diff options
| author | Igor Soarez <igor@soarez.org> | 2026-08-03 21:43:56 +0100 |
|---|---|---|
| committer | Igor Soarez <igor@soarez.org> | 2026-08-03 21:43:56 +0100 |
| commit | db69207c06e8d5233bff4996e3ef5b43332d533f (patch) | |
| tree | 032ebbc3983a6eb4f9e3c4ac162c60a9cc1edc75 /test/format.test.ts | |
| parent | 495de0d5283dd3e4a6ef715b596c4a2892e95915 (diff) | |
Support DuckDuckGo, Bing and Brave, switchable like the CDP host
Engine resolution mirrors the browser target: CCS_SEARCH_ENGINE, then a
/search-engine choice persisted machine-wide, then Google. A per-call
`engine` parameter sits above both so the agent can fall back when one
engine starts serving captchas — the one case where the model, not the
operator, has to make the call. Unlike the CDP target there is no safety
argument for the environment winning: driving the wrong browser means
automating someone's signed-in Chrome, choosing a different index does not.
An unsupported recency window is refused, naming the engines that support
it, rather than dropped. Silently returning unfiltered results is
indistinguishable from success, which is the failure this whole design is
trying to avoid.
Extraction grows a second mode. DuckDuckGo, Bing and Brave have clean
per-result containers; Google does not, so its heading-walk stays as its
own path rather than being bent into the item shape. DuckDuckGo and Bing
route links through redirectors, unwrapped in the page.
Third silent-failure trap found, alongside Google's two: on Bing, `count`
cancels `filters`. With ex1:"ez1" alone every result is hours old; add
count in either order and months-old results return, looking perfectly
ordinary. Bing now drops count whenever a date filter is present.
Challenge detection widened to Brave's "Verifying you're not a bot" and
"Quick check before you continue searching", which the previous Google-
shaped matcher missed entirely — found by tripping it.
Diffstat (limited to 'test/format.test.ts')
| -rw-r--r-- | test/format.test.ts | 20 |
1 files changed, 19 insertions, 1 deletions
diff --git a/test/format.test.ts b/test/format.test.ts index 3060a62..74e37e1 100644 --- a/test/format.test.ts +++ b/test/format.test.ts @@ -15,9 +15,16 @@ test("results are numbered, with the URL on its own line", () => { searchUrl: "https://www.google.com/search?q=q", finalUrl: "https://www.google.com/search?q=q", endpoint: "10.88.0.25:9223 (built-in default)", + engine: "Google", + engineSource: "built-in default", results, }); - assert.match(text, /^Search results for "q" \(2 hits, via Chrome at 10\.88\.0\.25:9223 \(built-in default\)\)/); + // The engine has to be named on every result set: the agent may have switched + // engines to get around a block, and two indexes are not interchangeable. + assert.match( + text, + /^Search results for "q" — 2 hits from Google \(built-in default\), via Chrome at 10\.88\.0\.25:9223 \(built-in default\)$/m, + ); assert.match(text, /^1\. First$/m); assert.match(text, /^ {3}https:\/\/example\.com\/1$/m); assert.match(text, /^2\. Second$/m); @@ -31,6 +38,8 @@ test("a redirect is reported, because the query URL is then not where we looked" searchUrl: "https://www.google.com/search?q=q", finalUrl: "https://www.google.com/search?q=q&sei=abc", endpoint: "e", + engine: "Bing", + engineSource: "/search-engine", results, }); assert.match(text, /^Landed on: https:\/\/www\.google\.com\/search\?q=q&sei=abc$/m); @@ -41,6 +50,8 @@ test("the challenge error tells the agent to hand off to a human, with the URL", challenge: "captcha-widget", pageUrl: "https://www.google.com/sorry/index?continue=x", endpoint: "10.88.0.25:9223 (built-in default)", + engine: "Google", + alternatives: ["duckduckgo", "bing", "brave"], detail: "Our systems have detected unusual traffic", }); assert.equal(err.name, "SearchChallengeError"); @@ -49,4 +60,11 @@ test("the challenge error tells the agent to hand off to a human, with the URL", assert.match(err.message, /https:\/\/www\.google\.com\/sorry\/index\?continue=x/); assert.match(err.message, /Ask the user/); assert.match(err.message, /10\.88\.0\.25:9223/); + // Switching engines is the agent's own move and must be offered before the + // human handoff, since a block on one engine says nothing about the others. + for (const alt of ["duckduckgo", "bing", "brave"]) assert.match(err.message, new RegExp(alt)); + assert.ok( + err.message.indexOf("retry the same query with engine") < err.message.indexOf("Ask the user"), + "the self-service fallback must come before the human handoff", + ); }); |
