Data Schemas
Preview schema
These schemas define the recommended shape for OptimAI data objects. They are intended for product, API, and integration alignment.
Professional AI docs need predictable objects. The following schemas describe the key entities builders will work with across Search, Claw, Persona Agent, Campaigns, Nodes, and Rewards.
Source
{
"id": "src_01JZ9K9B2MYYD2QJKF1EVW93X8",
"type": "web_page",
"url": "https://docs.optimai.network/",
"title": "OptimAI Network Docs",
"captured_at": "2026-06-22T08:20:00Z",
"freshness": "current",
"metadata": {
"domain": "docs.optimai.network",
"language": "en",
"content_type": "documentation"
},
"quality": {
"score": 0.91,
"signals": ["source_reputation", "freshness", "validator_sample"]
}
}
Citation
{
"source_id": "src_01JZ9K9B2MYYD2QJKF1EVW93X8",
"title": "OptimAI Network Docs",
"url": "https://docs.optimai.network/",
"quote": "OptimAI Network is a decentralized intelligence network...",
"confidence": 0.89
}
Claw Record
{
"id": "rec_01JZ9KP5CHQ5Q3T4WXSX0W8RZP",
"job_id": "claw_01JZ9KBC2HJ92A7BYGHDKZ7H01",
"data": {
"name": "Example Product",
"price": "$29",
"features": ["Semantic search", "Data extraction"],
"source_url": "https://example.com/products/example"
},
"provenance": {
"source_id": "src_01JZ9K9B2MYYD2QJKF1EVW93X8",
"node_id": "node_core_8f2a",
"captured_at": "2026-06-22T08:20:00Z"
},
"validation": {
"status": "sample_validated",
"quality_score": 0.87
}
}
Persona Memory
{
"id": "mem_01JZ9KD0D1S1DFKHMG0JFSQQ9A",
"persona_id": "persona_founder_qt",
"type": "preference",
"content": "Prefers concise technical docs with diagrams and API examples.",
"visibility": "private",
"source": "user_approved",
"tags": ["docs", "style", "product"],
"created_at": "2026-06-22T08:20:00Z",
"updated_at": "2026-06-22T08:20:00Z"
}
Campaign
{
"id": "camp_01JZ9KWBC5X2V1QV0C5K7EA4HF",
"name": "AI agent startup landscape",
"objective": "Create a structured dataset of AI agent startups, products, funding, and positioning.",
"status": "active",
"requirements": {
"sources": ["web", "social", "company_sites"],
"freshness": "14d",
"records_target": 500,
"require_citations": true
},
"reward_budget": {
"token": "OPI",
"amount": "25000"
},
"validation": {
"min_validators": 3,
"sample_rate": 0.2
}
}
Node Capability
{
"node_id": "node_core_8f2a",
"type": "core",
"capabilities": {
"browser_execution": true,
"claw_extraction": true,
"embedding_tasks": true,
"bandwidth": true,
"storage_gb": 128,
"gpu": false
},
"health": {
"status": "online",
"uptime_24h": 0.98,
"latency_ms": 82
},
"reputation": {
"score": 0.93,
"completed_tasks": 1842,
"validation_accuracy": 0.96
}
}
Reward Event
{
"id": "rew_01JZ9M0PXMQE7DEZ1RYCW8121P",
"account_id": "acct_01JZ9M12N0K0EQJZ7TE2S6MPQ4",
"node_id": "node_core_8f2a",
"task_id": "task_01JZ9M1KKR7YS4K7QGZV8G34MT",
"token": "OPI",
"amount": "42.50",
"reason": "validated_claw_extraction",
"quality_score": 0.91,
"status": "settled",
"created_at": "2026-06-22T08:20:00Z"
}
Schema Design Rules
- Use stable IDs with prefixes such as
src_,claw_,mem_,camp_,node_, andrew_. - Include
created_atorcaptured_attimestamps for auditability. - Preserve provenance for all data used by agents.
- Separate raw source metadata from validated quality signals.
- Keep private Persona memory scoped to the user unless explicitly shared.