Token导航 LogoToken导航TokenDH.com
前端设计只读github未标认证来源可访问许可证需确认审计通过

giving-presentations做演讲

Agent Skill

giving-presentations 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

776

周安装

33

GitHub Stars

3

下载量

272
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:giving-presentations(做演讲)
来源仓库:https://github.com/oldwinter/skills
仓库路径:skills/giving-presentations
安装命令:
npx skills add https://github.com/oldwinter/skills --skill giving-presentations
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/oldwinter/skills --skill giving-presentations

简介

giving-presentations 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态或协作事项进行整理。

  • 适用于需要可视化项目结构、流程或变更场景的 Agent 宿主环境。
  • 通过 npx skills add 命令从指定仓库安装,需结合原始 README 核验具体用法。
  • 安装前建议确认权限范围、维护状态,以及是否会触发联网或文件读写操作。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Giving Presentations

Scope

Covers

  • Turning a goal + audience into a clear talk objective and ask
  • Building a persuasive narrative using contrast (“what is” vs “what could be”)
  • Producing a slide/talk-track plan that is easy to deliver under time pressure
  • De-risking high-stakes talks via role-play, Q&A prep, and pre-briefs
  • Rehearsing for confidence (including visualization + record/review)
  • Delivery mechanics for in-person and Zoom (presence, pauses, looking up to think)

When to use

  • “Create an outline and talk track for my all-hands update.”
  • “Help me turn this doc into a 10-minute exec presentation with a clear ask.”
  • “I need a deck structure for a keynote / conference talk.”
  • “Prep me for Q&A and objections for a high-stakes review.”
  • “Build a rehearsal plan so I can deliver confidently.”

When NOT to use

  • You only need visual/brand design polish (use a design system or a designer; this skill focuses on narrative + delivery)
  • You need a long-form decision doc (write a memo/PRD first; then convert it to a talk)
  • You need deep stakeholder alignment on strategy from scratch (do alignment work first; this skill assumes a direction/ask)
  • You’re presenting on regulated/high-risk topics (medical/legal/financial advice) without expert review

Inputs

Minimum required

  • Presentation type + setting (all-hands, keynote, exec review, customer demo; in-person vs Zoom)
  • Audience (roles/seniority) + what they care about
  • Desired outcome (inform / align / decide / persuade) and the ask (decision, approval, next step)
  • Time limit and Q&A format (minutes; live Q&A vs async)
  • Core content (bullets, doc, notes, or an existing deck) + any must-include points
  • Constraints (deadline, level of polish, sensitive details to avoid)

Missing-info strategy

  • Ask up to 5 questions from references/INTAKE.md (3–5 at a time).
  • If details are missing, proceed with explicit assumptions and label unknowns.

Outputs (deliverables)

Produce a Presentation Pack in Markdown (in-chat; or as files if requested):

  1. Presentation brief (goal, audience, ask, constraints)
  2. Narrative outline (core message + “what is vs what could be” contrast)
  3. Slide-by-slide outline + talk track (each slide: takeaway, key points, evidence, speaker notes)
  4. Q&A / objection bank (top questions + crisp responses)
  5. Stakeholder pre-brief plan (who to pre-meet, what to align, how to de-risk)
  6. Rehearsal + delivery plan (visualization, record/review, timing, logistics, Zoom/in-person cues)
  7. Risks / Open questions / Next steps (always)

Templates: references/TEMPLATES.md Expanded guidance: references/WORKFLOW.md

Workflow (8 steps)

1) Intake: lock the objective, ask, and constraints

  • Inputs: user context + references/INTAKE.md.
  • Actions: Clarify audience, outcome, and the single most important ask. Confirm time limit and what is in/out of scope.
  • Outputs: Presentation brief (draft) + assumptions/unknowns list.
  • Checks: You can answer in one sentence: “After this talk, the audience will _____.”

2) Build the narrative spine using contrast

  • Inputs: brief + source content (doc/bullets/deck).
  • Actions: Define the “what is” current reality and the “what could be” future. Choose 2–4 supporting points and the call-to-action.
  • Outputs: Narrative outline (contrast table + story beats).
  • Checks: The contrast is concrete (not vague) and matches what the audience values.

3) Map the narrative to a slide/story structure

  • Inputs: narrative outline + time limit.
  • Actions: Select a structure (e.g., Context → Tension → Proposal → Proof → Ask). Create a slide list with 1 takeaway per slide and a rough time budget.
  • Outputs: Slide-by-slide outline (titles + takeaways + time plan).
  • Checks: The talk fits time with buffer; no slide has multiple competing takeaways.

4) Draft talk track and evidence (make it sayable)

  • Inputs: slide outline + evidence sources.
  • Actions: Write speaker notes (bullet talk track), add proof (metrics, examples, demos), and trim anything “nice to know.”
  • Outputs: Slide outline with speaker notes + evidence plan.
  • Checks: Each slide can be spoken without reading; jargon is translated for the audience.

5) Prepare for Q&A: role-play objections

  • Inputs: draft pack + stakeholder context.
  • Actions: Generate a Q&A / objection bank. Role-play the hardest audience member(s) and refine responses. Identify unanswered questions.
  • Outputs: Q&A bank + “unknowns to resolve” list.
  • Checks: Top 10 likely questions have concise answers and a fallback (“I’ll follow up by DATE”).

6) De-risk with stakeholder pre-briefs (no surprises)

  • Inputs: draft pack + stakeholder map.
  • Actions: Plan and run pre-meetings with key decision-makers/influencers. Capture objections early and update the narrative/ask.
  • Outputs: Pre-brief plan + change log (what changed and why).
  • Checks: No major stakeholder is seeing the core ask for the first time in the live meeting.

7) Rehearse for confidence (visualize + record + iterate)

  • Inputs: near-final outline/talk track.
  • Actions: Do a mental dress rehearsal (visualization), then a timed run. Record yourself, review, and iterate. Add delivery cues (pause, look up to think, avoid reading).
  • Outputs: Rehearsal plan + timing notes + delivery cues.
  • Checks: You can deliver within time twice in a row without major stumbles.

8) Finalize and run the quality gate

  • Inputs: final draft pack + logistics.
  • Actions: Run references/CHECKLISTS.md and score with references/RUBRIC.md. Confirm logistics (room/Zoom, backups). Produce the final pack.
  • Outputs: Final Presentation Pack + Risks/Open questions/Next steps.
  • Checks: A teammate can read the brief + slide outline and correctly predict the ask and flow.

Quality gate (required)

Examples

Example 1 (all-hands update): “Create a 7-minute all-hands presentation: what we shipped this quarter, what’s next, and what help we need from other teams.” Expected: brief, narrative contrast (current vs next), slide outline + talk track, Q&A, rehearsal + delivery plan.

Example 2 (exec review with decision): “I need a 12-minute exec review proposing a new onboarding flow. The ask is approval to run a 4-week experiment. Prep me for objections.” Expected: clear ask, proof points, objection bank, pre-brief plan for key execs, and a rehearsal plan.

Boundary example: “Make my slides prettier.” Response: clarify whether the problem is narrative/structure vs visual design; if it’s purely aesthetics, recommend design-system alignment or a designer and do not invent business content.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

39.35%
按下载量换算107

Claude

29.57%
按下载量换算80

Cursor

17.93%
按下载量换算49

Gemini CLI

9.44%
按下载量换算26

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills