skills.rest
xrcodexcode
xrcodexcodeCOMMUNITY
Explore all skills by this author →
GitHub

Local Knowledge Search

Searches Markdown notes via MOC navigation, grep patterns, tags, and Obsidian backlinks.

3|4|Updated Jul 20, 2026
Search Tags:#markdown#rag#pkm#obsidian#grep#zettelkasten#knowledge search
One-click install
npx skills add https://github.com/xrcodexcode/xr_node --skill local-knowledge-search-xrcodexcode
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Local Knowledge Search
Source: https://github.com/xrcodexcode/xr_node/tree/main/claude/skills/local-rag
Command: npx skills add https://github.com/xrcodexcode/xr_node --skill local-knowledge-search-xrcodexcode

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Retrieving answers from a large personal knowledge base usually requires heavy local vector databases and embedding models that consume RAM. This Skill retrieves knowledge directly from NexusDB Markdown files with zero background memory overhead, ideal for low-spec machines. ## Core Features & Use Cases - MOC-Guided Navigation: Starts retrieval from curated Maps of Content in 03_MOC/ to find relevant entry points quickly. - Direct File Search: Uses grep pattern matching across NODES/, NOTES/, and 02_NEW-KNOWLEDGE/ for exact terms and concepts. - Link & Tag Traversal: Follows Obsidian [[WikiLinks]] backlinks and filters by the controlled tag schema to gather related context. - Use Case: Ask a question about a studied concept and receive a grounded answer citing exact Markdown files, without running ChromaDB or Ollama. ## Quick Start Ask a question about any topic in your NexusDB knowledge base and request a cited answer from your local notes.

Frequently Asked Questions about Local Knowledge Search

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

FAQPage Schema
How do I search my Obsidian notes without a vector database?▼

Use direct file-based retrieval: consult Maps of Content for entry points, run grep pattern searches across note folders, follow [[WikiLinks]] backlinks, and filter by frontmatter tags. This returns cited answers with zero embedding infrastructure.

How to do RAG on a low-RAM laptop without ChromaDB or Ollama?▼

Replace vector retrieval with filesystem search over Markdown notes. MOC navigation, grep matching, and backlink traversal provide context retrieval with no background RAM usage, suitable for 4 GB RAM machines.

Can file-based search replace vector RAG for personal knowledge bases?▼

For curated atomic notes with MOCs, tags, and dense wikilinks, direct search often suffices and keeps notes portable. Vector RAG becomes advantageous at larger scale or for fuzzy semantic queries, and the notes remain ready for ChromaDB later.

What are the limitations of grep-based note search?▼

Grep matches exact strings or patterns, so it misses semantically related content phrased differently. Mitigate this by following backlinks, checking MOCs, and searching multiple synonyms for the target concept.

How does the skill cite sources in its answers?▼

Answers cite exact Markdown files using Obsidian-style references such as [[NODES/Atomic-Concept.md]]. If the requested information is absent from the local notes, the skill states that the knowledge is missing rather than guessing.

Related Skills

Explore More Education & Research Skills→
shiwenwenshiwenwen
1.6k

ha-knowledge

Organizes Markdown notes into linked knowledge bases using note_* tools.

Community
Intermediate
innV0innV0

Researcher

Gather and synthesize information from multiple sources into structured summaries with linked sources.

Official
Advanced
agenticnotetakingagenticnotetaking
3.5k

graph

Analyze Markdown knowledge graph structure and health metrics.

Official
Intermediate
Rawgrowth-ConsultingRawgrowth-Consulting

chroma

Store embeddings with metadata in a local Chroma vector database for similarity search.

Official
Intermediate