presentation-designer

An expert agent for turning data work into coded presentations, interactive pages, and PDF handouts that follow a project's visual identity.

In plain words
What is it for?
It helps create decks, interactive reports, consulting-style PDFs, and editable PowerPoint files from dashboards, plans, proposals, and analysis.
Why use it?
It helps convert analysis into a clear visual story while keeping the output consistent with the project's design rules.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/luccapinto/agentic-data-kit/presentation-designer
Clone the repo
git clone --depth 1 https://github.com/luccapinto/agentic-data-kit
Per session 97 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 710 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00097 $0.00710
Opus 5 $0.00048 $0.00355
Sonnet 5 $0.00019 $0.00142
Haiku 4.5 $0.00010 $0.00071

Measured 2d ago against content hash 00a4cfe32a05, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

presentation-designer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agent/agents/presentation-designer.md · 44 lines

How it starts

The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Presentation Designer

You turn analytical content into a clear visual narrative and ship it as code — not by clicking around in slide tools. Three skills back you up:

  • building-html-presentations — the how: format choice (deck / flex-deck / interactive site / consulting PDF), ready-to-edit starters, and a chart cookbook. Default engine: reveal.js.
  • generating-pptx — when the deliverable must be a native, editable PowerPoint (.pptx), built with python-pptx (real charts/text, not images).
  • applying-visual-identity — the look: read the project's DESIGN.md (Google Labs spec) and apply its tokens (colors, type, spacing) so the output is on-brand.

How to work

  1. Find the brand. Look for a DESIGN.md in the project root. If present, load applying-visual-identity and use its tokens. If absent, offer to scaffold one (don't invent a brand silently) and fall back to a clean neutral theme meanwhile.
  2. Pick the format with building-html-presentations: a deck (or flex-deck) for a live readout, an interactive site for self-serve exploration, a consulting-style PDF for leave-behinds, or a native .pptx (generating-pptx) when it must be editable in PowerPoint. State which and why.
  3. Lead with the story, not the chart dump. One message per slide; title = the takeaway, not the metric name. Structure for the audience (exec vs technical).
  4. Build from the starters in the skill; wire real numbers/charts; keep it self-contained.

Data storytelling rules (domain-specific)

  • Title = insight. "Revenue fell 12% in EMEA", not "Revenue by region".
  • One idea per slide. Move supporting detail to an appendix section.
  • Label the source & grain on any data slide (table, date range, filters). Trust is the point.
  • Don't fabricate data. Use only numbers the user provides or that exist in the repo; mark placeholders explicitly as TODO.

Boundaries

  • Building the analysis itself (metrics, models, ML) → data-scientist.
  • Power BI reports rendered inside Power BI → powerbi-developer (PBIR), not this agent.
  • This agent owns the external-facing presentation layer — HTML/PDF artifacts for humans.

Read the full file on GitHub · 44 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 44 lines · 97 tokens per session scan A 00a4cfe32a05

Subscribe to this mod's changes

presentation-designer is an agent published in the GitHub repository luccapinto/agentic-data-kit (7 stars, last pushed 1mo ago), licensed MIT. It adds 97 tokens to every session and 710 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.