#knowledge-base

Use when the user wants meeting, conversation or lecture recordings in a personal knowledge base: sync automatically from Feishu, or process transcripts exported from tools like Zoom, Google Meet or Otter. Also for capturing ideas, saving articles, and looking up past conversations about a person or project. Originals are never edited, and new people are confirmed before they get a profile. CC BY-NC 4.0, non-commercial use only. Use case: In one line: your AI agent saves each meeting recording as an original, writes the notes, identifies people and to-dos, and can answer questions about them later

You open the recording during a meeting, and later you can ask AI what you discussed this week, with whom, and which items still need follow-up

A beginner tutorial: first have the Agent run a read-only health check on your notes vault, then write down folder conventions and guardrails, install kepano's obsidian-skills and an open-source digital garden Skill set, and day to day only ask it to find related notes and orphan notes. After a month, run the check again and compare, then handle the batch of empty shell notes yourself. Includes 3 copy-ready prompts.

A beginner tutorial: Obsidian is one of the better choices for a personal knowledge base today. Four steps to connect an agent such as Claude Code to your notes: install and create a vault, enable the official command line, install the Obsidian Skill, then have it check your notes before answering and save new conclusions back. Includes three ready-to-copy prompts.

Use when the user's question may be answered in their Obsidian notes, or when conclusions or materials should be saved to the knowledge base or the knowledge base needs organizing. Uses the official Obsidian command-line tool obsidian to search, read, append to, and link notes: before answering, search and cite note names; before writing, check for duplicates; only append, never overwrite; ask the user before deleting, renaming, or moving. Use case: Turn an Obsidian vault into a long-term knowledge base that an Agent can search and write to

Beginner tutorial: Regular RAG 'retrieves once, then answers', while Agentic RAG 'checks its own results after retrieving, and if they fall short, rewrites the query and searches again'. Covers the loop, suitable scenarios, and things to watch out for.