
Proseify
io.github.DionisselamiPublished by GitHub account Dionisselamiv1.0.0Updated Oct 24, 2026
MCP book-writing server: 81 classics, genre recipes, drafting pipelines.
Overview
Hosted MCP server that turns an assistant into a book-writing engine: genre recipes, public-domain corpus search, outlining, drafting and scoring.
- What you can do
- Browse genres and search a public-domain book corpus with full-text queries, plus retrieve chapter and corpus metadata for reference passages and register examples drawn from a chosen genre's tradition, including pacing beats and dialogue ratios, to plan a chapter-by-chapter outline from a one-line premise, draft a whole book in one flow, and score the draft against genre benchmarks for revision.
- When to use it
- Use it when you want an assistant to produce a structured, editable manuscript rather than a single generic generation pass, and you want drafting grounded in genre conventions and real literary passages.
- Requirements
- Remote streamable HTTP endpoint at mcp.proseify.xyz, connected with an OAuth sign-in or a bearer key issued after signing in and choosing a paid plan. The README also describes installing a companion writing skill via npx.
Installation
In SourceWeft
- Open Proseify in the dashboard and add it to a workspace.
- Connect the required credentials. They are stored privately and never shown on public pages.
- Enable the server for the chats that should use its tools.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Use with an agent
Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.
Read https://sourceweft.com/mcp/io.github.Dionisselami%2Fproseify-mcp/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.Other MCP clients
Add this to your client's mcpServers config. The client completes OAuth sign-in on first connect.
{
"mcpServers": {
"proseify": {
"type": "http",
"url": "https://mcp.proseify.xyz/mcp"
}
}
}README
Proseify — the book-writing engine for AI agents
Proseify is a hosted MCP (Model Context Protocol) server that turns Claude Code, Codex, Cursor, or any MCP client into a disciplined book-writing machine — grounded in real literature instead of generic model defaults.
Give it a one-line premise. It picks the right genre recipe, pulls model passages from a curated public-domain library, plans the outline, drafts chapter by chapter, and scores its own work. You get a manuscript you can actually edit, not 40 pages of "it was a dark and stormy night."
🔗 Website & key issuance: https://proseify.xyz
[Listed on mcpservers.org] [M8ven Score]
Why genre-before-prose matters
Most "AI writes a book" tools generate immediately. The output reads the same way regardless of what you asked for — the same sentence lengths, the same dialogue rhythm, the same endings.
Proseify flips the order:
The genre recipe is derived from the actual tradition — pacing beats, dialogue ratio, sentence-length profile, how endings are handled. The agent reads it before touching prose. That single ordering decision is the difference between a horror book that reads like the gothic tradition and a horror book that reads like ChatGPT.
Tools exposed
…plus supporting tools for chapter retrieval and corpus metadata.
Setup (Claude Code, Codex, Cursor, and friends)
1. Get a key
Sign in at https://proseify.xyz (Google / GitHub / magic link), pick a plan. Monthly from $9, or one-time Founding Lifetime. Your key is issued the moment the purchase clears.
2. Connect the MCP server
(Use the client's native MCP config format for your agent — see the Agent SDK page for a filled-in block.)
3. Install the skill
The anti-prose-slop skill gives the agent the writing workflow — genre
recipe first, corpus grounding, chapter discipline, and a per-chapter
anti-slop editing pass — instead of just raw tool access.
4. Write
Tell the agent a premise:
"I want a gothic horror novel about a lighthouse keeper finding a body that shouldn't exist. 32 chapters. Budget 50,000 words."
The agent picks the right recipe, searches the corpus for register, plans, drafts, scores, and revises. You edit what it returns.
The corpus
a growing library of books / 11 genres, drawn from Project Gutenberg and similar public-domain collections — all verified against their original headers, so there is no copyright ambiguity anywhere in the library. Commercial use of output built on it is safe.
A partial list: works from Austen, Stevenson, Hugo,conrad, Verne, the Brontës, Shelley, gothic masters, and more — selected, grouped by genre, and indexed for search.
Why "anti-prose-slop"?
Because the failure mode of AI-written prose isn't bad grammar — models are
word-perfect. It's sameness: "seemed to", "began to", stacked adverbs,
repeated sentence constructions, filter words, dialogue-as-exposition,
chapter openings that always start with weather. The skill that ships with
Proseify has a concrete per-chapter checklist against each of these, and
evaluate_book scores the result against genre norms so the agent can
self-revise rather than ship its first pass.
Pricing
Cancel any time from your account (opens our payment provider's customer portal, signed in as you). See the refund policy for the 14-day money-back window.
License
MIT — see LICENSE. The public-domain corpus texts themselves are public domain (mostly Project Gutenberg-sourced).
Links
- Site / sign-up: https://proseify.xyz
- Agent SDK (filled-in config for your key): https://proseify.xyz/agent
- Terms · Privacy · Refund policy — footer of every page on the site
- Support: [email protected]
- Published in the official MCP registry:
io.github.Dionisselami/proseify-mcp
Source: README.md at commit 58c0cd3
Tools
0Version history
1- v1.0.0LatestSep 18, 2026
Score
85% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 daysRequired25 pts
- Auth declaredRequired10 pts
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts