research-topic

Chain vault_search, vault_context, vault_graph_neighborhood to synthesize insights from Obsidian notes.

Updated Mar 29, 2026
One-click install
npx skills add https://github.com/artislismanis/obsidian-agent-sandbox --skill research-topic-artislismanis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research-topic
Source: https://github.com/artislismanis/obsidian-agent-sandbox/tree/main/workspace/.claude/skills/research-topic
Command: npx skills add https://github.com/artislismanis/obsidian-agent-sandbox --skill research-topic-artislismanis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Synthesize vault insights by chaining MCP tools instead of reading files directly, enabling structured results that leverage Obsidian's metadata cache.

Core Features & Use Cases

  • Use vault_search to identify candidate notes related to a topic, returning a snippet per match.
  • Use vault_context on the top results to retrieve content, frontmatter, tags, headings, links, and backlinks in a single call.
  • Use vault_graph_neighborhood to explore adjacent notes at depth 1–2 and surface related context for deeper synthesis.
  • Example use case: ask "what do my notes say about X?" and synthesize a cohesive summary with cited paths.

Quick Start

Ask it to summarize what your notes say about a topic and review the synthesized insights.

Frequently Asked Questions about research-topic

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

FAQPage Schema
How do I summarize what my Obsidian vault notes say about a specific topic?

Vault insights synthesis uses MCP tools to search notes, retrieve context, and explore graph neighborhoods, chaining these operations instead of reading files directly to produce structured summaries with cited paths.

How does MCP discovery work for finding related notes in a vault?

MCP discovery works by using vault_search to find matching snippets, vault_context to pull frontmatter and links, and vault_graph_neighborhood to explore adjacent notes at depth 1–2, connecting related context without direct file reading.

Can I retrieve backlinks and frontmatter from my notes without opening the files directly?

Yes, you can retrieve backlinks and frontmatter without opening files by using the vault_context MCP tool, which returns content, tags, headings, links, and backlinks in a single call, optionally tracking citations with vault_backlinks.

What is the best way to explore connected context across multiple notes in a vault?

The best way to explore connected context is using vault_graph_neighborhood to map adjacent notes at depth 1–2, combined with vault_context to retrieve relational metadata, synthesizing broader insights across the vault graph.

Do I need any specific MCP tools installed to synthesize vault insights?

Yes, synthesizing vault insights requires the MCP tools vault_search, vault_context, and vault_graph_neighborhood to locate, contextualize, and connect notes, with optional use of vault_backlinks for citation tracking.