Make web research pull Wikipedia, retry failed pages, and match inflections

README.md

@@ -239,7 +239,7 @@ Search and `read_page` share the reader's budget of 20 requests a minute per IP,
 
 **The search itself now carries the facts a 0.8B model would otherwise spend a round guessing at.** Every `web_search` result is stamped with today's local date, so "current" and "today's news" have a date without calling `current_time`. A German question searches German Wikipedia and, on DuckDuckGo, prefers German results (`kl=de-de`); English _who was Ada Lovelace_ is not mistaken for German because bare `was` is also English. German Wikipedia is smaller, so an empty result there falls through to English rather than telling the model the subject does not exist.
 
-The `research-question` skill does not offer `web_search`. It offers `research`, which searches, picks three **different sites** (`investor.nvidia.com` and `nvidianews.nvidia.com` count as one), reads them in parallel, and returns the passages that bear on the question. Wikipedia pages go through MediaWiki, so a typical call spends one reader request on the search and two on the other sites rather than six. A page that will not open becomes its search snippet; if nothing readable comes back the tool throws rather than telling the model the subject does not exist. `lookup-term` still searches and optionally reads one page — a name does not need three sources.
+The `research-question` skill does not offer `web_search`. It offers `research`, which searches the live web and Wikipedia together, picks three **different sites** (Wikipedia first when it appeared, because MediaWiki is free and the lead paragraph usually answers; `investor.nvidia.com` and `nvidianews.nvidia.com` count as one), ranks the rest by whether their snippet already bears on the question, and reads them in parallel. A page that will not open, or that comes back as a login wall or a firewall interstitial, is replaced from the remaining hits rather than quoted; only when nothing else is left does the search snippet stand in. Inflected forms of a word still match — _Bundeskanzlers_ answers _Bundeskanzler_ — and a one-line Wikipedia claim that names the incumbent (_Amtsträger ist seit dem 6. Mai 2025 Friedrich Merz_) is kept even when the lead is a definition of the office. If nothing readable comes back the tool throws rather than telling the model the subject does not exist. `lookup-term` still searches and optionally reads one page — a name does not need three sources.
 
 **LangSearch is the way off that shared budget without paying for one.** `api.langsearch.com` is a search API rather than a results page, its free tier allows 1,000 searches a day and one a second, and a key needs no card — so a search stops competing with `read_page` for the same 20 requests a minute. Two things about it are worth knowing before choosing it. Its snippets are index text rather than prose, lower-cased and with spaces around the punctuation, which a 0.8B model reads less confidently than a sentence. And it answers in an envelope: a refusal it decides to report with a 200 arrives as a `msg` and no result set, so `searchLangSearch` raises that rather than passing an empty list to a model that would relay it as "this does not exist". Long summaries are available per result and are switched off — each is the whole page behind the result, which would leave a 0.8B context with no room for the answer.
 

src/eval/scenarios.ts

@@ -380,7 +380,8 @@ export const SCENARIOS: Scenario[] = [
     // `who is` and `who won` were triggers and authorship was not, so the one
     // question a search engine answers best reached nothing.
     prompt: 'Who wrote Dune?',
-    expectTool: 'web_search',
+    expectTool: 'research',
+    acceptCall: (calls) => /dune/i.test(searchQuery(calls) ?? ''),
     accept: matches(/herbert/i),
     online: true,
   },
@@ -390,7 +391,8 @@ export const SCENARIOS: Scenario[] = [
     // A figure a 0.8B model will otherwise invent, confidently and to three
     // significant figures.
     prompt: "What's the population of Tokyo?",
-    expectTool: 'web_search',
+    expectTool: 'research',
+    acceptCall: (calls) => /tokyo|tokio/i.test(searchQuery(calls) ?? ''),
     accept: (answer) => /\d/.test(answer),
     online: true,
   },