Searches machine manuals, bearing catalogs, and technical docs using RAG (FAISS + TF-IDF) retrieval via predictive-maintenance-mcp server. Extracts specs, bearing geometry, and fault frequencies.
How this skill is triggered — by the user, by Claude, or both
Slash command
/predictive-maintenance:documentation-searchThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Search and extract information from machine manuals, bearing catalogs, and
Search and extract information from machine manuals, bearing catalogs, and technical datasheets using RAG-based retrieval (FAISS + TF-IDF) and structured document parsing.
Prerequisite: The predictive-maintenance-mcp MCP server must be connected.
Call list_machine_manuals() to see all available manuals and catalogs in the
resources/ directory (PDF, TXT formats).
Call search_documentation(query=..., top_k=5) to perform semantic search
across all indexed documents.
The RAG engine uses FAISS vector similarity + TF-IDF keyword matching to find the most relevant document passages.
Call read_manual_excerpt(manual_name=..., start_page=..., end_page=...) to
read specific pages from a manual.
Useful when you know which manual contains the information but need to read a specific section.
Call extract_manual_specs(manual_name=...) to automatically extract structured
data from a manual:
Call search_bearing_catalog(query=...) to search the bearing database.
Call lookup_bearing_and_compute_tool(bearing_query=..., shaft_rpm=...) to
search the catalog AND compute characteristic fault frequencies (BPFO, BPFI,
BSF, FTF) in one step.
list_machine_manuals() to find the relevant manualextract_manual_specs(manual_name=...) to pull bearing designationssearch_bearing_catalog(query=...) with the found designationsearch_documentation(query="vibration limits {machine_name}")read_manual_excerpt(...) on the specific manuallookup_bearing_and_compute_tool(bearing_query="6205", shaft_rpm=1800)npx claudepluginhub lgdimaggio/predictive-maintenance-mcp --plugin predictive-maintenanceSearches indexed embedded systems documentation for registers, memory maps, and peripheral details using search_docs tool.
Builds RAG pipelines using LangChain: document loading, recursive text splitting, OpenAI embeddings, and vector stores (Chroma, FAISS, Pinecone).
Searches documents, codebases, and knowledge bases using BM25 keyword, semantic vector, hybrid, graph, and multi retrieval modes for dependencies, relationships, and references.