
AI ASSISTANTS
AI-Powered Research & Content Engine
Multi-agent automation from research to publication
Multi-agent automation for research, fact-checking, writing, and multi-channel content delivery.
THE PROBLEM
Researching topical AI automation trends, verifying sources, and generating consistent deliverables across channels (blog, email, LinkedIn, newsletter) was manual, inconsistent, and time-intensive. Without structured verification, content risked unverified claims or inaccuracies.
THE SOLUTION
- Built a multi-agent AI system using n8n with a supervisor agent orchestrating specialist agents
- Each step outputs structured JSON contracts for governable transitions between agents
- Agents handle: research, fact-checking, writing, repurposing, and distribution
AGENT ARCHITECTURE
Orchestrator
Manager- Reads input (topic, audience, tone, timeframe)
- Calls specialist agents in sequence
- Prevents hallucinations by enforcing tool usage
- Outputs final structured deliverables
Research Scout
Research Agent- Finds reputable sources (OpenAI, Anthropic, n8n, industry blogs)
- Extracts claims with excerpts
- Tags recency and risk
Fact Checker
Verification Agent- Confirms accuracy against sources
- Removes or flags weak claims
- Ensures evidence for every fact
Writer
Content Agent- Generates skimmable, audience-tailored articles
- Uses only verified claims
- Outputs blog draft + TL;DR bullets
Repurposer
Editor Agent- Polishes tone and structure
- Transforms draft into: blog markdown, LinkedIn post, newsletter snippet
- Generates source table markdown
WORKFLOW FLOW
IMPLEMENTATION STEPS
- 1.Input parameters (topic, audience, tone, timeframe)
- 2.Orchestrator AI assigns tasks
- 3.Research agent gathers sources and extracts claims
- 4.Fact-checker validates information
- 5.Writer composes content based on verified data
- 6.Repurposer formats outputs for multiple channels
- 7.Automated distribution to Email/Slack/Notion
IMPACT & RESULTS
- Reduced research-to-publish time from ~3 hours to ~20–30 minutes
- Standardized content structure across all outputs
- Improved reliability through fact-checking and source verification
- Scalable system for future content expansion
- Created a reusable AI content operations system