MCP App Store
by KeenEthics
WOIII.me icon

WOIII.me MCP Server for ChatGPT

By WOIII.me

Overview

WOIII.me finds the best pressing of a record on Discogs, for sound or for collectors, and shows the evidence behind each score: mastering engineers, matrix and runout stampers, pressing plants, label reputation, format, community ratings and demand. Ask about one pressing (paste a Discogs release link), survey an album's pressings, compare the ones you are deciding between, or rank your own wantlist by how well each album fits your collection. Open the WOIII.me home for the same workflows on one page. Read-only: WOIII.me never changes anything on Discogs, and it shows no prices or cover images. Scores read Discogs data; they do not measure sound. Data provided by Discogs; WOIII.me is not affiliated with Discogs.

How to connect

  1. ChatGPT: Open the WOIII.me listing in the ChatGPT app directory and choose Connect. ChatGPT listing WOIII.me exposes 13 tools here.

Tools

auth_status

ChatGPT
Show which Discogs account this session is connected to. The only tool that returns the username.

compare_pressings

ChatGPT
Side-by-side comparison of 2–8 specific pressings by release ID along a chosen axis: mastering pedigree & signals, format, ratings (incl. delta vs. the set average), collector demand, and overall evidence-weighted scores for both axes — each as a full evidence dossier. No prices. One Discogs call per uncached release.

find_best_pressing

ChatGPT
Find the best pressing of an album along a chosen axis. Identifies the master release, reads its version list for demand signals, and fetches full details for a small stratified candidate set (5 by default, up to 8) that ALWAYS includes audiophile reissues (Mobile Fidelity, Analogue Productions, Tone Poet, etc.) plus the most in-demand pressings. Each candidate is scored from mastering pedigree (reputable label by id, renowned engineer credits, matrix/runout stamper marks, pressing studio), format/medium quality, collector demand, and how its community rating compares to the album baseline — evidence-weighted so missing data doesn't penalise a pressing. No prices. Returns an evidence dossier per pressing (signals, mastering credits, matrix/runout, a provisional verdict, evidenceCoverage, scores for both axes and, when enabled, catalogClaims read from the notes), plus honest counts: the result is the best among the candidates checked, not every pressing. Optional filterCountry / filterFormat / yearFrom / yearTo narrow the survey (e.g. US vinyl 1970–1979). Costs about 8–12 Discogs calls cold; results are cached, and switching axis afterwards is free. Recommends vinyl only unless the user's settings allow CD and SACD (media in the result says which; a filterFormat argument overrides it, and mediaNote explains a fallback). For a discogs.com/master/<id> link pass that number as masterId; for a discogs.com/release/<id> link pass releaseId.

get_collection_stats

ChatGPT
Analytics over the authenticated user's collection: genre/style/decade/format/label distributions, rating stats, a taste-profile summary, and shelf: the collection in the side panel's terms (size, leading genres and styles, decade shares, format split, house labels, records added per month, recently added). The first call loads the collection (about 1 Discogs call per 100 records, cached for 4 hours); later calls are free.

get_release

ChatGPT
The verdict for one specific Discogs release (one concrete pressing/edition): community ratings, have/want, formats, labels, notes, the structured pressing evidence — matrix/runout stampers, mastering credits, pressing companies, a pedigree assessment — and a provisional verdict with scores on both axes (sonic, collector). No prices or images. One Discogs call when uncached. Call it directly for a discogs.com/release/<id> link (that number is releaseId): it returns the release's Discogs data, so no web search is needed to read the page. preferredLens, when present, is the lens the user chose in settings: lead with that axis's score. scoredAgainst says whether the scores compare this pressing with its album's surveyed pressings ('album', the same scores a survey shows) or stand alone ('pressing'). owned, wanted and tasteFit come from the user's already-loaded collection and wantlist at no extra call; each is null while that list is not loaded (unknown, not no).

get_release_versions

ChatGPT
List pressings/versions of a master release with optional country, year-range and format filters (filters are exact/inclusive and fetch deeper into long version lists). Rows carry label, catno, country, released, format and majorFormats, community demand (inCollection / inWantlist = how many Discogs users have / want it, not the caller) and inYourCollection (the caller's own collection; null when it isn't loaded), but NO ratings, runouts or credits (Discogs API limitation). totalVersions is every version Discogs lists; versionsLoaded is how many were read. To inspect filtered versions in detail — runouts, mastering credits, scores — call find_best_pressing with the same filters (candidates up to 8) instead of fetching releases one by one.

open_home

ChatGPT
Open the WOIII.me home page (workflows and tools). Takes no arguments and makes no Discogs requests: it shows only what is already cached.

rank_wantlist

ChatGPT
Rank the authenticated user's ENTIRE wantlist by taste fit in ONE call — server-side, deterministic, identical across clients. Groups alternate editions of the same album, scores each group against the collection's style/genre/decade profile (the same scale as tasteFit), assigns tiers (bullseye / good fit / off-profile), flags titles already in the collection, and returns the method and caveats. Use this instead of computing your own ranking from get_collection_stats or search_collection.

search_collection

ChatGPT
Search the authenticated user's Discogs collection. Free-text queries are mood-aware: queries like 'mellow sunday morning' map to matching genres/styles. Explicit genre/style/decade/rating filters can be combined with or used instead of the query. The first call loads the collection (about 1 Discogs call per 100 records, cached for 4 hours); later calls are free.

search_discogs

ChatGPT
Search the full Discogs catalog (releases, masters, artists, labels). Release results are marked inCollection when the user's collection is already loaded (null when it is not; search never loads it). Use it when you have a title, not a Discogs link: a release link goes straight to get_release, a master link to find_best_pressing.

settings_read

ChatGPT
Read the user's WOIII.me settings: the default lens (sonic or collector) for surveys and comparisons, how many pressings an album survey checks (5–8), whether surveys also recommend CD and SACD (default: vinyl only), and whether explanatory notes are shown. Returns the settings schema, current values and layout. No Discogs call.

settings_reset

ChatGPT
Delete the user's saved WOIII.me settings and go back to the defaults (sonic lens, 5 pressings per album, vinyl only, notes shown). Removes only WOIII.me preferences, nothing on Discogs.

settings_update

ChatGPT
Change one or more WOIII.me settings: lens ('sonic' | 'collector'), detailBudget (5–8; each pressing above 5 costs one more Discogs request per album survey), includeCd (true to also recommend CD and SACD; default false, vinyl only) or showCaveats (true/false). Send only the keys to change in set; returns every current value. Changes only WOIII.me preferences, never anything on Discogs. An explicit argument to a tool still beats a setting.

Capabilities

InteractiveWrites

Example Prompts

Click any prompt to copy it.

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/woiii-me"><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/woiii-me)