LoopCLI
Back to docsFeaturesPricingContact
Login / Sign Up
Back to docs
Templates / Ai

AI & Automation Templates

Workflow templates for AI-powered automation: content generation, data enrichment, sentiment analysis, and intelligent task routing—all orchestrated from the terminal.

AI Content Generation Pipeline

Automate blog post creation with Claude/GPT-4, generate social media variations, and publish across platforms.

Use Case

Generate weekly blog content using AI, create social media summaries, optimize for SEO, and publish to your CMS and social channels.

Workflow Configuration

name: ai-blog-content-pipeline
description: AI-powered content generation and distribution
trigger:
  type: schedule
  cron: "0 9 * * 1"  # Mondays at 9am
steps:
  - name: generate_blog_outline
    type: http
    config:
      method: POST
      url: https://api.anthropic.com/v1/messages
      headers:
        x-api-key: ${ANTHROPIC_API_KEY}
        anthropic-version: "2023-06-01"
        Content-Type: application/json
      body: |
        {
          "model": "claude-3-5-sonnet-20241022",
          "max_tokens": 2048,
          "messages": [{
            "role": "user",
            "content": "Generate a detailed outline for a blog post about: ${TOPIC}. Target audience: ${TARGET_AUDIENCE}. Include SEO keywords and key points."
          }]
        }
    output: outline

  - name: generate_full_article
    type: http
    config:
      method: POST
      url: https://api.anthropic.com/v1/messages
      headers:
        x-api-key: ${ANTHROPIC_API_KEY}
        anthropic-version: "2023-06-01"
        Content-Type: application/json
      body: |
        {
          "model": "claude-3-5-sonnet-20241022",
          "max_tokens": 4096,
          "messages": [{
            "role": "user",
            "content": "Write a complete blog post based on this outline:\n\n${outline.content[0].text}\n\nMake it engaging, include examples, and optimize for SEO. Format in markdown."
          }]
        }
    output: article

  - name: generate_social_posts
    type: http
    config:
      method: POST
      url: https://api.anthropic.com/v1/messages
      headers:
        x-api-key: ${ANTHROPIC_API_KEY}
        anthropic-version: "2023-06-01"
      body: |
        {
          "model": "claude-3-5-sonnet-20241022",
          "max_tokens": 1024,
          "messages": [{
            "role": "user",
            "content": "Based on this blog post, create:\n1. A Twitter thread (5 tweets)\n2. A LinkedIn post\n3. A Facebook post\n\nBlog content:\n${article.content[0].text}"
          }]
        }
    output: social

  - name: publish_to_cms
    type: http
    config:
      method: POST
      url: https://your-cms.com/api/posts
      headers:
        Authorization: Bearer ${CMS_API_KEY}
        Content-Type: application/json
      body: |
        {
          "title": "${TOPIC}",
          "content": "${article.content[0].text}",
          "status": "draft",
          "tags": ["${TAGS}"]
        }
    output: cms_post

  - name: store_social_content
    type: http
    config:
      method: POST
      url: https://your-project.supabase.co/rest/v1/social_queue
      headers:
        apikey: ${SUPABASE_KEY}
        Content-Type: application/json
      body: |
        {
          "blog_post_id": "${cms_post.id}",
          "twitter_thread": "${social.content[0].text}",
          "linkedin_post": "${social.content[0].text}",
          "scheduled_for": "${timestamp + 86400}"
        }

  - name: notify_content_team
    type: http
    config:
      method: POST
      url: ${SLACK_WEBHOOK_URL}
      body: |
        {
          "text": "📝 AI-generated content ready for review",
          "blocks": [
            {
              "type": "section",
              "text": {
                "type": "mrkdwn",
                "text": "*New Blog Post Generated*\n• Topic: ${TOPIC}\n• Status: Draft\n• <${cms_post.preview_url}|Review Post>\n• Social posts queued for tomorrow"
              }
            }
          ]
        }

Setup Instructions

  1. Create the workflow:

    loopcli loop create ai-blog-content-pipeline

  2. Set your secrets:

    loopcli loop secrets set ANTHROPIC_API_KEY
loopcli loop secrets set CMS_API_KEY
loopcli loop secrets set SUPABASE_KEY

  3. Deploy with schedule:

    loopcli loop deploy ai-blog-content-pipeline --activate

AI-Powered Data Enrichment

Enrich customer data using AI: extract company info, categorize industries, predict churn risk, and update CRM.

Use Case

When new leads are added to your CRM, use AI to research the company, categorize their industry, analyze fit score, and enrich with additional data points.

Workflow Configuration

name: ai-lead-enrichment
description: AI-powered lead data enrichment
trigger:
  type: webhook
  path: /webhooks/new-lead
steps:
  - name: research_company
    type: http
    config:
      method: POST
      url: https://api.anthropic.com/v1/messages
      headers:
        x-api-key: ${ANTHROPIC_API_KEY}
        anthropic-version: "2023-06-01"
      body: |
        {
          "model": "claude-3-5-sonnet-20241022",
          "max_tokens": 1024,
          "messages": [{
            "role": "user",
            "content": "Research this company and provide:\n1. Industry category\n2. Estimated company size\n3. Key products/services\n4. Tech stack (if visible)\n5. Potential use cases for our product\n\nCompany: ${webhook.company}\nWebsite: ${webhook.website}"
          }]
        }
    output: research

  - name: calculate_fit_score
    type: http
    config:
      method: POST
      url: https://api.openai.com/v1/chat/completions
      headers:
        Authorization: Bearer ${OPENAI_API_KEY}
        Content-Type: application/json
      body: |
        {
          "model": "gpt-4",
          "messages": [{
            "role": "system",
            "content": "You are a lead scoring expert. Analyze the company data and return a fit score from 0-100 and a brief explanation."
          }, {
            "role": "user",
            "content": "Company data:\n${research.content[0].text}\n\nOur ideal customer profile:\n- Industry: ${ICP_INDUSTRY}\n- Size: ${ICP_SIZE}\n- Tech stack: ${ICP_TECH}"
          }],
          "temperature": 0.3
        }
    output: fit_score

  - name: enrich_with_clearbit
    type: http
    config:
      method: GET
      url: https://company.clearbit.com/v2/companies/find?domain=${webhook.website}
      headers:
        Authorization: Bearer ${CLEARBIT_API_KEY}
    output: clearbit

  - name: update_crm
    type: http
    config:
      method: PATCH
      url: https://your-project.supabase.co/rest/v1/leads?id=eq.${webhook.lead_id}
      headers:
        apikey: ${SUPABASE_KEY}
        Content-Type: application/json
      body: |
        {
          "industry": "${research.content[0].text}",
          "company_size": "${clearbit.metrics.employees}",
          "fit_score": ${fit_score.choices[0].message.content},
          "enriched": true,
          "tech_stack": "${clearbit.tech}",
          "updated_at": "${timestamp}"
        }

  - name: notify_sales_high_fit
    type: http
    condition: fit_score.choices[0].message.content > 75
    config:
      method: POST
      url: ${SLACK_WEBHOOK_SALES}
      body: |
        {
          "text": "🎯 High-fit lead enriched!",
          "blocks": [
            {
              "type": "section",
              "text": {
                "type": "mrkdwn",
                "text": "*High-Fit Lead Alert*\n• Company: ${webhook.company}\n• Fit Score: ${fit_score.choices[0].message.content}/100\n• Industry: ${research.content[0].text}\n• Size: ${clearbit.metrics.employees} employees\n• Recommended action: Immediate outreach"
              }
            }
          ]
        }

Sentiment Analysis & Customer Support Routing

Analyze customer support tickets with AI, classify sentiment, detect urgency, and route to the right team.

Use Case

When support tickets arrive, use AI to analyze sentiment, detect urgent issues, categorize by topic, and route to specialized teams.

Workflow Configuration

name: ai-support-ticket-routing
description: AI-powered support ticket analysis and routing
trigger:
  type: webhook
  path: /webhooks/new-ticket
steps:
  - name: analyze_ticket
    type: http
    config:
      method: POST
      url: https://api.anthropic.com/v1/messages
      headers:
        x-api-key: ${ANTHROPIC_API_KEY}
        anthropic-version: "2023-06-01"
      body: |
        {
          "model": "claude-3-5-sonnet-20241022",
          "max_tokens": 512,
          "messages": [{
            "role": "user",
            "content": "Analyze this support ticket and return JSON with:\n1. sentiment (positive/neutral/negative/angry)\n2. urgency (low/medium/high/critical)\n3. category (billing/technical/feature-request/bug)\n4. suggested_team (support/engineering/sales/success)\n5. summary (1 sentence)\n\nTicket:\n${webhook.subject}\n${webhook.message}"
          }]
        }
    output: analysis

  - name: create_ticket_in_system
    type: http
    config:
      method: POST
      url: https://your-project.supabase.co/rest/v1/tickets
      headers:
        apikey: ${SUPABASE_KEY}
        Content-Type: application/json
      body: |
        {
          "customer_email": "${webhook.email}",
          "subject": "${webhook.subject}",
          "message": "${webhook.message}",
          "sentiment": "${analysis.content[0].text.sentiment}",
          "urgency": "${analysis.content[0].text.urgency}",
          "category": "${analysis.content[0].text.category}",
          "assigned_team": "${analysis.content[0].text.suggested_team}",
          "created_at": "${timestamp}"
        }
    output: ticket

  - name: alert_on_critical
    type: http
    condition: analysis.content[0].text.urgency == 'critical'
    config:
      method: POST
      url: ${PAGERDUTY_WEBHOOK_URL}
      body: |
        {
          "routing_key": "${PAGERDUTY_KEY}",
          "event_action": "trigger",
          "payload": {
            "summary": "Critical support ticket: ${webhook.subject}",
            "severity": "critical",
            "source": "support-ai",
            "custom_details": {
              "customer": "${webhook.email}",
              "sentiment": "${analysis.content[0].text.sentiment}",
              "ticket_id": "${ticket.id}"
            }
          }
        }

  - name: notify_team
    type: http
    config:
      method: POST
      url: ${SLACK_WEBHOOK_URL}
      body: |
        {
          "text": "New ticket: ${webhook.subject}",
          "blocks": [
            {
              "type": "section",
              "text": {
                "type": "mrkdwn",
                "text": "*New Support Ticket*\n• Customer: ${webhook.email}\n• Sentiment: ${analysis.content[0].text.sentiment}\n• Urgency: ${analysis.content[0].text.urgency}\n• Category: ${analysis.content[0].text.category}\n• Routed to: ${analysis.content[0].text.suggested_team}\n• Summary: ${analysis.content[0].text.summary}"
              }
            }
          ]
        }

  - name: send_auto_response
    type: http
    config:
      method: POST
      url: https://api.resend.com/emails
      headers:
        Authorization: Bearer ${RESEND_API_KEY}
      body: |
        {
          "from": "support@yourapp.com",
          "to": ["${webhook.email}"],
          "subject": "Re: ${webhook.subject}",
          "html": "<p>Thanks for reaching out! We've received your ${analysis.content[0].text.category} request and our ${analysis.content[0].text.suggested_team} team will respond within ${analysis.content[0].text.urgency == 'critical' ? '1 hour' : '24 hours'}.</p>"
        }

AI Code Review Assistant

Automate code review with AI: analyze pull requests, provide feedback, check for security issues, and comment on PRs.

Workflow Configuration

name: ai-code-review
description: AI-powered code review for pull requests
trigger:
  type: webhook
  path: /webhooks/github-pr
steps:
  - name: fetch_pr_diff
    type: http
    config:
      method: GET
      url: https://api.github.com/repos/${webhook.repository.full_name}/pulls/${webhook.number}/files
      headers:
        Authorization: token ${GITHUB_TOKEN}
    output: diff

  - name: ai_code_review
    type: http
    config:
      method: POST
      url: https://api.anthropic.com/v1/messages
      headers:
        x-api-key: ${ANTHROPIC_API_KEY}
        anthropic-version: "2023-06-01"
      body: |
        {
          "model": "claude-3-5-sonnet-20241022",
          "max_tokens": 2048,
          "messages": [{
            "role": "user",
            "content": "Review this code diff and provide:\n1. Security issues\n2. Performance concerns\n3. Code quality suggestions\n4. Best practice violations\n\nDiff:\n${JSON.stringify(diff)}\n\nFocus on critical issues only."
          }]
        }
    output: review

  - name: comment_on_pr
    type: http
    config:
      method: POST
      url: https://api.github.com/repos/${webhook.repository.full_name}/issues/${webhook.number}/comments
      headers:
        Authorization: token ${GITHUB_TOKEN}
        Content-Type: application/json
      body: |
        {
          "body": "## AI Code Review\n\n${review.content[0].text}\n\n---\n*Automated review by Claude Code*"
        }

  - name: check_security_issues
    type: cli
    config:
      command: |
        echo "${review.content[0].text}" | grep -i "security" | wc -l
    output: security_count

  - name: request_security_review
    type: http
    condition: security_count.stdout > 0
    config:
      method: POST
      url: https://api.github.com/repos/${webhook.repository.full_name}/pulls/${webhook.number}/requested_reviewers
      headers:
        Authorization: token ${GITHUB_TOKEN}
      body: |
        {
          "reviewers": ["security-team-lead"]
        }

AI Image Generation & Asset Pipeline

Generate marketing assets with AI image generation (DALL-E/Midjourney), optimize, and upload to CDN.

Workflow Configuration

name: ai-image-generation-pipeline
description: Generate and deploy AI images
trigger:
  type: manual
steps:
  - name: generate_images
    type: http
    config:
      method: POST
      url: https://api.openai.com/v1/images/generations
      headers:
        Authorization: Bearer ${OPENAI_API_KEY}
        Content-Type: application/json
      body: |
        {
          "model": "dall-e-3",
          "prompt": "${IMAGE_PROMPT}",
          "n": 1,
          "size": "1024x1024",
          "quality": "hd"
        }
    output: image

  - name: download_image
    type: cli
    config:
      command: |
        wget -O /tmp/generated-image.png "${image.data[0].url}"

  - name: optimize_image
    type: cli
    config:
      command: |
        convert /tmp/generated-image.png -quality 85 -strip /tmp/optimized-image.webp

  - name: upload_to_cdn
    type: http
    config:
      method: POST
      url: https://api.cloudflare.com/client/v4/accounts/${CF_ACCOUNT_ID}/images/v1
      headers:
        Authorization: Bearer ${CF_API_KEY}
      body_file: /tmp/optimized-image.webp
    output: cdn

  - name: store_metadata
    type: http
    config:
      method: POST
      url: https://your-project.supabase.co/rest/v1/assets
      headers:
        apikey: ${SUPABASE_KEY}
      body: |
        {
          "prompt": "${IMAGE_PROMPT}",
          "cdn_url": "${cdn.result.variants[0]}",
          "created_at": "${timestamp}"
        }

Getting Started

  1. Install LoopCLI:

    npm install -g loopcli

  2. Get API keys:

  3. Create your AI workflow:

    loopcli loop create my-ai-workflow

  4. Set your AI API keys:

    loopcli loop secrets set ANTHROPIC_API_KEY
loopcli loop secrets set OPENAI_API_KEY

  5. Deploy and run:

    loopcli loop deploy my-ai-workflow --activate

Need Help?

Related Documentation

Continue learning with these related topics

Starter Recipes

These lightweight recipes give you the “what” and “why” for popular automations without overwhelming you with YAML. Each one links to the matching persona template so you can grab the full workflow when you’re ready.

Developer & Engineer Templates

Workflow templates designed for software engineers and developers who want to automate deployment pipelines, testing workflows, and development operations from the terminal.

Founder & Solo Builder Templates

Workflow templates for founders and solo builders who need to automate customer onboarding, payment flows, analytics, and growth operations—all from the terminal.

Product Manager Templates

Workflow templates for product managers who need to automate feature releases, track metrics, coordinate launches, and keep stakeholders informed—all from the terminal.