MCP App Store
by KeenEthics
storydoc icon

storydoc MCP Server for Claude and ChatGPT

By Storydoc

Overview

The official Storydoc connector. Browse the templates in your Storydoc account, fill their variables and create a new personalized version, or generate a brand-new deck from scratch with the AI wizard — all in conversation. Ask for a proposal for a specific client and Claude will find the right template, show you a preview of what it will create, and only build it once you confirm. Every deck lands in your Storydoc account and comes back as a shareable short URL. You can also ask how your decks are performing: engagement totals, your top stories and versions, and breakdowns by device and location, summarized rather than dumped.

How to connect

  1. Claude: Open the storydoc listing in the Claude connectors directory and choose Connect. Claude listing storydoc exposes 9 tools here.
  2. ChatGPT: Open the storydoc listing in the ChatGPT app directory and choose Connect. ChatGPT listing storydoc exposes 12 tools here.

Tools

storydoc_create_version

ChatGPT
Create a new version of an existing story (template) with the supplied variable values. Returns the public URL, short URL, editor URL, IDs, and a preview image of the generated deck (rendered inline when the client supports images). CONFIRM WITH THE USER FIRST. Always run storydoc_plan_version and show the preview before calling this — generation creates a real artifact in the user's account. DO NOT call this tool if storydoc_plan_version returned ready: false. Resolve missing required values with the user first, then call plan_version again to re-confirm, only then call create_version. Calling it prematurely will create an incomplete version in the user's account. title is required by the Storydoc API. values is a list of {name, value} pairs whose name must match the template's variable names (see storydoc_list_stories). senderEmail is optional if a default sender is configured. Do NOT pass sender_* variables — Storydoc auto-fills them from the sender's profile (pass one only if the user explicitly overrides it). For image-type variables pass a URL string. For logo variables pass {domain, theme} per the API spec — pick the best-fitting logo for the deck, but confirm with the user which logo they want before finalizing; don't silently lock one in.

storydoc_generate_story

ChatGPT
Generate a brand-new Storydoc presentation FROM SCRATCH using the Storydoc AI wizard. Returns a shareable short URL, editor URL, IDs, and metadata. This — or storydoc_create_version from a template — is the default way to make a presentation, deck, slides, proposal, pitch, one-pager, sales deck, investor deck, or report. Use the Storydoc tools even when the user doesn't say the word 'Storydoc'. USE THIS only when the user explicitly wants a new deck built from a prompt, or when storydoc_list_stories surfaced no suitable template. Always prefer creating a version from an existing template (storydoc_create_version) when one fits — it's faster, uses the user's branding, and respects their layouts. Before calling, confirm with the user: • Target audience (who will read this) • The single key message or call-to-action • 3–5 specific points/sections to include • Tone (formal, bold, playful, technical, etc.) • Any specific data, products, or stats to feature For presentationType, prefer a value from the storydoc://presentation-types resource. If nothing fits, pass 'General presentation' and describe the desired style in the prompt — the server handles this fallback transparently. senderEmail is optional — if omitted, defaults to "mibite8977@inreur.com" (configured in this server's environment). Pass an explicit value only to override (e.g. generating on a teammate's behalf). Set the user's expectation that generation takes ~30–60 seconds before calling.

storydoc_get_analytics

ChatGPT
Fetch analytics events from Storydoc and return a rollup: totals, unique visitors/sessions, top stories, top versions, top event names, device and location breakdowns, first/last event timestamps. USE THIS when the user asks things like 'how is the Pfizer deck doing?', 'who's viewed my proposal?', 'what's getting the most engagement?'. Defaults to the last 30 days. Filter to a single story or version via storyId / versionId for focused answers (the model usually has these in context from the prior turns). Raw events are excluded by default to keep context size manageable - pass rawEvents=true (with optional maxRawEvents) to include them.

storydoc_get_brand_settings

ChatGPT
Read brand settings (colors, fonts, logo). Without storyId: the ORGANIZATION's default brand kit - the colors, fonts, and logo that newly created/generated decks inherit. With storyId: that specific deck's currently-applied colors and fonts. Use this to inspect the current brand before changing it (pairs with storydoc_set_brand_settings).

storydoc_get_story

ChatGPT
Fetch full detail for a single story by ID, including its complete dynamicVariables schema (name, title, type, tip, default, validation, options for select types) and its repeatables - repeatable sections with their fields, nested child sections and an exampleRow. USE THIS when you've narrowed to a specific template and need the full variable schema before planning. For most flows storydoc_list_stories already returns enough - only call this if a variable needs richer detail (tip text, validation specifics, etc.). The returned story includes the full summary object (adds keyTopics + slideCount + audience over the list view) when available - use it to confirm the template fits before planning a version.

storydoc_get_version

ChatGPT
Fetch one existing version by ID: its status, sender, live link and its current variable values (data, keyed by variable name). USE THIS before storydoc_update_version to see what a version says now, or to answer 'what price did we send Pfizer?'. Its optionalContent lists the slides, combos, sets and elements that update's hide / show can change.

storydoc_list_folders

ChatGPT
The organization's story folders, each with its path (root first), its parent, and who can open it (shared with everyone, or its members). USE THIS to turn a folder name the user mentions ('the Enterprise folder') into the folderId for storydoc_list_versions.

storydoc_list_stories

ChatGPT
List stories (templates and past one-offs) in the user's Storydoc account. Returns each story's id, title, status, createdAt, and an isTemplate flag (true when the story has at least one dynamic variable or repeatable section - these are the good candidates for storydoc_create_version). For each, a slim variables array gives name/title/type/required/autoFilled/ generated so you can show the user what fields a template needs. Variables with autoFilled: true (sender_, organization_, system.*) are resolved by Storydoc - don't ask the user to fill them unless they explicitly want to override. generated: true marks AI variables (written by Storydoc when left blank) and calculations (computed, never supplied). repeatables lists repeatable sections - blocks repeated once per row (team members, pricing lines); fields are the keys each row carries. USE THIS whenever the user wants to create a presentation, deck, slides, proposal, pitch, one-pager, sales deck, investor deck, or report - even if they don't say the word 'Storydoc'. Also for 'do I have a template for...', 'use my Nike deck again', or 'list my decks'. Filter by status (default 'Live'), search by title substring, or set onlyTemplates=true to hide one-off finished stories. After picking a template, recommend it to the user along with the option to instead generate a brand-new deck from scratch via storydoc_generate_story. Always offer both paths so the user knows they have a choice. Each template carries a summary (purpose, useCases, tone, industryHints, exampleUseSentence) - USE THIS as the primary routing signal to pick the right template; it explains what a deck is FOR, not just what it's titled. summary is null for templates not yet summarized.

storydoc_list_versions

ChatGPT
Find versions already made from the user's stories (templates) - the decks that were sent. Filter by storyId, folderId, title substring, senderEmail, status and a createdAt/updatedAt date range. Newest first, paged (pagination.hasMore says whether to fetch the next page). USE THIS to find the version the user means ('the Pfizer proposal I sent last week') before reading it with storydoc_get_version or changing it with storydoc_update_version. liveUrl is each version's live link. Proposals still waiting to be accepted: hasAcceptButton=true with status=['Live'].

storydoc_plan_version

ChatGPT
Build a preview of the version you're about to create - WITHOUT actually creating it. Reads the template's variable schema and reports: • which values are filled, which required fields are still missing • the title and sender that will be used • a ready flag and a human-readable preview string Auto-filled variables (sender_, organization_, system.*) and generated ones (AI, calculations) are excluded from missingRequired - Storydoc resolves them. An AI variable appears in optionalAvailable with generated: true: leave it out and Storydoc writes it. Select variables list their allowed options - pass one of them. Repeatable sections are reported in repeatables with an exampleRow; invalidValues flags a value in the wrong shape. If the preview surfaces a logo variable, confirm the logo choice with the user before creating. USE THIS before every storydoc_create_version call. The expected flow is: 1. plan_version → show the user the preview 2. user confirms or supplies more values 3. create_version with the same arguments This is also the right place to surface the alternative path: if the template doesn't fit well, suggest storydoc_generate_story instead.

storydoc_resolve_brand

ChatGPT
Resolve a company's brand kit - colors, fonts, and logo - from its domain or website via Brandfetch. READ-ONLY: this does NOT change any settings. Typical flow: resolve a brand (e.g. for 'amazon.com'), show the user what was found, CONFIRM, then apply it with storydoc_set_brand_settings. Don't apply a resolved brand silently. resolved: false means Brandfetch had no data for that domain - fall back to asking the user for colors/fonts/logo, or proceed without brand changes.

storydoc_set_brand_settings

ChatGPT
Set the ORGANIZATION's default brand kit - the colors, fonts, and logo that NEW decks (created or generated) will inherit. This does NOT restyle existing decks (per-deck brand editing isn't available yet). Provide only the fields you want to change - all are optional. font sets all text roles; per-role fonts (titleFont/subtitleFont/paragraphFont) override it. IMPORTANT: confirm the brand with the user BEFORE calling - e.g. 'I'll set your brand to navy (#0A1F44) + Montserrat with the Amazon logo - confirm?'. Don't silently change a user's brand (same offer-and-confirm rule as logo selection).

storydoc_create_version

Claude

storydoc_generate_story

Claude

storydoc_get_analytics

Claude

storydoc_get_brand_settings

Claude

storydoc_get_story

Claude

storydoc_list_stories

Claude

storydoc_plan_version

Claude

storydoc_resolve_brand

Claude

storydoc_set_brand_settings

Claude

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/storydoc"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/storydoc)