Menü
Late.dev publishing patterns for creating, scheduling, and cross-posting social media content
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
| name | late-publish |
| description | Late.dev publishing patterns for creating, scheduling, and cross-posting social media content |
node ../../../../.founderOS/scripts/late-tool.mjs posts create \
--accounts='["acc_123"]' \
--text="Post content here"
The CLI sets publishNow: true when neither --schedule nor --draft is provided.
node ../../../../.founderOS/scripts/late-tool.mjs posts create \
--accounts='["acc_123"]' \
--text="Scheduled post" \
--schedule="2026-03-15T09:00:00-05:00"
Always use ISO 8601 with timezone offset. Convert natural language times to ISO 8601 before calling the CLI.
node ../../../../.founderOS/scripts/late-tool.mjs posts create \
--accounts='["acc_123"]' \
--text="Draft content" \
--draft
Single API call with multiple account IDs:
node ../../../../.founderOS/scripts/late-tool.mjs posts create \
--accounts='["acc_linkedin_123","acc_twitter_456"]' \
--text="Cross-platform post"
Late.dev handles per-platform delivery. For higher-quality adaptation, the Content Adapter agent adapts content per platform before the API call, using separate posts create calls per platform.
Pass as JSON via --platform-options:
# LinkedIn: first comment + org posting
--platform-options='{"orgs":["org_123"],"firstComment":"Link: https://..."}'
# X/Twitter: thread
--platform-options='{"threadItems":[{"text":"Tweet 1"},{"text":"Tweet 2"}]}'
When a cross-post partially fails (some platforms succeed, others fail):
node ../../../../.founderOS/scripts/late-tool.mjs posts retry --post-id=post_abc123
This retries only the failed platform deliveries within the post.
{
"id": "post_abc123",
"status": "published|partial|failed|scheduled|draft",
"platforms": [
{ "accountId": "acc_123", "platform": "linkedin", "status": "published", "url": "https://..." },
{ "accountId": "acc_456", "platform": "twitter", "status": "failed", "error": "..." }
]
}
Researches individual competitors via web search to gather pricing, features, positioning, and reviews. Activates when the user wants to research a competitor, gather intel on a company, or asks 'what does [company] offer?' Covers multi-source intelligence gathering across official sites, review platforms, and community discussions.
Synthesizes competitive research into strategic insights, SWOT analyses, and positioning recommendations. Activates when the user wants a competitive analysis, market positioning review, competitor comparison matrix, or asks 'how do we stack up against the competition?' Identifies market gaps and strategic opportunities.
Extracts structured terms from legal contracts across 7 categories (Payment, Duration, IP, Confidentiality, Liability, Termination, Warranty). Activates when the user has a contract to analyze, wants to extract terms, review an agreement, or asks 'what does this contract say?' Supports PDF, DOCX, MD, and TXT formats with auto contract-type detection.
Evaluates contracts for harmful, unusual, or one-sided clauses using Red/Yellow/Green risk classification. Activates when the user wants to find contract red flags, assess legal risks, check for unfair terms, or asks 'is this contract safe to sign?' Provides plain-English explanations and concrete mitigation suggestions for freelancers and agency founders.
Creates and manages activity records in the CRM Communications database with AI-generated summaries. Activates when the user wants to log an email or meeting to CRM, record a client interaction, or asks 'save this to the client record.' Handles idempotent operations, deduplication, and AI summarization.
Matches email addresses and meeting attendees to CRM contacts using progressive matching. Activates when the user needs to identify which client an email belongs to, resolve a contact, or look up a client from an email address. Uses a 5-step algorithm from exact email match to fuzzy name matching with confidence scoring.