memory-bridge

Determine which AI tool produced or updated each Obsidian wiki page.

4|Updated May 9, 2026
One-click install
npx skills add https://github.com/itsmarsss/memento --skill memory-bridge
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-bridge
Source: https://github.com/itsmarsss/memento/tree/main/obsidian-wiki/.cursor/skills/memory-bridge
Command: npx skills add https://github.com/itsmarsss/memento --skill memory-bridge

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you understand how different AI tools contributed to your Obsidian wiki by exposing the provenance of each page (which tool originally produced or touched it), so you can quickly spot knowledge gaps and blind spots between tools.

Core Features & Use Cases

  • Cross-tool provenance browsing: View wiki pages grouped by the tool that sourced or updated them, using .manifest.json plus page frontmatter.
  • Targeted searching by tool: Search for a topic within pages sourced from a specific tool without scanning the whole vault.
  • Diff-based insight for gaps: Compare tool silos to find what’s unique to each tool and what’s shared, highlighting surprising asymmetries.
  • Origin matrix mapping: Generate a compact table showing which tools touched each page, capped for readability.

Quick Start

Run the memory-bridge skill with the command "/memory-bridge diff claude codex" to see what knowledge is unique to Claude versus Codex in your wiki.

Frequently Asked Questions about memory-bridge

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

FAQPage Schema
How do I track which AI tool generated specific pages in an Obsidian wiki?

You can track Obsidian wiki provenance by parsing .manifest.json files and page frontmatter to map each page to its originating AI tool. This groups pages into per-tool sets for browsing without scanning the entire vault.

Can I search for a topic only within pages created by a specific AI tool?

Yes, you can search topics within tool-scoped page subsets. By filtering source mappings into per-tool sets, you query a specific AI tool's contributed pages directly, avoiding full-page reads across the whole vault.

How do I compare knowledge gaps and blind spots between different AI tools?

You compare AI tool blind spots by diffing per-tool page sets to find unique and shared knowledge. Generating a page×tool origin matrix highlights surprising asymmetries and gaps between tools like Claude and Codex.

Does this provenance discovery method require reading full Obsidian page contents?

No, provenance discovery works without unnecessary full-page reads. It parses source mappings from .manifest.json and frontmatter to present counts and context, computing an origin matrix capped for readability.

What is an origin matrix for AI knowledge mapping in Obsidian?

An origin matrix is a compact table mapping which AI tools touched each Obsidian wiki page. It parses .manifest.json and frontmatter sources to show page provenance across multiple tools, capped for readability.

How do I start a diff analysis between two AI tools in my Obsidian vault?

Start a diff analysis by running a command like "/memory-bridge diff claude codex". This compares tool silos to output unique versus shared knowledge, highlighting gaps between the specified tools.