Logo
Cockpit
Content Platform
Get Cockpit
pro

Detektivo

Detektivo is a powerful full-text search and indexing addon for Cockpit CMS. It provides lightning-fast search capabilities with support for fuzzy matching, faceted filtering, and multiple storage backends.

Overview

Detektivo streamlines content search without requiring third-party services like Algolia or Elasticsearch for small to mid-size projects. For larger deployments, it seamlessly integrates with Meilisearch for enhanced performance.

Screenshot of an index

Features

  • Full-Text Search: Search across all indexed content
  • Fuzzy Matching: Find results even with typos
  • Faceted Filtering: Filter results by categories, tags, etc.
  • Real-Time Indexing: Content updates are immediately searchable
  • Multi-Language Support: Search works across all locales
  • Two Storage Backends: SQLite FTS5 or Meilisearch

Storage Backends

SQLite FTS5 (Default)

The default backend uses SQLite's Full-Text Search 5 extension. No additional configuration is needed - it works out of the box.

Best for:

  • Small to mid-size projects
  • Simple deployment requirements
  • Projects without external dependencies

Meilisearch

Meilisearch is an open-source search engine that provides advanced features and better performance for larger datasets.

Best for:

  • Large content repositories
  • High-traffic applications
  • Advanced search requirements (typo tolerance, synonyms, etc.)

Configuration

SQLite (Default)

No configuration required. Detektivo automatically uses SQLite FTS5.

Meilisearch

Add to /config/config.php:

<?php

return [
    'search' => [
        'server' => 'meilisearch://localhost:7700',
        'options' => [
            'api_key' => getenv('MEILISEARCH_KEY')
        ]
    ]
];

Set in your .env file:

MEILISEARCH_KEY=your-master-key

Meilisearch with Custom URL

'search' => [
    'server' => 'meilisearch://search.yourdomain.com',
    'options' => [
        'api_key' => getenv('MEILISEARCH_KEY')
    ]
]

Creating Indexes

Via Admin Interface

  1. Navigate to Detektivo in the admin menu
  2. Click Add Index
  3. Configure the index:
Field Description
Name Unique identifier for the index
Label Display name
Source Content model to index (Collection, Singleton, Tree)
Fields Fields to include in the index
Filters Optional conditions for which items to index
  1. Click Save to create the index
  2. Click Rebuild to populate the index with existing content

Index Configuration

When creating an index, you specify:

  • Source Model: Which collection, singleton, or tree to index
  • Indexed Fields: Which fields should be searchable
  • Filter Criteria: Optional filters to include only certain items

Managing Indexes

Rebuilding Indexes

If content was added before the index existed, or if you change index configuration:

  1. Open the index in Detektivo
  2. Click Rebuild Index

This reprocesses all matching content and updates the search index.

Automatic Updates

Once an index is created, Detektivo automatically:

  • Adds new content when created
  • Updates index entries when content changes
  • Removes entries when content is deleted

Search Features

Search across all indexed fields:

laptop computer

Search for exact phrases:

"gaming laptop"

Fuzzy Search (Meilisearch)

Meilisearch automatically handles typos:

laptpo  → finds "laptop"

Filtering (Meilisearch)

Combine search with filters:

?q=laptop&filter=category:electronics

Best Practices

Index Design

  • Be Selective: Only index fields that need to be searchable
  • Group Related Content: Consider creating multiple indexes for different search contexts
  • Use Filters: Limit indexes to relevant content (e.g., published items only)

Performance

  • SQLite: Works well for up to ~50,000 documents
  • Meilisearch: Recommended for larger datasets or heavy search traffic

Content Considerations

  • Index text content, not complex nested structures
  • Consider which locales need to be searchable
  • Use meaningful field names for faceted navigation

Troubleshooting

Search Returns No Results

  • Verify the index has been rebuilt after content changes
  • Check that queried terms exist in indexed fields
  • Ensure content matches any index filters

Meilisearch Connection Issues

  • Verify server URL is correct
  • Check API key permissions
  • Ensure Meilisearch is running and accessible

Index Not Updating

  • Check Cockpit logs for indexing errors
  • Verify content model hasn't changed
  • Try rebuilding the index manually