Version: 1.0.0 Category: communication Port: 3107 Status: Production Ready

Advanced Search plugin provides semantic search with AI-powered relevance, vector similarity, and faceted filtering for comprehensive content discovery in ɳChat.

• Semantic Search: Natural language queries with AI understanding
• Vector Search: Embedding-based similarity matching
• Full-Text Search: Traditional keyword matching with fuzzy search
• Faceted Filtering: Filter by user, channel, date, file type
• Auto-Suggestions: Real-time query completion
• Search History: Personal search history and saved searches
• Multi-Language: Support for 50+ languages

cd backend
nself plugin install advanced-search

# .env.plugins
SEARCH_ENABLED=true
SEARCH_PORT=3107
SEARCH_ENGINE=meilisearch
SEARCH_VECTOR_ENABLED=true
SEARCH_VECTOR_PROVIDER=qdrant
SEARCH_VECTOR_DIMENSION=1536

GET  /api/search?q=query&filters=...
POST /api/search/semantic
GET  /api/search/suggest?q=partial
POST /api/search/save
GET  /api/search/history

# Basic
hello world

# Exact phrase
"team meeting"

# User filter
from:alice

# Channel filter
in:general

# Date range
after:2026-01-01

# File type
has:image

# Boolean
urgent AND (bug OR issue)

GET /api/search?q=from:alice has:file after:2026-01-01&limit=20

Response:

{
  "results": [
    {
      "id": "msg-123",
      "content": "Check out this file...",
      "user": { "id": "alice", "name": "Alice" },
      "channel": { "id": "general", "name": "General" },
      "timestamp": "2026-01-15T10:30:00Z",
      "relevance": 0.95
    }
  ],
  "facets": {
    "users": [{ "id": "alice", "count": 15 }],
    "channels": [{ "id": "general", "count": 20 }]
  },
  "total": 45
}

import { useSearch } from '@/hooks/use-search'

function SearchPage() {
  const { search, results, loading } = useSearch()

  return (
    <SearchBar onSearch={search} />
    <SearchResults results={results} loading={loading} />
  )
}

Full Documentation: See /docs/plugins/ADVANCED-SEARCH-PLUGIN.md