deepset-mcp
OfficialRelated Servers
Alternatives to deepset-mcp
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityCmaintenanceDeploys and serves Haystack Pipelines and Agents as REST APIs and MCP tools, enabling AI assistants to interact with custom NLP pipelines for tasks like RAG, document processing, and intelligent agents through natural language.4,522 PyPI152Apache 2.0
- FlicenseAqualityBmaintenanceEnables Claude to inspect, configure, run, and modify individual RAG/agent pipeline stages (retrieve, rerank, generate, eval) as structured typed tool calls, and execute the full pipeline end to end.5-
- FlicenseAqualityBmaintenanceConnects AI agents to live observability stacks including Sentry, GitHub, Vercel, Better Stack, and Cloudflare, enabling end-to-end incident investigation, deployment correlation, root cause analysis, and regression triage.14-
- AlicenseBqualityDmaintenanceEnables AI agents to interact with Dagster instances, explore data pipelines, monitor runs, and manage assets.924Apache 2.0
- FlicenseNot gradedqualityCmaintenanceEnables document Q&A and knowledge retrieval through hybrid semantic and keyword search, with tools for document ingestion, chunking, summarization, PII redaction, and RAGAS-based evaluation.-
- AlicenseNot gradedqualityBmaintenanceEnables running durable, traceable AI agents via LangGraph through a universal MCP interface, integrating with Hatchet for orchestration, logging, and retries. Provides tools for knowledge management (ingestion, RAG) and Kubernetes operations (diagnosis, auto-fix).MIT
TDQS
Scored across 50 tools
Most tools are clearly separated by resource and action, but the three search_pipeline variants (plain, with_filters, with_params) heavily overlap, and list_pipeline_traces, list_pipeline_search_history, and list_search_history all expose query-run data that could be confused. A few other pairs like get_custom_components vs list_custom_component_installations also require careful reading to distinguish.
The vast majority of tools follow a predictable verb_noun pattern (list_workspaces, create_pipeline, get_index, deploy_pipeline), making navigation easy. The object store utilities break the pattern with yq_object_store, sed_object_store, and grep_object_store using Unix-command-style names, and get_from_object_store/get_slice_from_object_store are slightly awkward prepositional forms, but these are a small minority.
At 50 tools, this server is well above the 25+ threshold that indicates an overly large surface. Many tools could be consolidated, such as the three search_pipeline variants and the multiple trace/history retrieval tools, making the set feel more like a raw API dump than a curated MCP tool set.
The server covers the core lifecycle for pipelines, indexes, and workspaces with create, get, list, update, validate, and deploy operations, plus solid tracing, logging, and search history support. However, there are no delete operations for any resource, no way to create or update secrets, and no tool to install or delete custom components, leaving notable dead ends in the workflow.