teneo-book-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@teneo-book-mcp Write a 10-chapter book on habit building"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
teneo-book-mcp
MCP server for Teneo. Lets any MCP-capable agent — Claude Desktop, Cursor, Claude Code, or your own — generate a complete, publish-ready non-fiction book: full manuscript, cover, and PDF/EPUB/DOCX files.
{
"mcpServers": {
"teneo-book": {
"command": "npx",
"args": ["-y", "teneo-book-mcp"],
"env": { "TENEO_API_KEY": "tk_live_..." }
}
}
}Tools
Tool | What it does |
| Discover auth, pricing, request shape, workflow, and output shape. Start here. |
| Start a book (async) → returns a |
| Poll a job; when |
| Generate an outline only, without running the full pipeline. Free. |
| Niche research from a seed topic. Free. |
| Generate a cover only. |
| Run quality evaluation on an existing generated book. |
Related MCP server: context-book
What it costs
Priced in Teneo credits. One credit is $69 singly, $59 each in a 3-pack, $44.90 each in a 10-pack; credits never expire.
requiredCredits = bookCredits + coverCredits + qualityCreditsReserve
bookCredits = wordBandCredits(estimatedWordCount) x modelMultiplier(model_version)
estimatedWordCount = number_of_chapters x topics_per_chapter x subtopic_count x 350A default request (12 chapters / 3 topics / 3 subtopics) estimates 37,800 words → 1.05
credits, about $72 at list. teneo_generate_outline and teneo_research_niche cost
nothing, so validate direction before committing to a paid generation.
A funded Teneo balance is required before the first paid call — there is no
pay-per-call endpoint yet. The live numbers, the full model-multiplier table, and the
request schema are always at
/api/agent/capabilities, which is
authoritative if anything here disagrees with it.
Setup
Get a credential, then add the config block above to your MCP client.
Preferred (delegated): start at
auth.teneo.io/auth.md, register forhttps://teneo.io/, have the user approve the claim, and exchange it for a delegated agent token. Pass it asTENEO_AGENT_TOKEN. The delegated user is charged.Legacy: create an API key in your Teneo dashboard and pass
TENEO_API_KEY=tk_live_....Point at staging with
TENEO_BASE_URL=https://staging.teneo.io.
Spend semantics
Worth knowing before you wire this into an autonomous loop:
Quoted before charged. A request that would exceed the balance returns
402with the full quote and charges nothing — a safe way to read the exact price.Idempotent. Send an
Idempotency-Keyheader; replaying a request replays the original debit instead of charging twice.Refunded on failure. A failed generation is refunded in full.
Async. A book takes ~45 minutes. Poll
teneo_book_status, or passcallback_urlfor a completion webhook. Download URLs expire in ~1 hour; re-poll for fresh ones.
Not for
A single short answer, a blog post, or an article — answer those directly instead. Fiction and novels are not supported by this API.
Develop
npm test # unit tests for the HTTP client (mocked fetch, no network)
npm start # run the MCP server over stdioclient.mjs is a dependency-free HTTP client; server.mjs wraps it as MCP tools. Both
mirror the published
OpenAPI contract.
Links
Agent front door: https://teneo.io/agents
Capability manifest: https://teneo.io/api/agent/capabilities
OpenAPI: https://teneo.io/openapi/agent-book-generation.yaml
Also installable as an agent skill:
npx skills add https://teneo.ioOfficial MCP registry:
io.github.Traviseric/teneo-book-mcp
License
MIT
Available Tools
7 toolsteneo_book_statusA
Check a book generation job. When status is "completed", the response includes outputManifest with coverUrl and time-limited download URLs for each format. Poll every ~30-60s.
Inputs: jobId (required).
teneo_capabilitiesA
Discover what Teneo can generate and how: auth, pricing (credits), request fields, the workflow, use cases, and the output shape. Call this first if unsure.
Inputs: no named parameters.
teneo_evaluate_bookC
Run a Teneo evaluation on an existing generated book.
Inputs: fullBookId (required), forceRestart, evaluationMode, rewriteVersion.
teneo_generate_bookA
Generate a COMPLETE, publish-ready non-fiction book (full manuscript + cover + PDF/EPUB/DOCX) from a topic/spec. Asynchronous: returns a jobId immediately (~45 min to finish). Then poll teneo_book_status. Charged in Teneo credits to the funded account. Use for a finished book/manuscript or KDP-ready files — not for a short answer or fiction.
Inputs: title (required), description (required), target_audience (required), author_name, +4 more.
teneo_generate_coverC
Generate a book cover only. This charges the account for cover generation credits.
Inputs: title (required), mood, style, theme, +2 more.
teneo_generate_outlineB
Generate an outline candidate for a nonfiction book without running the full book pipeline.
Inputs: title (required), description, target_audience, number_of_chapters.
teneo_research_nicheC
Start a niche research job from a seed topic.
Inputs: topic (required), count, audience, category.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 7 tools
The tools target distinct actions: capabilities (discovery), generate_book, research_niche, book_status, generate_outline, generate_cover, evaluate_book. The main ambiguity is generate_cover vs generate_book (which already produces a cover) and whether book_status also tracks the research_niche job, but descriptions largely clarify these.
All tools share a teneo_ prefix and follow a mostly verb_noun convention (generate_book, research_niche, evaluate_book). Minor deviations: teneo_capabilities is noun-only and teneo_book_status reverses to a noun_status pattern rather than the verb_noun style.
Seven tools is well-scoped for a book-generation service, covering discovery, generation (book/outline/cover), status polling, research, and evaluation. Each tool earns its place without obvious redundancy.
The surface covers the core lifecycle: discover, generate (book/outline/cover), poll status with output manifest, research niche, and evaluate. Minor gaps: no dedicated status/cancel for the research_niche job and no job listing/history, but agents can work around these.
Maintenance
Related MCP Connectors
Agentic AI book creator: plan, write, illustrate, narrate, and publish books from any MCP client.
Build and publish full-stack apps from your coding agent: models, rules, pages, auth, per-app MCP.
Chinese web novel MCP: 36 tools (outline, prose, review, coach, KD export). BYOK, no API key.
MCP book-writing server: 81 classics, genre recipes, drafting pipelines.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceProvides MCP tools for AI-powered code review, research, and book writing via a fleet of coding agents.Apache 2.0
- AlicenseNot gradedqualityDmaintenanceProvides AI agents with a persistent, searchable knowledge library via MCP tools, allowing them to create books, manage pages, perform semantic search, and retrieve usage guides.5MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI-driven long-form novel creation and management, including chapter generation, character and timeline tracking, semantic memory retrieval, version savepoints, and deep consistency checking across multiple projects through MCP tools, with support for local LM Studio or any OpenAI-compatible API.16MIT
- FlicenseNot gradedqualityCmaintenanceExposes a fleet of specialized publishing and marketing agents — including manuscript authoring, ebook production, editorial, market research, cover direction, sales copy, and knowledge/Drive management — as MCP tools served over FastAPI, one server per agent. Each tool calls shared local Ollama models, so clients can drive end-to-end book creation, launch strategy, and analytics workflows through natural language.-