{
  "schema_version": 2,
  "kind": "shortening-method",
  "format": "agent-skill",
  "id": "slide-copy",
  "name": "Slide copy",
  "category": "Writing",
  "summary": "Reduce a section to one slide claim and a few readable supporting points.",
  "use_cases": [
    "Presentation slides",
    "Briefing decks",
    "One-slide summaries"
  ],
  "word_count": 144,
  "url": "https://sho.rten.it/methods/slide-copy/",
  "instructions_url": "https://sho.rten.it/methods/slide-copy/SKILL.md",
  "skill_url": "https://sho.rten.it/methods/slide-copy/SKILL.md",
  "json_url": "https://sho.rten.it/methods/slide-copy/llms.txt",
  "plain_text_url": "https://sho.rten.it/methods/slide-copy/prompt.txt",
  "license": "MIT",
  "sources_url": "https://sho.rten.it/sources/#slide-copy",
  "skill_name": "slide-copy",
  "skill_description": "Reduce a section to one slide claim and a few readable supporting points. Use for Presentation slides, Briefing decks, One-slide summaries.",
  "agents_md_url": "https://sho.rten.it/methods/slide-copy/AGENTS.md",
  "sources": [],
  "instructions": "Condense the supplied section into text for one presentation slide. Make the title state the main finding or decision, rather than just name the topic. Use up to three brief supporting points unless the user supplies a different slide structure.\n\nSelect evidence that directly supports the title. Keep units, baselines, dates, sample scope, and conditions needed to interpret numbers. Place a material limitation beside the claim it qualifies; do not hide it in nonexistent speaker notes. Use parallel phrases where they remain clear. If the input contains unrelated claims, choose the requested focus or identify that more than one slide is needed.\n\nRemove spoken transitions, repeated explanation, decorative language, and detail the audience does not need on screen. Do not increase certainty to make a punchier title. Do not invent a chart, statistic, decision, or recommendation. Return only the slide title and supporting points.",
  "example": {
    "context": "One slide for an operations review",
    "before": "The purpose of this slide is to explain what happened in the evening support pilot. During August, the team tested an additional evening shift for the UK queue. Median first-response time in that queue fell from 90 minutes in July to 55 minutes in August. Customer satisfaction stayed at 4.3 out of 5. August ticket volume was 18 percent lower than July, so we cannot attribute the change in response time entirely to the additional shift. This was a UK-only pilot.",
    "after": "UK response time fell during the August evening-shift pilot\n\n- Median first response: 90 min in July → 55 min in August.\n- Satisfaction stayed at 4.3/5.\n- Ticket volume fell 18%; the shift's effect is uncertain.",
    "must_preserve": [
      "UK-only August evening-shift pilot",
      "July-to-August response comparison and units",
      "Satisfaction unchanged at 4.3/5",
      "18% lower volume limits causal interpretation"
    ],
    "omitted": [
      "Removes spoken setup and repeated pilot scope."
    ],
    "kind": "illustrative"
  }
}