Ingest local files, git histories, and web crawls into the gnosis-mcp knowledge base.
npx skills add https://github.com/nicholasglazer/gnosis-mcp --skill ingest-nicholasglazer
Please help me install this Agent Skill. Skill: ingest Source: https://github.com/nicholasglazer/gnosis-mcp/tree/main/skills/ingest Command: npx skills add https://github.com/nicholasglazer/gnosis-mcp --skill ingest-nicholasglazer
Ingests and indexes docs into the gnosis-mcp knowledge base.
Ingest the contents of the docs folder to populate and update the knowledge base.
High-intent search queries and answers about installing and using this skill.
To index local documentation files, you ingest the contents of your docs folder to populate and update the knowledge base. This builds a searchable index directly from your local files.
Yes, you can ingest git histories and web crawl results alongside local files. The ingestion process builds a searchable index combining all three source types into one knowledge base.
The best way to keep an indexed knowledge base accurate is using incremental updates and pruning of missing sources. This maintains index accuracy by removing outdated entries during reingest cycles.
Yes, the ingestion process supports an optional embedding workflow for richer search capabilities. You can enable embeddings using the --embed flag during the ingestion lifecycle.
Available flags to control the reingest lifecycle include --force, --prune, --wipe, --embed, and --include-crawled. These single-entry point flags manage index updates and safe pruning options.
Use the --prune flag to safely remove missing sources during incremental updates, or use --wipe to clear the index entirely. These options maintain index accuracy when original documentation sources are deleted or moved.
arbisoftCentralize knowledge management across local files, MCP memory, vector stores, and Git repositories.
nicholasglazerSearches indexed gnosis-mcp content with keyword, hybrid semantic+keyword, and git history modes.
steipeteIndex local files and search them with BM25, vector, and hybrid queries.
1thirteeng3Extract text from PDFs and documents using OCR or text-based parsers.
