Skip to main content

Compliance Agents

ZebraTruth uses a pipeline of specialized AI agents, each focused on a specific compliance domain. Agents are added regularly — always use GET /agents to discover available agents dynamically.

Current Agents

AgentPhaseInputWhat It Checks
Jurisdiction Detection1TextDetects applicable laws (FTC, UCPD, ASA, etc.)
Advertising Law2TextClaim substantiation, endorsement rules, targeting
Platform Policy2TextYouTube, Instagram, Facebook, TikTok, LinkedIn rules
Metadata & Labeling3TextC2PA Content Credentials, AI disclosure labels
Rights Clearance (Text)3TextCelebrity references, trademarks, copyrighted characters
Rights Clearance (Image)3ImageCelebrity faces, logos, watermarks via computer vision
Compliance Score4AllDeterministic scoring from all checks

Pipeline Phases

In full mode, agents run in 4 sequential phases with content enrichment:
  1. Phase 1 — Jurisdiction agent detects applicable laws
  2. Phase 2 — Advertising Law + Platform agents (parallel), enriched with Phase 1 findings
  3. Phase 3 — Metadata + Rights agents (parallel), enriched with Phase 1+2 findings
  4. Phase 4 — Score agent aggregates all checks
In fast mode, all agents run in parallel without enrichment.

3-Model Sub-Agent Pipeline

Each text agent internally runs a 3-stage pipeline:
StageModelRoleCacheable
APerplexity ProLegal research & citationsYes (30-day TTL)
BGPT-5.2Content-specific risk analysisNo
Cclaude-opus-4-6Structured compliance writingNo

Invoking a Single Agent

Call any agent directly via POST /agents/{agentId}. Costs ~8-12 credits vs ~47 for a full check. See Individual Agents guide for details.