Content Agent

Purpose

The Content Agent autonomously generates SEO-optimized blog posts, social media content, lead magnets, and email nurture sequences. It uses the Claude API for content generation, publishes to the Vercel-hosted marketing site, and distributes across social channels. Human review is optional but recommended for high-visibility pieces.

Runtime: AgentCore Runtime (marketplace-content-agent)

The Content Agent runs as an AgentCore Runtime deployment. The pipeline that was previously a Step Function is now the agent’s reasoning loop — the agent decides what to generate, generates it, and publishes it in a single long-running session (up to 8 hours covers a full weekly content run).

AgentCore services used by this agent: * AgentCore Runtime — hosts the agent; weekly sessions handle the full generate → review → publish → distribute pipeline * AgentCore Memory (content-calendar store) — published article index, keyword coverage map, scheduled topics, internal link map; replaces DynamoDB content tracking * AgentCore Memory (cross-agent-leads shared store) — writes new email nurture sequences so the Email Agent can pick them up * AgentCore Identity — manages Claude API key, LinkedIn API token, HubSpot token, Vercel deploy token, Ahrefs/Semrush API key * AgentCore Gateway — exposes hubspot_create_contact (for lead magnet captures), ses_send_email (newsletter blast coordination) as MCP tools

Trigger: EventBridge weekly Monday 6am ET → dispatch-content Lambda → InvokeAgentRuntime

Content Generation Pipeline

AgentCore Runtime: content-pipeline-session

Trigger: EventBridge weekly on Monday at 6am ET

1. Topic Selection
   → Query SEO keyword API (Ahrefs/Semrush) for volume + difficulty data
   → Read AgentCore Memory (content-calendar): existing keyword coverage, published topics
   → Agent reasons to select 2–4 topics for the week, assigns content type
   → Write topic plan to AgentCore Memory for session continuity

2. Draft Generation (per topic)
   → Call Claude API (via AgentCore Identity) with topic brief, keyword, word count
   → Generate full draft: meta title, meta description, H2/H3 structure, body, CTAs
   → Store draft in S3; write draft reference to AgentCore Memory

3. Review Queue (optional)
   → If auto-publish enabled: proceed directly to publishing
   → If manual review: agent writes draft status to AgentCore Memory, session suspends.
     Resumes when Vercel review webhook calls InvokeAgentRuntime with approval signal.
     (AgentCore Runtime supports async session suspension and resumption natively)

4. Publishing
   → Blog posts: push to CMS via Vercel/headless CMS API
   → Social: push to LinkedIn company page via LinkedIn API
   → Lead magnets: generate PDF via Vercel API route, upload to S3, create landing page
   → Email nurture: write sequences to AgentCore Memory (cross-agent-leads store);
     Email Agent reads them during its next send session

5. Distribution
   → Post to LinkedIn profiles via LinkedIn API
   → Coordinate newsletter blast via `ses_send_email` tool
   → Generate social snippets; store in S3 for manual posting on other platforms
   → Submit URL to Google Search Console API
   → Update internal link map in AgentCore Memory (content-calendar store)
   → Log all published content to AgentCore Memory for future keyword deduplication

Content Types & Prompts

Blog Posts (1,500–2,500 words)

System: You are a content writer specializing in Amazon and Walmart marketplace
seller education. You write for an audience of mid-market sellers ($100K–$1M/year)
who are business-savvy but not finance experts.

Rules:
- Write in seller language: SKUs, ASINs, PPC, ACoS, FBA, FBM, BSR, Buy Box
- Use specific numbers and examples throughout
- Never say "factoring" or "receivables financing" — say "daily payouts" or
  "early payout services"
- Include actionable takeaways the reader can implement immediately
- Mention [Company] product naturally as one solution, not as a hard sell
- Write in a practical, peer-to-peer tone — like advice from an experienced seller

Article brief:
- Target keyword: {keyword}
- Secondary keywords: {secondary_keywords}
- Content pillar: {pillar}  (Cash Flow Management | Payment Education | Seller Growth)
- Word count: {word_count}
- Include: {specific_sections_or_angles}
- Internal links to include: {urls}
- CTA: {cta_type} (calculator | demo | lead_magnet)

Generate: meta title (under 60 chars), meta description (under 155 chars),
full article with H2/H3 structure, and a CTA paragraph.

Priority Content Pillars

Pillar 1 — Cash Flow Management for Sellers: * "How to Calculate Your Amazon Cash Flow Gap" * "5 Ways Payment Delays Hurt Your Amazon Business" * "Cash Flow Planning for Q4: A Seller’s Guide" * "Why Your Amazon Revenue Doesn’t Match Your Bank Account" * "The Hidden Cost of Waiting for Amazon Payouts"

Pillar 2 — Payment & Payout Education: * "Amazon Payout Schedule Explained (2026 Update)" * "When Does Walmart Pay Marketplace Sellers?" * "What Happens When Amazon Holds Your Payments" * "Amazon Payment Reserve: What It Is and What to Do" * "Amazon vs Walmart: Comparing Seller Payout Schedules"

Pillar 3 — Seller Growth & Scaling: * "How to Fund Your Next Product Launch Without a Loan" * "Scaling Amazon PPC When Cash Flow Is Tight" * "Seller Financing Options Compared" * "How to Stock Up for Q4 When Cash Is Tied Up"

LinkedIn Posts (150–300 words)

System: Write a LinkedIn post for the [Company] profile targeting Amazon/Walmart
sellers. The post should be educational and engaging, not salesy. Use short
paragraphs (1–2 sentences each). Include a hook in the first line. End with
a question to drive engagement.

Topic: {topic}
Angle: {educational | case_study | industry_news | data_insight | poll}

Lead Magnets

Generate as long-form content, then convert to PDF using a Next.js API route on Vercel that renders HTML to PDF (using Puppeteer or react-pdf).

  1. Cash Flow Calculator — Interactive React component on Vercel. Input: monthly revenue, marketplace. Output: locked capital estimate, daily payout amount, annual fee, net benefit. Gate behind email capture.

  2. "Getting Paid Faster on Amazon" Guide — 15–20 page PDF covering all options.

  3. "Q4 Cash Flow Playbook" — Seasonal guide, regenerated annually with fresh data.

  4. "Payment Hold Survival Guide" — Tactical resource.

Email Nurture Sequences (for content leads)

When a seller downloads a lead magnet or subscribes to the blog, the Content Agent generates a 6-email nurture sequence written to AgentCore Memory (cross-agent-leads shared store). The Email Agent reads and sends them:

  • Day 0: Deliver the resource. Thank them.

  • Day 3: Related blog post or additional resource.

  • Day 7: Case study email.

  • Day 14: Soft introduction to daily payouts.

  • Day 21: Offer a demo or free cash flow analysis.

  • Day 30: Breakup + testimonial.

SEO Optimization

Each blog post is optimized before publishing: * Primary keyword in: title tag, H1, first 100 words, one H2, meta description, URL slug * Secondary keywords distributed naturally throughout * Internal links to at least 2 other blog posts and the product/demo page * External links to authoritative sources (Amazon Seller Central help pages, industry reports) * FAQ schema markup for relevant articles (generated by Claude API) * Image alt text optimized for keywords

Publishing Infrastructure

Vercel Marketing Site

  • Next.js app with MDX blog engine or headless CMS integration (Sanity, Contentful, or custom)

  • Content Agent pushes new posts via Git commit (using GitHub API) or CMS API

  • Automatic build and deploy on Vercel

  • Each blog post has: Open Graph meta tags, Twitter cards, schema markup, and analytics tracking

Content CDN

  • All static assets (PDFs, images, infographics) served from S3 via CloudFront

  • Lead magnet PDFs stored in S3 with signed URLs (expire after download)

Distribution Automation

After publishing, the Content Agent automatically: . Posts to LinkedIn company page via LinkedIn API . Generates 3 social media snippets for manual posting on other platforms . Adds to the next email newsletter batch (weekly digest to subscriber list) . Updates the internal link map in AgentCore Memory (content-calendar store) so future articles can reference it . Submits URL to Google Search Console for faster indexing

Expected Metrics

Metric Month 3 Target Month 6 Target

Blog posts published/month

8–12

12–16

LinkedIn posts/week

3–5

5–7

Monthly organic traffic

2,000–5,000

8,000–15,000

Blog-to-lead conversion

2–4%

3–6%

Lead magnet downloads/month

100–300

400–800

Content-sourced demos/month

10–20

25–50

Keywords ranking page 1

10–20

30–60