | No | GitHub 请求重试次数 | 0 |
| UV_EXTRA_INDEX_URL | No | PyTorch 官方 CPU 仓库地址，Linux/Windows 用户可配置为 https://download.pytorch.org/whl/cpu | |
| EMBEDDING_LAZY_LOAD | No | 是否启用懒加载 | true |
| CACHE_DURATION_HOURS | No | 智能缓存配置（小时） | 24 |
| EMBEDDING_MODEL_NAME | No | 本地向量化模型名称 | BAAI/bge-small-zh-v1.5 |
| SEARCH_TIMEOUT_SECONDS | No | 检索超时秒数 | 30 |
| CHROMA_ANONYMIZED_TELEMETRY | No | Chroma 匿名遥测开关 | false |
| GITHUB_HTTP_TIMEOUT_SECONDS | No | GitHub 请求超时时间（秒） | 15 |
| UPDATE_MAX_DURATION_SECONDS | No | 单次更新最大持续时间（秒） | 3600 |
| UPDATE_INTENT_ACTION_KEYWORDS | No | 更新动作关键词（逗号分隔） | 强制更新,更新知识库,同步知识库,重建索引,force update,update knowledge base,rebuild index,sync knowledge base |
| UPDATE_INTENT_TARGET_KEYWORDS | No | 更新目标关键词（逗号分隔） | 知识库,索引,向量库,knowledge base,index,vector |
| UPDATE_INTENT_SEARCH_ONLY_PATTERNS | No | 检索优先短语（命中后不触发更新） | 更新日志,changelog,release note,release notes,版本更新,最新更新 |
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
No tools | |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
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/didengren/micro-app-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server