Menu
Seeds a mahresources instance with semantic content across all entity types, creating enough items to trigger pagination (>50 per type)
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Operate a mahresources server from the command line with the `mr` CLI — list, create, edit, tag, and link Resources, Notes, and Groups; run MRQL queries and full-text search; manage download jobs, resource versions, series, note blocks, note sharing, taxonomy (tags/categories/note types/relations), plugins, group export/import, audit logs, and admin settings. Use whenever a task involves the `mr` binary or driving a mahresources instance over its HTTP API from a shell.
Design Mahresources Categories, Resource Categories, and Note Types from a data model. Use when asked to create or update MetaSchema JSON Schema, SectionConfig, or CustomHeader/CustomSidebar/CustomSummary/CustomAvatar/CustomMRQLResult layouts, or when you need to understand built-in shortcodes, built-in plugin shortcodes, category template slots, or x-display behavior.
Retake all docs-site screenshots with richly seeded data. Builds the app, starts an ephemeral server, seeds realistic data, captures 26 screenshots via Playwright, and updates the manifest.
| name | seed-data |
| description | Seeds a mahresources instance with semantic content across all entity types, creating enough items to trigger pagination (>50 per type) |
| disable-model-invocation | true |
| argument-hint | ["base-url"] |
| allowed-tools | Bash, Read |
Seeds a mahresources instance with rich semantic content covering all entity types. Creates 60+ items of each type to ensure pagination is triggered.
/seed-data http://localhost:8181
If no URL is provided, defaults to http://localhost:8181.
When this skill is invoked, run the seed script to populate the mahresources instance:
.claude/skills/seed-data/seed.sh $ARGUMENTS
If $ARGUMENTS is empty, the script defaults to http://localhost:8181.
The script will:
Despite these limitations, the script successfully creates pagination-triggering content for:
Create entities in this order to satisfy foreign key relationships:
Use these semantic themes to create meaningful, interconnected data:
Base path: /v1/
Use curl or equivalent to POST form data:
# Tag
curl -X POST "$BASE_URL/v1/tag" -d "name=TagName&Description=Tag description"
# Category
curl -X POST "$BASE_URL/v1/category" -d "name=CategoryName&Description=Description"
# NoteType
curl -X POST "$BASE_URL/v1/note/noteType" -d "name=NoteTypeName&Description=Description"
# RelationType
curl -X POST "$BASE_URL/v1/relationType" -d "name=RelationName&Description=Description&fromCategory=1&toCategory=2"
# Group (with category and tags)
curl -X POST "$BASE_URL/v1/group" -d "name=GroupName&Description=Description&categoryId=1&tags=1&tags=2"
# Note (with owner, type, tags)
curl -X POST "$BASE_URL/v1/note" -d "Name=NoteName&Description=Content&ownerId=1¬eTypeId=1&tags=1"
# Resource from remote URL
curl -X POST "$BASE_URL/v1/resource/remote" -d "url=https://example.com/image.jpg&name=ResourceName&ownerId=1&tags=1"
# Relation
curl -X POST "$BASE_URL/v1/relation" -H "Content-Type: application/json" -d '{"fromGroupId":1,"toGroupId":2,"groupRelationTypeId":1}'
# Query
curl -X POST "$BASE_URL/v1/query" -d "name=QueryName&Text=SELECT * FROM resources LIMIT 10"
Create at least these quantities to trigger pagination (threshold is 50):
| Entity Type | Minimum Count |
|---|---|
| Tags | 60 |
| Categories | 60 |
| NoteTypes | 60 |
| RelationTypes | 60 |
| Groups | 60 |
| Notes | 60 |
| Resources | 60 |
| Relations | 60 |
| Queries | 60 |
Create tags across these categories:
Create categories for:
Create note types for:
Create relation types like:
Create hierarchical groups using the categories above, with meaningful names like:
Create notes with realistic content:
Create resources using public domain images and content:
https://picsum.photos/seed/{name}/800/600https://httpbin.org/image/pnghttps://via.placeholder.com/800x600.png?text=ResourceNameCreate useful saved queries:
SELECT * FROM resources ORDER BY created_at DESC LIMIT 20SELECT * FROM groups WHERE id NOT IN (SELECT group_id FROM group_tags)SELECT note_types.name, COUNT(*) FROM notes JOIN note_types ON notes.note_type_id = note_types.id GROUP BY note_types.idAfter seeding, verify pagination by checking:
curl "$BASE_URL/v1/tags?page=1" | jq '.length' # Should be 50
curl "$BASE_URL/v1/tags?page=2" | jq '.length' # Should be 10+