One-click install
npx skills add https://github.com/AdamTylerLynch/obsidian-agent-memory-skills --skill obs-memory-adamtylerlynch
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: obs-memory
Source: https://github.com/AdamTylerLynch/obsidian-agent-memory-skills/tree/main/skills/obs
Command: npx skills add https://github.com/AdamTylerLynch/obsidian-agent-memory-skills --skill obs-memory-adamtylerlynch

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coding agents lose all session context, pending TODOs, and project architecture knowledge when a session ends, forcing you to re-explain past work, design decisions, and component relationships every time you start a new session.

Core Features & Use Cases

  • Automatic session orientation: Reads pending TODOs and project overviews at session start without being asked, so the agent immediately understands current work priorities.
  • Graph-based vault navigation: Uses the Obsidian CLI to look up component dependencies, architecture decisions, domain knowledge, and related notes on demand, minimizing context token usage.
  • Bidirectional relationship management: Maps dependencies, extensions, and implementation relationships between code components with BFS tree walking for full dependency visibility.
  • Session and project management: Tools to write session summaries, manage TODOs, scaffold new projects, and hydrate the vault with project metadata from existing repository documentation. Use case: For a multi-session software project, the agent automatically picks up where you left off, remembers past architecture decisions, and logs new discoveries to the vault for future reference.

Quick Start

Use the obs-memory skill to initialize your Obsidian memory vault and enable persistent cross-session context for your coding agent.

Frequently Asked Questions about obs-memory

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

FAQPage Schema
How do I give a coding agent persistent memory across sessions?

A coding agent loses session context, pending work, and project architecture knowledge when a session ends. You can solve this by using a local Obsidian knowledge vault to store session summaries and project metadata, allowing the agent to automatically pick up where you left off.

How do coding agents remember past architecture decisions and component relationships?

Coding agents remember architecture decisions and component relationships by querying a local Obsidian knowledge vault. The vault stores graph-structured memory and maps bidirectional dependencies between code components, minimizing context token usage during lookups.

How do I automatically orient a coding agent at the start of a new session?

You can orient a coding agent automatically at session start by using a skill that reads pending TODOs and project overviews from an Obsidian vault. This provides the agent with immediate context on current work priorities without being explicitly asked.

Can I use Obsidian to manage project architecture knowledge for AI coding assistants?

Yes, you can use Obsidian to manage project architecture knowledge for AI coding assistants. The vault stores component notes and uses bidirectional relationship management with BFS tree walking to map dependencies, extensions, and implementation relationships between code components.

How do I hydrate a knowledge vault with existing repository documentation for an agent?

You can hydrate a knowledge vault with existing repository documentation by running repository analysis tools. This scaffolds new projects and extracts project metadata to automatically generate component notes and map architecture within the vault.

Does obs-memory work with any coding agent or only specific platforms?

The obs-memory skill applies to all coding workflows using agents compatible with the Agent Skills specification. It leverages a local Obsidian knowledge vault and uses CLI-first querying to deliver persistent, cross-session context to the compatible agent.