Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From semantic-search
Indexes a folder of documents for semantic search: scans supported files, extracts text, stores embeddings in local vector database. Reports files indexed, skipped, and errors.
npx claudepluginhub zhubit/cowork-semantic-searchHow this command is triggered — by the user, by Claude, or both
Slash command
/semantic-search:index <folder path>The summary Claude sees in its command listing — used to decide when to auto-load this command
Index the specified folder for semantic search. This scans all supported document files, extracts text, and stores embeddings in a local vector database. Call the `index_folder` tool with the provided folder path. Report the results including how many files were indexed, skipped, and any errors encountered.
/agent-brain-indexIndexes documents and optionally code from a path for semantic and keyword search, creating vector embeddings and BM25 index. Supports incremental updates.
/index-docsIndexes documents from directories, URLs, databases, or APIs into vector stores like Pinecone, Weaviate, Chroma, or pgvector using semantic chunking, embeddings, and metadata for RAG.
/pdf-researchIndexes PDFs using LightRAG for semantic search in hybrid/local/global/naive modes, checks index status, and configures directories.
/indexGenerates project documentation, knowledge base, API docs, structure analysis, or README for a target path. Supports --type (docs|api|structure|readme) and --format (md|json|yaml).
/indexIndexes codebase into Supermemory for persistent context: detects tech ecosystem, scans manifests/configs/entry points/architecture.
/indexIndexes codebase components and external resources into searchable specs in `specs/.index/`. Supports --path, --type, --exclude, --dry-run, --force, --changed, --quick flags.
Share bugs, ideas, or general feedback.
Index the specified folder for semantic search. This scans all supported document files, extracts text, and stores embeddings in a local vector database.
Call the index_folder tool with the provided folder path. Report the results including
how many files were indexed, skipped, and any errors encountered.