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

Seed Data Skill

Seeds a mahresources instance with rich semantic content covering all entity types. Creates 60+ items of each type to ensure pagination is triggered.

Usage

/seed-data http://localhost:8181

If no URL is provided, defaults to http://localhost:8181.

Instructions

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:

Check server connectivity
Create 60+ entities of each type in the correct dependency order
Verify pagination is working
Print a summary of created entities

Known Limitations

Relation Types: May fail to persist due to an FTS trigger bug in the application. The script reports success but entities may not be saved.
Relations: Require groups with matching categories to the relation type's category constraints. Most will fail due to category mismatches with the default relation types.

Despite these limitations, the script successfully creates pagination-triggering content for:

Tags (78+ items)
Categories (68+ items)
Note Types (63+ items)
Groups (64+ items)
Notes (60+ items)
Resources (60+ items)
Queries (60+ items)

Execution Order (Dependencies)

Create entities in this order to satisfy foreign key relationships:

Tags (no dependencies)
Categories (no dependencies)
NoteTypes (no dependencies)
RelationTypes (depends on Categories)
Groups (depends on Categories, Tags)
Notes (depends on Groups, Tags, NoteTypes)
Resources (depends on Groups, Tags, Notes)
Relations (depends on Groups, RelationTypes)
Queries (no dependencies, but useful after data exists)

Content Themes

Use these semantic themes to create meaningful, interconnected data:

Theme 1: Software Development Company

Categories: Departments, Projects, Employees, Meetings, Documents
Tags: urgent, in-progress, completed, review-needed, approved, archived
Groups: Engineering Team, Design Team, Marketing Team, HR Department, Finance
Notes: Meeting minutes, project specs, code reviews, performance reviews
Resources: Design mockups, architecture diagrams, reports, presentations

Theme 2: Research Library

Categories: Books, Journals, Authors, Publishers, Research Topics
Tags: peer-reviewed, open-access, citation-needed, primary-source, archived
Groups: Computer Science, Biology, Physics, Mathematics, Philosophy
Notes: Research summaries, literature reviews, annotations, bibliographies
Resources: PDFs, images, datasets, supplementary materials

Theme 3: Media Production

Categories: Films, TV Shows, Actors, Directors, Studios
Tags: released, in-production, pre-production, cancelled, award-winning
Groups: Action Movies, Documentaries, Animated Films, TV Dramas, Shorts
Notes: Script notes, production logs, reviews, cast lists
Resources: Posters, trailers, stills, behind-the-scenes

Theme 4: E-commerce Inventory

Categories: Electronics, Clothing, Home & Garden, Sports, Books
Tags: in-stock, out-of-stock, on-sale, new-arrival, bestseller, clearance
Groups: Smartphones, Laptops, Men's Wear, Women's Wear, Kitchen, Outdoor
Notes: Product descriptions, customer reviews, shipping info, returns
Resources: Product images, manuals, size charts, videos

API Endpoints Reference

Base path: /v1/

Creating Entities

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&noteTypeId=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"

Required Counts (for pagination)

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

Semantic Data Lists

Tags (60+ items)

Create tags across these categories:

Status: draft, pending, in-review, approved, published, archived, deprecated, active, inactive, suspended
Priority: critical, high, medium, low, backlog, urgent, important, normal, deferred
Type: feature, bug, enhancement, documentation, research, experiment, prototype, production
Department: engineering, design, marketing, sales, support, hr, finance, legal, operations
Quality: verified, unverified, tested, untested, stable, unstable, beta, alpha, release-candidate
Timeline: q1-2024, q2-2024, q3-2024, q4-2024, planned, in-progress, completed, overdue

Categories (60+ items)

Create categories for:

Organization: department, team, project, initiative, program, portfolio, division, unit
Content: document, report, presentation, spreadsheet, diagram, mockup, prototype, specification
People: employee, contractor, vendor, customer, partner, stakeholder, advisor, board-member
Events: meeting, conference, workshop, training, review, retrospective, planning, demo
Products: software, hardware, service, subscription, license, support-plan, add-on, bundle
Assets: image, video, audio, code, data, model, template, component

NoteTypes (60+ items)

Create note types for:

Documentation: readme, changelog, api-doc, user-guide, tutorial, faq, troubleshooting, reference
Communication: email-summary, meeting-notes, announcement, memo, newsletter, status-update
Analysis: report, review, assessment, evaluation, audit, benchmark, comparison, survey
Planning: roadmap, sprint-plan, milestone, objective, key-result, goal, strategy, tactic
Technical: architecture, design-doc, rfc, adr, postmortem, incident-report, runbook, sop

RelationTypes (60+ items)

Create relation types like:

Hierarchy: parent-of, child-of, contains, part-of, belongs-to, owns, managed-by
Dependencies: depends-on, blocks, required-by, enables, triggers, follows, precedes
Association: related-to, similar-to, alternative-to, replaces, derived-from, based-on
Workflow: assigned-to, reviewed-by, approved-by, created-by, modified-by, archived-by
Business: customer-of, vendor-for, partner-with, competes-with, acquired-by, merged-with

Groups (60+ items)

Create hierarchical groups using the categories above, with meaningful names like:

"Engineering - Backend Team", "Engineering - Frontend Team", "Engineering - DevOps"
"Q1 2024 Roadmap", "Q2 2024 Roadmap", "Product Launch Initiative"
"Customer: Acme Corp", "Customer: TechStart Inc", "Vendor: Cloud Services Ltd"

Notes (60+ items)

Create notes with realistic content:

Meeting notes with dates, attendees, action items
Technical specifications with requirements
Project status updates with metrics
Research findings with citations
Review feedback with recommendations

Resources (60+ items)

Create resources using public domain images and content:

Use URLs from picsum.photos for placeholder images: https://picsum.photos/seed/{name}/800/600
Use httpbin.org for test files: https://httpbin.org/image/png
Use placeholder.com for labeled images: https://via.placeholder.com/800x600.png?text=ResourceName

Queries (60+ items)

Create useful saved queries:

"Recent Resources": SELECT * FROM resources ORDER BY created_at DESC LIMIT 20
"Untagged Groups": SELECT * FROM groups WHERE id NOT IN (SELECT group_id FROM group_tags)
"Notes by Type": SELECT note_types.name, COUNT(*) FROM notes JOIN note_types ON notes.note_type_id = note_types.id GROUP BY note_types.id
Various filtered views and reports

Execution Strategy

Batch creation: Create entities in batches of 10-20 using parallel curl commands
Track IDs: Store created entity IDs for use in relationships
Verify counts: After creation, query each endpoint to verify pagination works
Report results: Output a summary of created entities

Verification

After 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+

Error Handling

If the server is not reachable, report the error and stop
If an entity creation fails, log the error and continue with the next
At the end, report success count vs failure count for each entity type

Name	seed-data
Description	Seeds a mahresources instance with semantic content across all entity types, creating enough items to trigger pagination (>50 per type)

seed-data

SKILL.md