nemo-retriever-mcp
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseNeMo Retriever MCP
NeMo Retriever MCP
Use the NeMo Retriever MCP tools exposed under instead
of the CLI.
retrieverretrieverThe server exposes MCP tools, not MCP resources. Do not call
to discover documents or decide whether Retriever is
available; an empty MCP resource list is expected.
list_mcp_resources使用 下公开的 NeMo Retriever MCP 工具,而不是 CLI。
retrieverretriever该服务器公开的是 MCP 工具,而不是 MCP 资源。不要调用
来发现文档或判断 Retriever 是否可用;MCP 资源列表为空是预期结果。
list_mcp_resourcesWorkflow
工作流
- Use to search documents already available to Retriever.
query - Never ingest documents that are already searchable, including documents in a prebuilt index.
- Use only when required documents are not yet searchable and server-readable paths are explicitly available. Do not assume that paths visible to the agent are also readable by the Retriever server.
ingest_documents - After ingestion completes, use to retrieve evidence from the added documents.
query - Start with and
top_k=5. Request more only when the first successful response does not contain enough evidence to answer the task.format="hits" - Read evidence hits from the tool response and ground your answer in them.
- Preserve available document identifiers such as ,
document_id, orsource_id, along with page numbers, when the task requests citations or structured output.pdf_basename - If fails, inspect the error and retry with corrected parameters or a shorter equivalent query. If it still fails, report the error before using another retrieval method.
query
- 使用 搜索 Retriever 中已经可用的文档。
query - 切勿摄取已经可搜索的文档,包括预构建索引中的文档。
- 仅当所需文档尚不可搜索,且明确提供了服务器可读取的路径时,才使用 。不要假定 Agent 可见的路径也可由 Retriever 服务器读取。
ingest_documents - 摄取完成后,使用 从新增的文档中检索证据。
query - 初始设置使用 和
top_k=5。仅当首次成功响应未包含足够证据来回答任务时,才请求更多结果。format="hits" - 从工具响应中读取证据命中项,并以这些证据为依据回答。
- 当任务要求引用或结构化输出时,保留可用的文档标识符,如 、
document_id或source_id,以及页码。pdf_basename - 如果 失败,请检查错误,并使用修正后的参数或更短的等效查询重试。如果仍然失败,请先报告错误,再使用其他检索方法。
query