Profile-Aware Social Media Pipeline
Content Creator Official v1.0.19Generate personalized multi-platform social content using your voice and audience profiles
by skrptiq
Sign in to import this skrpt into your workspace.
Sign inWhat's included 7 nodes
Profile-Aware Social Media Pipeline
Generate personalized multi-platform social content using your voice and audience profiles
Core Message Drafting
Drafts the core message for a social media topic using the creator's voice profile
Platform Adaptation
Adapts the core message for a specific social media platform while maintaining the creator's voice
Voice Verification
Checks all platform adaptations against the creator's voice profile for authenticity
Adapt for Platform
Adapts the core message for a specific platform — runs in for_each loop
Draft Core Message
Drafts the core social media message in the creator's voice
Verify Voice Consistency
Checks all platform adaptations against the creator's voice profile
Requirements 1 service, 1 data type
- LLM Service
- pii
Security Passed
Detected
- Services: llm-service
- Permissions:
- Data Handling:
Version history 16 releases
GH#863 Wave 3 (Tier-3 tail) — the output_step (language-polish) was polishing its positional-previous step (a validation/review), not the deliverable, because polish-language was pinned pre-`source`-slot. Repinned polish-language 1.0.1→1.0.6 and bound language-polish `source` ← "Platform Adaptation" (the real deliverable producer). No new steps; no other change.
GH#845 — republish with American English (en-US) content, completing the source-only GH#805 flip that never reached the Hub. Copy only — no functional or behaviour change.
GH#745 — declare per-step `output: {name, type}` on every execution step (core_message/text, platform_content/text, voice_verdict/decision, polished_posts/text, consistency_verdict/decision). Lights up the #744 rich flow-map. Content-only; no bindings or logic changes.
GH#711 — fix the `for_each` platform loop. The `platform-loop` iterates `{{input.platforms}}`, but the input was documented and exampled as a comma-separated string (`LinkedIn, Twitter, …`), which resolves to prose the loop cannot iterate (the engine now fails this loud — GH#557). The input contract is now a JSON array (or one platform per line), mirroring the proven `batch-competitor-analysis` for_each pattern, so the expression resolves to an iterable array. Content-only: Inputs table, Stage 2 wording (notes `{{loop.item}}`), and Example Input updated; no contract or engine change.
Fix-forward after Row 3b v1.0.14 publish failure. The v1.0.14 per-skrpt CI's "Register version with Hub API" step failed because the consumer's source `manifest.id` (5697d19d…) did not match the D1 catalogue row's id (912dbfcd…) — a legacy drift from before Action 6 (`0bcc5ae0`) made publish-skrpt.mjs Step 2 INSERT use `manifest.id` for the D1 id column. v1.0.15 reconciles the source `manifest.id` to the catalogue authoritative value (Row-5-equivalent for consumers) and republishes. Per Adj-1: no re-tag of v1.0.14; the orphaned GitHub release artefact stays inert (no D1 versions row, no consumer pinned it).
Wave 2: re-signed with canonical engine signing pipeline.
Signature fix — RELEASE_NOTES.md now included in integrity checksum.
Initial catalogue release with full structural and content-quality validation. All scanner checks pass.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
More from Content Creator
View all →Social Media Campaign
Multi-platform content creation: briefing, content generation, caption writing, and scheduling
Editorial Calendar Build
Audience analysis, ideation, calendar planning, and brief creation for editorial planning
Cross-Provider Technical Review
Draft with one AI model, critique with another, revise until the reviewer passes — better output through model diversity