Every change Wellknown observed on this MCP server, newest first, with what it was before and what it became. Tool-surface changes carry the definition diff. Nothing here is edited after the fact.
Changed the definition of "list_professions", "recommend_model_for_profession" and "search_all"
⟨42 unchanged words⟩ ,"name":{"type":"string"},"not_measured":{"description":"What no benchmark in the catalogue evaluates at all. Applies even when `unmeasured_tasks` is 0.","type":"string"},"resource_uri":{"type":"string"},"unmeasured_tasks":{"description":"How many of its listedtasksTASKS no benchmark measures yet. Zero is complete task coverage, not a complete measurement — `not_measured` is a different axis and applies anyway.","type":"number"}},"required":["id","name","group","evaluation","first","unmeasured_tasks","not_measured","canonical_url","dataset_version","resource_uri"]," ⟨9 unchanged words⟩
⟨12 unchanged words⟩ ,"caveat":{"type":"string"},"coverage":{"additionalProperties":false,"properties":{"dimensions_not_measured":{"description":"What no benchmark evaluates at all — independent of task coverage.","type":"string"},"how_to_present":{"description":"Follow this when quoting the limits: it says which of the two axes is complete and which is not.","type":"string"},"tasks_listed":{"type":"number"},"tasks_measured":{"type":"number"},"tasks_unmeasured":{"type":"number"}},"required":["tasks_listed","tasks_measured","tasks_unmeasured","dimensions_not_measured","how_to_present"],"type":"object"},"dataset_version":{"type":"string"},"evaluation" ⟨35 unchanged words⟩ {"type":"string"},"not_measured":{"description":"Dimensions no benchmark in the catalogue evaluates at all. A different axis from `unmeasured_tasks`, which is about this profession's listed tasks; this one applies even when that list is empty.","type":"string"},"picks":{"additionalProperties" ⟨175 unchanged words⟩ ,"type":"array"},"unmeasured_tasks":{"description":"The listed tasks with no benchmark behind them. An empty list means every task is covered, not that the ranking measures everything that matters — see `coverage`.","items":{"type":"string"},"type" ⟨6 unchanged words⟩ ","resolved_locale","evaluation","tasks","unmeasured_tasks","coverage","picks","ranking","caveat","not_measured ⟨7 unchanged words⟩
⟨4 unchanged words⟩ "content_locales":{"description":"Keep only resultsactuallywhosewrittentext is entirely in these languages. `locale` is a presentation preference ⟨5 unchanged words⟩ this is the hard filter. A result thatcannot declaremixesitsa language you did not allow isdropped.dropped, and so is one that cannot declare its languages.","items":{"type":"string"}, ⟨88 unchanged words⟩
⟨44 unchanged words⟩ {"type":"string"},"content_locale":{"anyOf":[{"type":"string"},{"type":"null"}],"description":"The single
Changed the definition of "search_all"
{"$schema":"https://json-schema.org/draft/2020-12/schema","properties":{"content_locales":{"description":"Keep only results actually written in these languages. `locale` is a presentation preference and still returns substitutions, labelled; this is the hard filter. A result that cannot declare its language is dropped.","items":{"type":"string"},"minItems":1,"type":"array"},"limit_per_surface":{"description":"Maximum hits from each ⟨78 unchanged words⟩
⟨6 unchanged words⟩ {"count":{"type":"number"},"empty_reason":{"description":"Why a strict content_locales filter returned nothing.","type":"string"},"query":{"type":"string"},"requested_locale":{"description":"The locale this search asked for.","type":"string"},"results":{"items":{"additionalProperties":{}, ⟨7 unchanged words⟩ ,"canonical_url":{"type":"string"},"content_locale":{"description":"The language this result is actually in.","type":"string"},"
Added "compare_models", "get_benchmark", "list_benchmarks" and 2 more; changed the definition of "get_overview" and "search_all" (31 tools before, 36 now)
⟨12 unchanged words⟩ plus the separate Article, Labs, Homeric Atlas, claim-registry andclaim-registryAI-benchmark surfaces, with their tools, identifiers, citation ⟨6 unchanged words⟩
Search the core corpus, first-party essays, executable Labs, epistemic claimsand, the Homeric Atlas and the AI benchmarks (professions and benchmark catalogue)
Added "get_author_profile"; changed the definition of "get_architecture", "get_governance", "get_handbook" and 9 more (30 tools before, 31 now)
⟨59 unchanged words⟩ :"string"},"type":"array"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"knowledge" ⟨94 unchanged words⟩
⟨59 unchanged words⟩ :"string"},"type":"array"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"knowledge" ⟨94 unchanged words⟩
Certificate changed, valid to 2027-01-02
Certificate recorded, valid to 2026-10-31
Authorization not required
First tool surface recorded: 30 tools (server version 0.5.0)
Showing the latest 8 events. The API returns up to 500 and filters by kind: ?kind=tool_surface_changed
⟨61 unchanged words⟩ :"Restrict the search. Omit to search allfivesix surfaces.","items":{"enum":["core","articles","labs","claims","homeric_atlas","benchmarks"],"type":"string"},"minItems": ⟨7 unchanged words⟩
⟨67 unchanged words⟩ ,"articles","labs","claims","homeric_atlas","benchmarks"],"type":"string"},"title":{ ⟨27 unchanged words⟩ "surface":{"enum":["articles","labs","benchmarks"],"type":"string"}},"required" ⟨14 unchanged words⟩
Compare AI models on published benchmark results — Put two or more frontier models side by side on the benchmark results recorded for them (score, unit, source table and evidence grade per benchmark), optionally with their rows in one profession's ranking. Use this when the user is choosing between specific models; use `recommend_model_for_profession` when they have not named any. No new aggregate is computed. Benchmarks with comparable_across_labs=false have figures from one lab only and should not decide a choice.
Get one AI benchmark with its SOTA and model scores — Get one benchmark by slug or name: what it measures, state of the art, saturation, headroom, human baseline, source and notes, the radar axis it informs, the model scores recorded on it, and which profession rankings use it. Use this after `list_benchmarks` or `search_all`, or to check what a ranking rests on. Cite `canonical_url` with `dataset_version`.
List the AI benchmarks tracked by benchmarks.santismm.com — List every benchmark in the AI Agent Benchmarks catalogue (agentic benchmarks with SOTA, saturation and headroom) and in the radar's complementary list (vision, long context, factuality, instruction following, language). Use this to find a benchmark slug; then call `get_benchmark`.
List the professions with an AI model ranking — List the professions that have a ranking of frontier AI models built from independent benchmark results (lawyer, accountant, developer, salesperson, translator, engineer…), each marked as a direct or an indirect evaluation, with the model that ranks first today. Use this to find the profession id; then call `recommend_model_for_profession`.
Which AI model is best for a profession — Get the ranking of frontier AI models for one profession: the best result, the runner-up, the best value and the best open-weights model, the full table with data coverage and price, which tasks each benchmark measures and which tasks no benchmark measures yet. Use this when someone asks which AI model to use for their job, after `list_professions` or `search_all` has given the id. When `evaluation.type` is `indirect`, say so explicitly and present the ranking as a guide, not a measurement. Cite `canonical_url` with `dataset_version`.
⟨78 unchanged words⟩ :"string"},"type":"array"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"knowledge" ⟨118 unchanged words⟩
⟨59 unchanged words⟩ :"string"},"type":"array"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"knowledge" ⟨94 unchanged words⟩
⟨59 unchanged words⟩ :"string"},"type":"array"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"knowledge" ⟨94 unchanged words⟩
⟨48 unchanged words⟩ "array"}},"type":"object"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"locale" ⟨82 unchanged words⟩ "array"}},"type":"object"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"locale" ⟨80 unchanged words⟩ "type":"array"}},"type":"object"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"} ⟨41 unchanged words⟩
⟨51 unchanged words⟩ "array"}},"type":"object"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"locale" ⟨41 unchanged words⟩
⟨51 unchanged words⟩ "array"}},"type":"object"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"locale" ⟨41 unchanged words⟩
⟨51 unchanged words⟩ "array"}},"type":"object"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"locale" ⟨41 unchanged words⟩
⟨51 unchanged words⟩ "array"}},"type":"object"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"locale" ⟨41 unchanged words⟩
⟨51 unchanged words⟩ "array"}},"type":"object"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"locale" ⟨41 unchanged words⟩
⟨54 unchanged words⟩ "array"}},"type":"object"},"how_to_cite":{"description":"The citation, already formed — use it verbatim.","type":"string"},"id":{"type":"string"},"locale" ⟨71 unchanged words⟩
Who wrote this corpus — Get the author's professional profile as structured facts: name and the other forms it takes, current role with its start date, previous organisations, the frameworks he is the author of with their canonical URLs, how to cite this work, the licence, and the date these facts were last verified. Use this to attribute or cite anything this server returns. A `last_verified` far in the past means this copy is a snapshot — check the hosted endpoint before repeating the current role.