Curated consulting and case competition slide deck generation for Claude, ChatGPT, etc.
Scored across 5 tools
The two find_* searches target clearly different granularity (deck-level flows vs. slide-level layouts), and descriptions explicitly state when to use each and in what order. The only mild ambiguity is between find_references and get_reference, but the descriptions ('search by what a slide must do' vs. 'one library slide by id') resolve it.
All five tools follow a consistent snake_case verb_noun pattern (find_deck_flows, find_images, find_references, get_playbook, get_reference). Search vs. fetch verbs are used predictably: find_* for queries, get_* for single-item retrieval.
Five tools map exactly onto the distinct phases of the workflow (methodology, deck-flow search, slide search, single-reference fetch, image sourcing). Nothing is redundant and nothing feels padded; the set is tight and well-scoped.
The retrieval pipeline is covered end to end: get_playbook gives the method, find_deck_flows and find_references supply search, get_reference expands a hit, and find_images fills visual slots. Minor gaps exist — no tool to fetch a full playbook in one call or to validate/assemble a finished deck plan — but these are workaround-able by the agent.
Use after the deck's end belief and style are settled, to inform the storyline in the text plan (see get_playbook). Searches the library's deck flows (how strong decks order their sections) by what the whole deck must achieve, before find_references. Returns up to 3 summaries (deck id, arc, length, industry, style, section sequence, why the flow works, score). Call get_reference with a deck id for its steps and their slide ids.
Inputs: purpose (required), style, length, industry.
Use while building an approved slide to source a logo, icon or photo for an image slot (see get_playbook). Searches public-domain logo SVGs (Wikimedia Commons, Simple Icons), permissive icon SVGs (Iconify) and CC0, public-domain and CC BY photos (Openverse). Returns up to k results, each with its title, licence and either the SVG text or the photo's download and page links, plus a credit line only when the licence requires one; photo thumbnails follow as images.
Inputs: kind (required), query (required), k.
Use after the slide's claim is settled, to inform the layout options in the text plan (see get_playbook). Searches the curated library of slide design DNA by what a slide must do. Returns up to k short summaries (id, slide type, score, purpose, why it works, the layout's component kinds, impact device if any, style, when to use it), at most one per deck; an impact filter returns few results. Call get_reference with an id for the full layout. Reference, never copy.
Inputs: purpose (required), k, style, impact, +3 more.
The 80-20 slide-design methodology for case competition decks. Call it before planning a deck. With no stage it returns the index of stages and the workflow; pass a stage for one chapter.
Inputs: stage.
One library slide: its labels (purpose, why it works, archetype, insight device, impact), its component layout (v3), its design DNA (the 40 largest region boxes with colour and type roles, text structure) and its mockup image. Use an id from find_references. Build your own slide from it; never copy it. A deck id from find_deck_flows returns that deck's flow instead: its steps in order, each with a slide id and whether that slide is in the library (available).
Inputs: id (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Add one secure layer between your agents and this server.
