DEDUPED Hot AI repos. Real use cases. What people build. Get the Monday 5

How to grow a personal knowledge graph and import notes with Learnharness

Let lessons accumulate as propositions and reasoning chains, and import your own notes.

Short answer: Install LearnHarness and add a provider key; then Complete teaching sessions so lessons accumulate as propositions and reasoning chains; open the graph to see the structure with links back to source sessions. All steps.

Who it is forA researcher or lifelong learner who wants learning to carry across sessions.

The problemNotes and chats pile up without showing how ideas connect.

The outcomeA graph of claims and the premises behind them, readable by future teaching sessions.

What you need first

  • macOS (Apple Silicon), Windows (x64) or Linux (x64)
  • An API key for one of the 30+ supported provider options, such as OpenAI, Anthropic, Google, DeepSeek or OpenRouter
  • Optional: a Deepgram API key with the Member role or higher for dictation

Steps

  1. Install LearnHarness and add a provider key
  2. Complete teaching sessions so lessons accumulate as propositions and reasoning chains
  3. Open the graph to see the structure with links back to source sessions
  4. Import your notes into the graph
  5. Start a new session; it can read the graph instead of starting from scratch

Prompt for your agent

In LearnHarness, import my notes on <topic> into the knowledge graph, then start a new session that builds on them.

Caveat. Local storage does not mean offline: model requests, web research and dictation still contact external services.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for amosblomqvist/learnharness and "Grow a personal knowledge graph and import notes"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer