Penwright MCP server

• Registry nameink.penwright/discoveryverified in the MCP registry

The registry name sits under penwright.ink. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

• Stateanswerschecked 2026-10-20
• Tools9
• Server calls itselfpenwright-discovery
• Server version1.2.0
• Protocol version2025-06-18

On 2026-10-20 the server answered initialize and tools/list without a sign-in. First answer in 99 ms.

• Whether its tools do the product's job has not been judged yet.
• Checked on 3 of the 7 days it takes to judge reliability.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

• INITE MCP scoreafter a weeka score needs 7 daily checks: 3 so far
• Answeredmeasuring3 of the 7 daily checks it takes to print a share
• Median first answer142 ms
• Space in an agent's context≈ 1,173 tokensthe definitions of all its tools (9), which an agent carries on every turn
• Tool definitionsA · 100 of 100
• Typed results9 of 9 toolsdeclared outputSchema: results come typed, read once

• Every tool passes all six checks.

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (30%), doing the product's job (25%), definitions (20%), context (10%) and use (15%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

• search_library

  Search or list Penwright's writing-craft book library by title/author substring and/or property tag. No filters returns the first page.

• get_library_work

  Returns one writing-craft library work's bibliographic record and (when profiled) its thesis.

• get_guide

  Returns one live guide's title, subtitle, access, book count, and reading time — card-level metadata only, never the paid body/depth content. Known slugs (21): memoir-writing, nonfiction-writing, fiction-writing, getting-published…

• get_magazine_piece

  Returns one published magazine piece's question, per-writer sections, and editor's coda — the same content the human page renders, ungated. Known slugs (1): the-whole-room--my-memoir-keeps-stalling-i-can-write-scenes.

• get_debate

  Returns one debate — a named tension between poles argued by different books on the same capability guide, with stakes, guidance, and the library ids of resolvable authors on either side. Known slugs (38)…

• choose_camp

  Given a guide slug and one pick per debate ('A', 'B', or skip, in debate order, as a single string e.g. 'ABAB'), returns which library ids argue the picked side of each debate and the union reading list. Pure compute over public debate…

• get_answer

  Get one Answer — question, grounded answer, visible citations — by id or slug, in the Answer contract v0 shape (ANSWERS-PROGRAM.md). Read-only. Returns null if not found.

• list_answers

  List Answers (contract v0), paginated with an opaque cursor. Read-only.

• search_answers

  Search Answers by free-text query over question/answer text. Returns contract v0 records. Read-only.