Skip to main content

​
social_ai_analyze

Analyze social media posts for relevance, sentiment, complaints, and urgency using Claude.
  • Category: action
  • Estimated Duration: 60s
  • Capabilities: supportsRerun, supportsBulkActions
  • Services:
    Required: anthropic
    Optional analyzedPosts, notifications

​
Input

​
topic
string
required
Original topic for context.
​
userIntent
string
required
What the user is looking for.
​
twitterPosts
SocialPost[]
Posts from Twitter monitor.
​
redditPosts
SocialPost[]
Posts from Reddit monitor.
​
linkedinPosts
SocialPost[]
Posts from LinkedIn monitor.
​
posts
SocialPost[]
Legacy field (backwards compat).
​
monitoringConfigId
string
For storage service.
{
  id: string;
  platform: string;
  url: string;
  text: string;
  title?: string;
  authorName: string;
  authorHandle: string;
  authorUrl: string;
  authorFollowers?: number;
  engagement: { likes: number; comments: number; shares: number; views?: number };
  postedAt: string;
}

​
Output

FieldTypeDescription
analyzedPostsAnalyzedPost[]All analyzed posts (sorted by relevance desc)
highPriorityPostsAnalyzedPost[]Posts with urgency=high or relevance≥80
complaintsAnalyzedPost[]Posts flagged as complaints
totalAnalyzednumberCount
highPriorityCountnumberCount
complaintCountnumberCount
averageRelevancenumberAverage relevance score
Extends SocialPost with: relevanceScore, sentiment ('positive'|'negative'|'neutral'), isComplaint, urgencyLevel ('low'|'medium'|'high'), aiSummary, matchedKeywords.

​
sora_video

Generate AI video using OpenAI Sora 2.
  • Category: integration
  • Estimated Duration: 60s
  • Capabilities: supportsRerun
  • Services:
    Required: openai

​
Input

​
prompt
string
required
Video description.
​
model
enum
default:"sora-2"
'sora-2' | 'sora-2-pro'
​
seconds
4 | 8 | 12
default:"4"
Video duration.
​
size
enum
default:"1280x720"
'720x1280' | '1280x720' | '1024x1792' | '1792x1024'

​
Output

FieldTypeDescription
videoobject{ url, durationSeconds, size, model }
processingTimeSecondsnumberTime taken to generate

​
seo_keyword_research

Research keywords with search volume, difficulty, CPC, and intent data.
  • Category: integration
  • Estimated Duration: 10s
  • Capabilities: supportsRerun
  • Services:
    Required: dataForSeo

​
Input

​
seedKeywords
string[]
required
Seed keywords to research.
​
locationCode
number
default:"2840"
Location code (default: US).
​
languageCode
string
default:"en"
Language code.
​
limit
number
default:"30"
Max keywords per seed.

​
Output

FieldTypeDescription
keywordsarray{ keyword, searchVolume, keywordDifficulty, cpc, searchIntent }
totalResearchednumberTotal keywords returned
searchIntent values: 'informational' | 'commercial' | 'navigational' | 'transactional'

​
seo_audit

Run comprehensive SEO audit on a URL.
  • Category: integration
  • Estimated Duration: 30s
  • Capabilities: supportsRerun
  • Services:
    Required: dataForSeo
    Optional notifications

​
Input

​
url
string (URL)
URL to audit. Can also come from context.variables.site_url.

​
Output

FieldTypeDescription
overallScorenumber | nullSEO score 0–100
issuesarray{ id, title, description, score, displayValue? }
passedAuditsnumberCount of passed checks
failedAuditsnumberCount of failed checks
urlstring | nullAudited URL
metaobject?{ title?, description?, canonical?, htags? }
performanceobject?{ timeToInteractive?, largestContentfulPaint?, domComplete? }
linksobject?{ internal, external, broken }
resourcesobject?{ images, scripts, stylesheets }
contentobject?{ wordCount, textRatio, readabilityScore? }
extractedKeywordsstring[]?Keywords found on page

​
social_keyword_generator

Generate platform-specific search keywords from a topic description using Claude.
  • Category: action
  • Estimated Duration: 15s
  • Capabilities: supportsRerun
  • Services:
    Required: anthropic
    Optional notifications

​
Input

​
topic
string
required
Natural language topic description.
​
userKeywords
string[]
User-specified keywords to include.

​
Output

FieldTypeDescription
topicstringEcho of input topic
twitterobject{ keywords: string[], searchQuery: string }
redditobject{ keywords: string[] }
linkedinobject{ keywords: string[], searchQueries: string[] }
allKeywordsstring[]All unique keywords across platforms