docs: add terminal demo GIF to README

Recorded via vhs against a sample vault. Shows hybrid search with
--explain, person context bundle, and status output.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: devwhodevs

README.md

@@ -8,6 +8,10 @@
 
 engraph turns your markdown vault into a searchable knowledge graph that AI agents can query through [MCP](https://modelcontextprotocol.io). It combines semantic embeddings, full-text search, and wikilink graph traversal into a single local binary. No API keys, no cloud — everything runs on your machine.
 
+<p align="center">
+  <img src="assets/demo.gif" alt="engraph demo: hybrid search with 3-lane RRF, person context bundles" width="800">
+</p>
+
 ## Why engraph?
 
 Plain vector search treats your notes as isolated documents. But knowledge isn't flat — your notes link to each other, share tags, reference the same people and projects. engraph understands these connections.

assets/demo.gif

@@ -0,0 +1,63 @@
+Output assets/demo.gif
+Set Shell zsh
+Set FontSize 15
+Set Width 1100
+Set Height 700
+Set Padding 20
+Set Theme "Catppuccin Mocha"
+Set TypingSpeed 50ms
+Set WindowBar Colorful
+Set WindowBarSize 40
+
+Hide
+Type "export PATH=/Users/devwhodevs/Documents/engraph/target/release:$PATH"
+Enter
+Sleep 500ms
+Show
+
+# Title
+Type "# engraph — local knowledge graph for AI agents"
+Enter
+Sleep 1s
+
+# Index
+Type "engraph index /tmp/engraph-demo-vault"
+Enter
+Sleep 2s
+
+Type ""
+Enter
+Sleep 500ms
+
+# Search
+Type "# Hybrid search: semantic + keyword + graph expansion"
+Enter
+Sleep 800ms
+
+Type "engraph search 'how does authentication work' --explain -n 5"
+Enter
+Sleep 4s
+
+Type ""
+Enter
+Sleep 500ms
+
+# Person context
+Type "# Person context bundle — note + mentions + connections"
+Enter
+Sleep 800ms
+
+Type "engraph context who 'Sarah Chen'"
+Enter
+Sleep 3s
+
+Type ""
+Enter
+Sleep 500ms
+
+# Status
+Type "engraph status"
+Enter
+Sleep 2s
+
+Sleep 3s