onote

Summarize Claude sessions into structured Obsidian-ready markdown notes with YAML frontmatter.

Updated Jun 4, 2026
One-click install
npx skills add https://github.com/PeteRichardson/skills --skill onote-peterichardson
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: onote
Source: https://github.com/PeteRichardson/skills/tree/main/onote
Command: npx skills add https://github.com/PeteRichardson/skills --skill onote-peterichardson

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Learning sessions with an AI assistant produce valuable insights, decisions, and code artifacts that are easily lost once the conversation ends. This Skill converts a conversation into a permanent, structured Obsidian note so learnings, trade-offs, and open questions are preserved for future recall. ## Core Features & Use Cases - Session Summarization: Analyzes the full conversation and extracts 3-5 significant learnings using explicit significance criteria (new concepts, changed approaches, solved blockers, reusable tools, mental model shifts). - Structured Output Template: Generates a markdown file with YAML frontmatter (tags, date, related links, status, confidence), key learnings with code artifacts, a decisions-made table, remaining questions, and next-step actions. - Confidence Tagging: Labels each learning as Confirmed, Provisional, or Hypothesis so future readers know what to trust. - Use Case: After a long debugging session on Rust async patterns, invoke the skill to produce a note saved directly to your Obsidian vault, complete with code snippets, a decision table comparing Tokio vs async-std, and follow-up tasks. ## Quick Start Ask the assistant to create an Obsidian note from this session, optionally scoping it with a focus topic such as the Rust patterns discussed.

Frequently Asked Questions about onote

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I turn a Claude conversation into an Obsidian note?

Invoke the skill with /onote or ask to create an Obsidian note from the session. It analyzes the conversation, extracts 3-5 significant learnings, and saves a structured markdown file directly to your Obsidian vault directory.

How do I scope a session summary to one topic?

Add a focus modifier to the invocation, such as "/onote focus on the Rust async patterns". You can also request separate linked notes for multiple domains, limit the learning count, or ask for a minimal condensed version.

When should I not use a session summarization skill?

Skip it for quick lookups like a single syntax question, since it is designed for deeper learning sessions with multiple substantive exchanges. Sessions covering many unrelated topics should be split into separate linked notes instead.

What does the generated Obsidian note contain?

Each note includes YAML frontmatter with tags, date, related links, status, and confidence, plus sections for context, key learnings with code artifacts and references, a decisions-made table, remaining questions, and next-step actions.

How are confidence levels used in generated notes?

Each learning is tagged Confirmed, Provisional, or Hypothesis based on whether it was tested, verified once, or remains a theory. This lets you revisit notes later and promote provisional items to confirmed after real-world validation.