← PrePilot LiteCONTENT HISTORYWHAT CHANGED · RULE-BASED ANALYSIS
Update to PrePilot Lite
Snapshot Sep 30, 2026 · 23:14 UTC · version 0.1.0
Collection source: not recorded for this historical snapshot.
First saved snapshot
No earlier snapshot is available to establish a change.
Compare saved observations
Download comparison JSONFull technical diff · 0 changed fields
Full snapshot data
{
"description": "Use when the user wants social media strategy or content for LinkedIn, X, Instagram, TikTok, Facebook, YouTube Shorts, or similar channels, including hooks, post concepts, scripts, calendars, and repurposing.",
"included_files": [
{
"relative_path": "agents/openai.yaml",
"size_in_bytes": 229
}
],
"name": "social-content",
"skill_md_contents": "---\nname: social-content\ndescription: Use when the user wants social media strategy or content for LinkedIn, X, Instagram, TikTok, Facebook, YouTube Shorts, or similar channels, including hooks, post concepts, scripts, calendars, and repurposing.\n---\n\n# Social Content\n\n## Use it for\n\n- Social strategy\n- Post or short-video creation\n- Founder or brand thought leadership\n- Content calendars\n- Repurposing long-form work into social assets\n\n## Operating rules\n\n- Use context already present in the conversation before asking for more input.\n- Separate known facts, reasonable assumptions, and unknowns. Do not present assumptions as evidence.\n- When the answer depends on current platform rules, market conditions, pricing, benchmarks, or competitors, verify them with current sources when tools are available.\n- Prefer concrete decisions, examples, and next actions over generic marketing advice.\n- Do not invent campaign performance, customer quotes, research findings, testimonials, rankings, or competitor claims.\n- Keep the requested market, language, funnel stage, audience awareness, budget, and channel constraints visible throughout the work.\n- Respect platform-native behavior without chasing every trend.\n- Write for a specific audience tension or useful idea.\n- Avoid fake personal stories, invented anecdotes, or manufactured controversy.\n- For current platform limits or formats, verify current rules when tools are available.\n\n## Workflow\n\n1. Define channel, audience, objective, creator/brand voice, available proof, and production constraints.\n2. Choose content angles based on audience tension, insight, proof, or useful demonstration.\n3. Match format to idea: text, carousel, short video, image, thread, or live interaction.\n4. Create the hook, body structure, proof/example, and action.\n5. Plan reuse across channels without copy-pasting the same asset.\n6. Define what to measure for the post's actual job.\n\n## Output contract\n\n- Channel-ready content or calendar\n- Hook and structure\n- Production notes when needed\n- Repurposing options\n- Measurement note\n\n## Quality gate\n\n- The hook earns attention without baiting\n- The body delivers the promised value\n- Voice is consistent\n- CTA is proportionate to the content\n\n## Voice and performance modes\n\nIf the user provides 3 or more representative writing samples, calibrate voice from recurring signals and absences before drafting. Preserve distinctive vocabulary, rhythm, bluntness, humor, uncertainty, and recurring structures rather than polishing every post into the same style.\n\nIf the user provides historical post performance, compare the draft with the user's own high- and low-performing patterns. Use the data as evidence, not as a universal formula. Report the dataset, metric definition, sample size, and age of the data before drawing a conclusion. Never substitute another creator's benchmark and present it as the user's expected performance.\n\nFor LinkedIn profile requests, treat the profile as a conversion surface: clarify the desired action, audience, offer, proof, headline, about section, featured proof/links, and consistency with the user's content voice.\n\n## Handoff\n\nIf the task is part of a larger marketing request, return the completed deliverable plus the evidence or decisions the next PrePilot skill needs. Do not repeat upstream analysis unless it changes the result.\n"
}SHA-256 of public snapshot: 8dd740fe907a474f9df8378bde5b38af0f1b64b608b1f7084178707174c260b6