obsidian-mind

What the README says you get

• AI
  Persistent agent memoryAn Obsidian vault that gives AI coding agents persistent memory
• AI
  Content classification hooksClassifies content (decision, incident, win, 1:1, architecture, person, project update)
• AI
  Semantic search via QMDSemantic recall. Find "what did we decide about caching"
• AI
  Local offline modelsQMD runs three small models locally, so there is no API key
• AI
  Wrap-up commandthe agent runs `/om-wrap-up` — verifying notes, updating indexes
• AI
  Pre-compaction backupPreCompact | Before context compaction | Backs
• API
  MCP server toolsRegistered as a Model Context Protocol server in `.mcp.json`
• Deploy
  ShardMind installerInstall via `shardmind install` or `git clone`
• Other
  Frontmatter and link validationValidates frontmatter + wikilinks, blocks misplaced memory files
• Other
  Vault-first memoryVault-first memory keeps context across sessions and machines
• Other
  Automatic note linkinglinking work notes to people, decisions, and competencies automatically
• Other
  GitHub template optionOr use it as a GitHub template

Extracted from the kit's README by a local AI model (deepseek-flash) on Oct 19, 2026. Every item is checked against an exact README quote; 2 unverifiable items were discarded. A README claim is the maker's description, not something we tested.