agentleFS
kastriasanikastriasani/google-ads-skills/skills/google-ads-presentation/SKILL.md

google-ads-presentation

Skill1 starsChanged 5 months agoMITView on GitHub

Structure a client-ready Google Ads proposal or QBR deck outline that a slide generator can render. Use whenever someone needs a pitch deck, a sales proposal, a client-ready slide outline, a quarterly business review deck, a QBR presentation, or wants Google Ads results and a plan turned into slides. Trigger phrases include build a deck, proposal deck, pitch deck, QBR deck, slide outline, present these results, turn this report into slides, client presentation, and structure a Google Ads presentation. Pairs with google-ads-reporting, which produces the underlying numbers.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Pick the mode: Use "proposal" to pitch a new or expanded engagement (objective -> baseline -> gaps -> plan -> impact -> ask).

    line 35
  2. Very specific:

    Build the outline: Run python3 scripts/deck_outline_builder.py --demo to see the structure, then --data engagement.json --mode …

    line 40
  3. Distinctive:

    No live connection is required.

    line 62

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Google Ads Presentation

Turn an engagement or a quarter's results into a structured, client-ready deck outline that a slide generator can render. This skill does not draw slides; it produces the ordered argument - each slide with a title, a slide-type, the key data points, a one-line narrative, and the ask - so the story is right before a single pixel is rendered. It covers two modes: a proposal that sells an engagement, and a QBR that accounts for a period.

When to use

Reach for this skill when someone needs a Google Ads proposal or pitch deck, a client-ready slide outline, or a QBR/quarterly-review deck, or wants results and a plan turned into slides. If the request is to compute the numbers first (metrics, winners/losers, recommendations), run google-ads-reporting and feed its output here. This skill is the deliverable layer that sits on top of the report.

Workflow

  1. Gather engagement inputs. Collect client, objective, budget, current results (spend, conversions, ROAS, CPA, period), targets, the proposed plan, and the ask. If a google-ads-reporting JSON exists, lift the totals, winners, losers, and recommendations from it directly.
  2. Pick the mode. Use "proposal" to pitch a new or expanded engagement (objective -> baseline -> gaps -> plan -> impact -> ask). Use "qbr" to review a period with the 4-act arc (where we said we'd be -> where we are -> why the gap -> what next). See references/deck-structure.md for both slide orders.
  3. Build the outline. Run python3 scripts/deck_outline_builder.py --demo to see the structure, then --data engagement.json --mode {proposal|qbr}. It emits structured JSON (consumable by a PptxGenJS renderer) by default, and a readable Markdown outline with --markdown. Each slide carries title, slide_type, data_points, narrative, and ask.
  4. Tighten the narrative. Make every slide title a takeaway, not a topic, so each follows Headline -> Data -> Narrative -> Ask. Lead with the headline or objective; never bury a miss; make the ask specific and assign an owner. See references/narrative-frameworks.md.
  5. Hand off to rendering. The outline JSON maps cleanly to slides: KPI dashboards become tables, projected impact and trends become charts, plans and gaps become numbered or two-column lists. references/pptxgenjs-recipes.md gives the PptxGenJS recipes, chart-styling notes, palette/typography guidance, the corruption pitfalls to avoid, and the required visual-QA loop. Build the renderer separately; this skill stops at the outline.

Inputs and data sources

The skill runs standalone on a hand-written engagement JSON or pasted numbers. No live connection is required.

Optional sources that enrich the deck:

  • google-ads-reporting output (JSON) for the KPI dashboard, winners/losers, and recommendations - the most natural upstream.
  • Google Ads API / UI for the raw current-period results and targets.
  • Google Sheets as the place a client keeps targets and budget.

Outputs

  • A structured deck outline in JSON: an ordered list of slides, each with title, slide_type, data_points, a narrative line, and the ask, plus design notes.
  • The same outline as a readable Markdown document for quick review or sign-off.
  • A clean handoff contract for a PptxGenJS (or any) slide renderer.

References

  • references/deck-structure.md - the proposal vs QBR slide orders, the general rules, and how each slide type maps to a layout.
  • references/narrative-frameworks.md - the Headline-Data-Narrative-Ask per-slide formula, the 4-act QBR arc, how to deliver bad news, and the specific-ask rule.
  • references/pptxgenjs-recipes.md - a distilled how-to for rendering the outline to .pptx with PptxGenJS, including table/chart recipes, chart styling, palette and typography notes, corruption pitfalls, and the visual-QA loop (reference only; do not build the renderer here).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.