data-cleanup
Clean and standardize messy tabular data (CSV, spreadsheet paste, system exports) into an analysis-ready dataset — consistent…
Summarize a long document into an executive summary, key points, and role-specific implications. Use when the user says "summarize this document", "give me the key points", "do I need to read this?", or drops a long PDF, report, or contract and asks what matters in it.
Claude Code
$ npx -y skills add sgharlow/claude-code-recipes --skill summarize-document --agent claude-code
Codex
$ npx skills add sgharlow/claude-code-recipes --skill summarize-document -a codex -y
Cursor
$ npx skills add sgharlow/claude-code-recipes --skill summarize-document -a cursor -y
Gemini CLI
$ npx skills add sgharlow/claude-code-recipes --skill summarize-document -a gemini-cli -y
GitHub Copilot
$ npx skills add sgharlow/claude-code-recipes --skill summarize-document -a github-copilot -y
OpenCode
$ npx skills add sgharlow/claude-code-recipes --skill summarize-document -a opencode -y
Windsurf
$ npx skills add sgharlow/claude-code-recipes --skill summarize-document -a windsurf -y
name: summarize-document description: Summarize a long document into an executive summary, key points, and role-specific implications. Use when the user says "summarize this document", "give me the key points", "do I need to read this?", or drops a long PDF, report, or contract and asks what matters in it.
Read a long document and return a decision-ready digest tuned to the reader's role. The summary must be faithful: every claim in it traces to the document, with section or page references where the format allows.
1. Confirm the target file and the reader's context — role, why they are reading it, and what they will do with it (present, decide, share). If context is missing, ask these three questions first; a summary without them is generic and half as useful. 2. Read the full document. For very long documents, read in passes: structure first (headings, tables, executive sections), then the sections that bear on the reader's stated purpose. 3. Produce, in order:
it matters to this reader.
stated purpose.
4. Verify: re-scan the document for any section not represented in the key points; either add it or state why it was omitted. Never present an inference as the document's claim — label reader-specific interpretation as "Implication", not "Finding".
file) — say exactly what was unreadable instead of summarizing the fragment silently.
Full walkthrough, examples, and variations: `recipes/Recipe-004-Document-Summarization.md`.
Claude Code artifacts graded by Skill Crossroads — click the badge for the evidence-cited scorecard. Your Complete Guide to AI-Powered Productivity
Clean and standardize messy tabular data (CSV, spreadsheet paste, system exports) into an analysis-ready dataset — consistent…
Turn raw meeting notes into a shareable summary, an action-item table with owners and deadlines, and follow-up email drafts. Use…
Turn a process description into an operational quality-control checklist — binary pass/fail checks, acceptance criteria,…
Synthesize multiple sources — articles, reports, notes, transcripts — into one structured brief with themes, agreements,…
Turn scattered weekly updates (notes, emails, task exports, meeting notes) into a structured status report with highlights,…
Shape output for a reader with ADHD: lead with the next action, number multi-step work, restate state across turns, suppress…
Set up the owner profile so a personal agent knows who it works for (first-run setup, \"get to know me\"), or distill colleague,…