diff --git a/lib/search/api.sx b/lib/search/api.sx
index 84918b5e..cef49db4 100644
--- a/lib/search/api.sx
+++ b/lib/search/api.sx
@@ -4,7 +4,8 @@
 ;; interpreter. Public Haskell entry points: indexDoc, lookupTerm, deleteDoc,
 ;; docFreq, allTerms, tokens, positioned, evalQuery, parseQuery, searchQuery,
 ;; rankTfIdf, rankBm25, topNTfIdf, topNBm25, fedIndex, aclFilter, searchTfIdfAcl,
-;; topNTfIdfAcl, searchBm25Acl, prefixTerms, prefixDocs, prefixRankTfIdf.
+;; topNTfIdfAcl, searchBm25Acl, prefixTerms, prefixDocs, prefixRankTfIdf,
+;; paginate, pageTfIdf, pageBm25, resultCount.
 
 (define
   search/src
@@ -21,4 +22,6 @@
     "\n"
     search/fed-src
     "\n"
-    search/prefix-src))
+    search/prefix-src
+    "\n"
+    search/page-src))
diff --git a/lib/search/conformance.conf b/lib/search/conformance.conf
index c5d09b5c..79b14819 100644
--- a/lib/search/conformance.conf
+++ b/lib/search/conformance.conf
@@ -25,6 +25,7 @@ PRELOADS=(
   lib/search/rank.sx
   lib/search/fed.sx
   lib/search/prefix.sx
+  lib/search/page.sx
   lib/search/api.sx
   lib/search/testlib.sx
 )
@@ -36,4 +37,5 @@ SUITES=(
   "rank:lib/search/tests/rank.sx"
   "integration:lib/search/tests/integration.sx"
   "prefix:lib/search/tests/prefix.sx"
+  "page:lib/search/tests/page.sx"
 )
diff --git a/lib/search/page.sx b/lib/search/page.sx
new file mode 100644
index 00000000..93b57dd4
--- /dev/null
+++ b/lib/search/page.sx
@@ -0,0 +1,11 @@
+;; search pagination — Haskell source fragment. Depends on rank.
+;; Windows a ranked result list by offset/limit (offset >= length -> empty;
+;; limit clamps to what remains).
+;;   paginate    :: Int -> Int -> [DocId] -> [DocId]   (offset, limit)
+;;   pageTfIdf   :: Int -> Int -> [Term] -> Index -> [DocId]
+;;   pageBm25    :: Int -> Int -> Float -> Float -> [Term] -> Index -> [DocId]
+;;   resultCount :: [Term] -> Index -> Int
+
+(define
+  search/page-src
+  "paginate off lim docs = take lim (drop off docs)\npageTfIdf off lim ts idx = paginate off lim (rankTfIdf ts idx)\npageBm25 off lim k1 b ts idx = paginate off lim (rankBm25 k1 b ts idx)\nresultCount ts idx = length (rankTfIdf ts idx)\n")
diff --git a/lib/search/scoreboard.json b/lib/search/scoreboard.json
index df5e60d7..16472224 100644
--- a/lib/search/scoreboard.json
+++ b/lib/search/scoreboard.json
@@ -1,15 +1,16 @@
 {
   "lang": "search",
-  "total_passed": 136,
+  "total_passed": 148,
   "total_failed": 0,
-  "total": 136,
+  "total": 148,
   "suites": [
     {"name":"index","passed":18,"failed":0,"total":18},
     {"name":"boolean","passed":28,"failed":0,"total":28},
     {"name":"parse","passed":32,"failed":0,"total":32},
     {"name":"rank","passed":23,"failed":0,"total":23},
     {"name":"integration","passed":21,"failed":0,"total":21},
-    {"name":"prefix","passed":14,"failed":0,"total":14}
+    {"name":"prefix","passed":14,"failed":0,"total":14},
+    {"name":"page","passed":12,"failed":0,"total":12}
   ],
-  "generated": "2026-06-06T20:21:41+00:00"
+  "generated": "2026-06-06T20:54:50+00:00"
 }
diff --git a/lib/search/scoreboard.md b/lib/search/scoreboard.md
index 0578f296..9cdc93b3 100644
--- a/lib/search/scoreboard.md
+++ b/lib/search/scoreboard.md
@@ -1,6 +1,6 @@
 # search scoreboard
 
-**136 / 136 passing** (0 failure(s)).
+**148 / 148 passing** (0 failure(s)).
 
 | Suite | Passed | Total | Status |
 |-------|--------|-------|--------|
@@ -10,3 +10,4 @@
 | rank | 23 | 23 | ok |
 | integration | 21 | 21 | ok |
 | prefix | 14 | 14 | ok |
+| page | 12 | 12 | ok |
diff --git a/lib/search/tests/page.sx b/lib/search/tests/page.sx
new file mode 100644
index 00000000..6ad77310
--- /dev/null
+++ b/lib/search/tests/page.sx
@@ -0,0 +1,53 @@
+;; Extension — result pagination (offset / limit) over ranked results.
+;; Corpus (tf of "x" descending): 1 x4  2 x3  3 x2  4 x1  5 y(no x)
+;; rankTfIdf ["x"] -> [1,2,3,4]
+
+(define
+  page-setup
+  "idx = indexDoc 5 \"y\" (indexDoc 4 \"x\" (indexDoc 3 \"x x\" (indexDoc 2 \"x x x\" (indexDoc 1 \"x x x x other\" emptyIndex))))\n")
+
+(define
+  page-cases
+  (list
+    (list "first page" "pageTfIdf 0 2 [\"x\"] idx" (list 1 2))
+    (list
+      "second page"
+      "pageTfIdf 2 2 [\"x\"] idx"
+      (list 3 4))
+    (list
+      "sliding window"
+      "pageTfIdf 1 2 [\"x\"] idx"
+      (list 2 3))
+    (list
+      "limit exceeds remaining"
+      "pageTfIdf 3 10 [\"x\"] idx"
+      (list 4))
+    (list "offset past end" "pageTfIdf 4 2 [\"x\"] idx" (list))
+    (list "limit zero" "pageTfIdf 0 0 [\"x\"] idx" (list))
+    (list
+      "whole result"
+      "pageTfIdf 0 10 [\"x\"] idx"
+      (list 1 2 3 4))
+    (list
+      "paginate raw list"
+      "paginate 1 2 [10, 20, 30, 40]"
+      (list 20 30))
+    (list "paginate raw past end" "paginate 9 2 [10, 20]" (list))
+    (list
+      "bm25 page window size"
+      "[length (pageBm25 0 2 1.5 0.75 [\"x\"] idx)]"
+      (list 2))
+    (list "result count" "[resultCount [\"x\"] idx]" (list 4))
+    (list "result count zero" "[resultCount [\"zzz\"] idx]" (list 0))))
+
+(define
+  page-results
+  (search-batch page-setup (map (fn (c) (nth c 1)) page-cases)))
+
+(map-indexed
+  (fn
+    (i c)
+    (hk-test (nth c 0) (nth page-results i) (nth c 2)))
+  page-cases)
+
+{:fail hk-test-fail :pass hk-test-pass :fails hk-test-fails}
diff --git a/plans/search-on-sx.md b/plans/search-on-sx.md
index 2f440bd0..32444f20 100644
--- a/plans/search-on-sx.md
+++ b/plans/search-on-sx.md
@@ -112,12 +112,17 @@ lib/search/index.sx                     lib/search/eval.sx
 
 - [x] prefix / wildcard queries (`prefixTerms`, `prefixDocs`, `prefixRankTfIdf`) — 14 tests
 - [ ] fuzzy matching — edit distance term expansion
-- [ ] result pagination (offset / limit)
+- [x] result pagination (offset / limit) — `paginate`, `pageTfIdf`, `pageBm25`,
+  `resultCount` — 12 tests
 - [ ] snippet / highlight generation
 - [ ] stemming (suffix stripping) — recall-improving normalizer
 
 ## Progress log
 
+- **Extension: pagination (148/148 total).** `paginate off lim` windows a ranked list
+  (take lim . drop off); `pageTfIdf`/`pageBm25` + `resultCount`. 12 tests. Note the
+  full conformance now runs 8 suites sequentially and needs an overall timeout ~1900s
+  under the heavy box load.
 - **Extension: prefix/wildcard queries (136/136 total).** `prefixTerms` matches every
   indexed term starting with a prefix (via allTerms + isPrefixOf); `prefixDocs` unions
   their docs; `prefixRankTfIdf` ranks treating the matched terms as the query. 14 tests.