Persistent, searchable memory for AI agents — an MCP server on MongoDB with hybrid Vector Search + full-text retrieval.
6 files for Claude Code: add-mcp-tool, embedding-model-migration, indexed-text-matching, ingestion-pipeline and 2 more — 7,556 tokens loaded in every session.
CLAUDE.md A 7,556 tok .claude/skills/add-mcp-tool/SKILL.md A 77 tok .claude/skills/embedding-model-migration/SKILL.md A 91 tok .claude/skills/indexed-text-matching/SKILL.md A 110 tok .claude/skills/ingestion-pipeline/SKILL.md A 75 tok .claude/skills/vector-index-management/SKILL.md A 88 tok These files are jtv4k/mongodb-memory-mcp's own configuration — they tell Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.