본문으로 건너뛰기
cccskills
무료GitHub에 공개

codexkit-document-summarizer

Summarize document-heavy inputs such as memos, policies, PDFs, research notes, proposals, and multi-document packets. Use when users need key points, obligations, questions, and action implications from documents.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

설치 방법 보기

포함된 파일(2)

  • SKILL.md3.7 KB
  • agents/openai.yaml180 B

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Document Summarizer

When to Use

  • Summarizing long documents that are not formal business reports.
  • Extracting key points from policies, proposals, PDFs, research notes, or document packets.
  • Preparing a quick readout before deciding whether to read the full document.
  • Comparing multiple documents for themes, contradictions, or action items.

Procedure

Step 1 - Identify Document Type

Classify the source: policy, proposal, memo, research, contract-like document, transcript, academic paper, operating document, or mixed packet.

Step 2 - Extract Core Content

Capture:

  • purpose
  • main claims or findings
  • dates, parties, obligations, or deadlines
  • decisions requested
  • open questions
  • risks or caveats

Step 3 - Separate Summary From Interpretation

Keep source-backed summary separate from analysis or recommendation. Mark anything inferred.

Step 4 - Produce Audience-Specific Output

Adapt format for executive, manager, student, legal/compliance reviewer, or team member.

Step 5 - Add Reading Guidance

State which sections deserve human review and why.

Inputs

InputRequiredFormat
Document text or excerptYesPasted text, extracted notes, OCR
Document typeRecommendedProposal, policy, memo, research, etc.
AudienceRecommendedExecutive, manager, student, team
Desired lengthOptional5 bullets, one page, detailed brief
Focus areaOptionalRisks, obligations, actions, decisions

Output

## Document Summary - [Document Name]

### One-Line Takeaway
[Core point]

### Key Points
- [Point backed by source]

### Actions / Decisions
| Item | Owner | Deadline | Source Note |
|------|-------|----------|-------------|

### Risks / Caveats
- [Risk]

### Needs Human Review
- [Section or issue]

Quality Criteria

  • The summary stays faithful to the document and does not invent missing facts.
  • Important dates, obligations, numbers, and decisions are preserved.
  • Source-backed summary is separated from interpretation.
  • The output states what was not reviewed or could not be verified.
  • High-risk documents are flagged for expert review.

Verification (4C)

CheckQuestion
CorrectnessAre all summarized claims traceable to the document text?
CompletenessAre key points, actions, deadlines, risks, and review needs captured?
Context-fitDoes the summary match the audience and intended decision?
ConsequenceWhat could be misread if the user relies only on the summary?

Edge Cases

  • OCR or extraction errors - Flag low confidence and avoid definitive claims.
  • Legal, financial, or compliance content - Summarize only and recommend expert review before acting.
  • Multiple documents conflict - Show a contradiction table instead of choosing a winner.
  • Very long packet - Summarize by section and list skipped or unread portions.

Examples

Prompt: "Summarize this vendor proposal for an operations manager. Focus on deliverables, pricing assumptions, risks, and questions for the vendor."

Good pattern: "The proposal assumes customer data export is available by June 1. This is a dependency, not a confirmed commitment."

Definition of Done

  • The user can decide whether to read the full document.
  • Important actions and risks are visible.
  • Uncertainty and review needs are explicit.

Changelog

  • v1.0.0 - Initial release

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

Design rigorous A/B test plans with hypothesis, sample size calculation, Minimum Detectable Effect (MDE), randomization strategy, and decision rules. Includes guardrail metrics and rollout playbook. Use when planning product experiments, conversion optimization, or data-driven feature decisions.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hoavdc/CodexKit252026년 10월 11일 업데이트

Review REST and GraphQL API designs for consistency, usability, and best practices. Covers naming conventions, versioning strategy, error format, pagination, authentication patterns, and breaking change detection. Use when reviewing API specs, designing new APIs, or auditing existing endpoints.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hoavdc/CodexKit252026년 10월 11일 업데이트

Write Architecture Decision Records (ADRs) following the Michael Nygard format. Captures context, options considered, decision rationale, and consequences. Use when making technology choices, framework selections, or any architectural decision that future developers need to understand.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hoavdc/CodexKit252026년 10월 11일 업데이트

Assess organizational readiness for financial audits (internal or external). Map assertions to account balances, check evidence completeness, score readiness using a Red/Amber/Green framework, and generate a remediation timeline. Aligned with SOX, IFRS, and GAAP audit standards. Use before scheduled audits or when preparing for first-time compliance.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hoavdc/CodexKit252026년 10월 11일 업데이트

Design safe recurring Codex automations with clear prompts, outputs, schedules, and gating rules.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hoavdc/CodexKit252026년 10월 11일 업데이트

Refine Product Backlog Items to meet INVEST criteria. Write User Stories with Acceptance Criteria in Given/When/Then format, estimate with Story Points, and flag dependencies. Use before sprint planning when backlog items need grooming. Do not use to prioritize the backlog — that is the Product Owner's decision.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hoavdc/CodexKit252026년 10월 11일 업데이트

hoavdc의 스킬 모두 보기

이 스킬의 문제 신고