agentsclimarketplace

Deck Lens

Skill BohanCui/Deck-Lens

Audience-aware presentation framing for pitches, reports, and technical handoffs

Install
npx -y skills add BohanCui/Deck-Lens

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use this skill to choose the right narrative lens for a presentation based on audience, context, and goal, then generate the matching slide structure, PPTX, or speech script. Triggers include "make me a slide deck", "prepare a pitch", "help me present X", "write a speech script", "build a presentation", "帮我做个PPT", "帮我写汇报", "做个presentation", "写个演讲稿", "我要汇报", "比赛要交PPT". Also triggers when the user describes a product, feature, or project they want to present — even without saying "PPT". Always run this skill before generating any presentation content.

SKILL.md

5.6 KB, as published. Nobody here has run it

Deck Lens

Deck Lens decides how a presentation should be told before producing anything. It is not a generic slide generator — it first picks the right narrative lens for the audience, context, and goal, then generates the matching structure, PPTX, or speech script.

The same project told through the wrong lens fails. A budget request told as a competition pitch feels like a performance; a competition pitch told as a government submission feels lifeless. Choosing the lens is the job.

Step 1 — Read the Situation

Before anything else, understand who the deck is for and why it exists. Ask the user, or infer from what they've already said:

  • Audience — Who is in the room? Investors, judges, clients, internal leadership, a government review panel, engineers?
  • Context — What's the occasion? Competition, funding round, internal review, formal submission, technical handoff?
  • Goal — What outcome do you want? Win, get funded, get approval, get sign-off, transfer knowledge?

If the user already made this clear, don't re-ask — move to Step 2.

Step 2 — Recommend a Lens (auto-detect, then confirm)

Deck Lens has four lenses. Match the situation to one:

LensUse forCore logic
PitchCompetition, fundraising, client demo, product launchWin attention → feel the pain → solution → pricing → call to action
Internal ReportLeadership / management / cross-department reportingConclusion first, then evidence (pyramid principle)
Government / InstitutionGovernment bodies, public institutions, regulatorsLegitimacy → objective evidence → reliability → verifiable outcomes → risk control
Technical HandoffDevelopers, architects, technical client stakeholdersRequirement discovery → design → logic → implementation → testing → delivery

Auto-detection signals:

  • "for leadership / management / report to my boss / 给领导汇报" → Internal Report
  • "competition / judges / investors / clients / pitch / 比赛 / 路演 / 融资" → Pitch
  • "government / public institution / regulator / formal submission / 体制内 / 政府 / 事业单位" → Government / Institution
  • "technical lead / architecture review / handoff / developers / 技术负责人 / 交接 / 架构评审" → Technical Handoff

Always confirm before proceeding. State your pick and the reasoning, then ask. Example:

This looks like Internal Report mode — the audience is internal leadership, so they'll want the conclusion up front, then the supporting evidence. Shall I proceed with this lens?

Do not skip confirmation even when the signal is strong. The user may know something about the room that you don't.

Hybrid situations: If a deck genuinely spans two lenses (e.g. a technical pitch, or an internal report that also requests budget), pick one primary lens and borrow at most one beat from the other. Name the borrowed beat explicitly. Do not blend all four — a four-way hybrid has no shape.

Step 3 — Ask for Output Format

Once the lens is confirmed:

What output format do you need?

  1. PPTX file — A ready-to-open slide deck
  2. Speech script — A detailed written script for speaking or memorizing
  3. Both — PPTX + matching speech script

Step 4 — Gather Content

Ask the user to describe (skip any they've already provided):

  1. Topic — What is this about?
  2. Pain points — What problem does it solve? What's broken today?
  3. Solution — What does your product, feature, or approach do?
  4. Key highlights — The top 3–5 things the audience should remember
  5. Future value — Any roadmap, potential, or scalability worth mentioning?

Default content depth is non-technical (no code, no architecture, no implementation detail) for every lens except Technical Handoff, which has its own depth rules.

Step 5 — Generate

  1. Load the confirmed lens file from modes/ and follow its structure beat by beat.
  2. If PPTX or Both: apply a design template. Default is templates/design-default.md. If the user has their own template, use theirs instead.
  3. If Speech script or Both: write natural spoken language per section, with transition lines and estimated speaking time (~150 words/min English, ~120 characters/min Chinese).
  4. If Both: generate the PPTX first, then derive the script from the slide content.
  5. For PPTX generation, read /mnt/skills/public/pptx/SKILL.md first and follow its instructions.

Lens Files

  • modes/pitch.md — Pitch
  • modes/internal.md — Internal Report
  • modes/government.md — Government / Institution
  • modes/technical.md — Technical Handoff

Design Templates

  • templates/design-default.md — Default lens (Royal Blue, bright background). Replaceable — copy it, change the values, and point the skill at the new file.

Reference Example

  • examples/spectrum-analyzer.md — A worked example using the Pitch lens.

Gives 0 of the 12 instructions most slides presentations skills give

Counted across 568 of the 571 authors here whose files we hold, read 2026-08-06

  • include a visual element on every slidein 52 of 568, across 21 files
  • put one idea per slidein 50 of 568, across 42 files
  • state the design approach before writing codein 38 of 568, across 8 files
  • validate XML immediately after each editin 37 of 568, across 7 files
  • rasterize gradients and icons as PNG images before referencing themin 37 of 568, across 7 files
  • generate and inspect thumbnails to validate layoutin 37 of 568, across 7 files
  • commit to a single visual motif across every slidein 37 of 568, across 12 files
  • use web-safe fonts onlyin 36 of 568, across 7 files
  • keep 0.5 inch minimum marginsin 35 of 568, across 9 files
  • use two-column layout for slides with charts or tablesin 34 of 568, across 5 files
  • save a template inventory analysis to a filein 33 of 568, across 4 files
  • use subagents to visually inspect rendered slidesin 33 of 568, across 9 files

Said here and by no other author read

  • run this skill before generating presentation content
  • determine the audience context and goal first
  • match the situation to one of four lenses
  • state the chosen lens and reasoning
  • confirm the chosen lens with the user
  • pick only one primary lens for hybrid situations

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.