diff --git a/performance/projects/PROCESS.md b/performance/projects/PROCESS.md
index 36666ee..eeae699 100644
--- a/performance/projects/PROCESS.md
+++ b/performance/projects/PROCESS.md
@@ -123,6 +123,10 @@ The existing study READMEs own their exact installation commands and bounds:
    prefix. The public SDK seeds invented documents; actual CLI subprocesses run under the existing
    external-I/O guard. Keep the bound of one initial call and two resumes plus the separate tiny control.
    Record the result as a known-limitation reproduction, not a passing retrieval-quality measurement.
+5. [Evidence on large documents](evidence-large-document/README.md#setup): a separate test beside the frozen
+   historical evidence, which stays unchanged. The runner takes a built Memo checkout like the historical one,
+   but runs its CLI under the external-I/O guard and refuses unless every state folder is inside one fresh
+   `/var/tmp` directory. Labels are in its own `baseline.json`.
 
 For example, after creating the reviewed isolated package prefix as the Drive guide describes:
 
diff --git a/performance/projects/README.md b/performance/projects/README.md
index 6ae3d29..e900622 100644
--- a/performance/projects/README.md
+++ b/performance/projects/README.md
@@ -25,5 +25,8 @@
 
 - [evidence](evidence/README.md): what `memo projects evidence` finds and misses on a synthetic store against a
   frozen baseline — the project's names, extra words and people, the 60-day window, resume, unread sources.
+- [evidence-large-document](evidence-large-document/README.md): the same command on large imported documents —
+  paragraphs found, paragraphs shown around a hit, pages, charged bytes, `oversized` and resume. A separate test: the
+  frozen one above is unchanged.
 - [review](review/README.md): what reviewing an agent's proposals costs with and without context, and how a changed
   cited text and agent decisions are handled.
diff --git a/performance/projects/evidence-large-document/README.md b/performance/projects/evidence-large-document/README.md
new file mode 100644
index 0000000..1837bef
--- /dev/null
+++ b/performance/projects/evidence-large-document/README.md
@@ -0,0 +1,171 @@
+# Project evidence on large documents
+
+## Why we ran it
+
+The [frozen project evidence test](../evidence/README.md) holds 13 short texts, so it cannot show how
+`memo projects evidence` behaves on a large imported document. Two things were never checked: what the command
+returns when one document is far larger than a page and its matching paragraphs lie far apart, and which
+paragraphs the `around` context shows for a document hit. The risk was recorded as RISK-1268. The owner decided
+(NEED-1572, option A) to measure it in a separate test and leave the frozen fixture, its `baseline.json` and its
+results unchanged. The decision this feeds: whether a reader who gets a document hit can trust the excerpt and the
+`around` paragraphs to point at what matched, and whether a long search finishes when it is cut into steps.
+
+## Setup
+
+- AMD Ryzen AI 9 HX 470, Linux 7.0, Node v24.19.0.
+- Memo 0.82.0, built from `main` at `579e0af` (`v0.82.0-3-g579e0af`), with `@wirecat/cli-core` 0.371.0 and
+  `@wirecat/cli-base` 0.47.1. Measured 2026-10-20.
+- Limits from Core's [project gathering guide](https://github.com/WireCatLabs/cli-core/blob/main/docs/dev/PROJECT-GATHERING.md):
+  a page is 64 KiB (65,536 bytes); a record above that is read alone, up to 256 KiB; above 256 KiB it is listed
+  under `oversized` and not read; the read budget is 2 MiB per invocation.
+- Synthetic store of 11 documents in one imported notes folder, project "Lantern Works" with key `LNW` and alias
+  "Beacon Desk" (three searches, shown as `l1` name, `l2` key, `l3` alias). Paragraphs are split at blank lines and
+  start with a marker (`p004`), so the paragraph a result shows can be read off its text.
+
+  | Document | Bytes | Paragraphs that name the project |
+  | --- | --- | --- |
+  | `large` | 144,829 | p4 (name), p60 (key), p150 (alias), p215 (name) |
+  | `huge` | 284,459 | p10 (name) |
+  | `m1` to `m6` | 42,829 each | p30 (name) |
+  | `small` | 76 | p2 (alias) |
+  | `decoy-lantern` | 29,865 | none: "lantern" and "works" in the same paragraph |
+  | `decoy-garden` | 29,841 | none |
+
+- The right answers are in [`baseline.json`](baseline.json), written before the first run: which documents must be
+  found, which paragraphs `around` must show, that `huge` is `oversized`, and what the resume must do. The runner
+  [`evidence-large-document.ts`](evidence-large-document.ts) builds the store through the build's store module, runs
+  the built `memo projects evidence LNW --json` three ways (default; `--before-n 5 --after-n 5`; `--max-pages 2` and
+  `--cursor` until it stops) and prints the scores. It runs under the existing
+  [external-I/O guard](../../documents/drive-shared-store/guard.mjs). It does not touch the owner's store; it
+  refuses to start unless HOME, the XDG folders, `MESSAGING_STORE`, `WIRECAT_STATE_DIR` and `TMPDIR` are inside
+  one fresh `/var/tmp` folder.
+- The runner reuses no helper from [`../fixture.ts`](../fixture.ts): that file seeds the Harbour Bureau store and
+  cannot take other documents. The same sandbox idea is repeated, as in the frozen test.
+
+Command, from the repository root, with a built Memo checkout:
+
+```sh
+S=$(mktemp -d -p /var/tmp ld-XXX) && HOME=$S XDG_CONFIG_HOME=$S/config XDG_DATA_HOME=$S/data XDG_CACHE_HOME=$S/cache MESSAGING_STORE=$S/store.db WIRECAT_STATE_DIR=$S/state TMPDIR=$S node performance/projects/evidence-large-document/evidence-large-document.ts /path/to/cli-memo > performance/projects/evidence-large-document/result.json
+```
+
+The full output of this run is [`result.json`](result.json).
+
+## Results
+
+Default run, one invocation: 13 reads, 12 pages, `complete: true`, no checkpoint, 104 ms.
+
+| Check | Label (before the run) | Observed |
+| --- | --- | --- |
+| Documents found | `large`, `m1` to `m6`, `small` | all eight found |
+| Decoys not found | `decoy-lantern`, `decoy-garden` | `decoy-garden` not found; **`decoy-lantern` found** by the name search |
+| `huge` (284,459 bytes) | listed under `oversized`, charged 0 | `oversized` lists it (284,515 bytes, name search); no item |
+| `large` found by | three searches | all three (`l1`, `l2`, `l3`) |
+| Charged bytes | at most 1.2 times the size of the documents read | 432,246 for 431,744 bytes of found documents: 1.00 |
+| Pages | name search at least 4 | name search 9 (one record per page), key search 1, alias search 2 |
+| Mediums and `small`: `around` | p28 to p32 and p1 to p3 | exactly as labelled, the hit paragraph marked `found` |
+| `large`: `around` | p2 to p6, around the first paragraph that names the project (p4) | **p58 to p62, around p60** (the key paragraph) |
+| `large` with `--before-n 5 --after-n 5` | p1 to p9 | **p55 to p65**, around p60 |
+| Paragraphs that name the project, shown in `around` | 8 of 12 | 8 of 12, but one of them is the wrong one: `large` p60 is shown instead of p4; p150, p215 and the paragraph in `huge` (not read) are not |
+| `around` bytes (`usage.around_bytes`) | not labelled | 8,171 default; 17,887 with 5 and 5 |
+
+The excerpt of a hit is the first 300 characters of the document, and `around` shows 200-character excerpts
+(Core's guide). So of the four paragraphs in `large` that name the project, a reader sees one, and which one depends
+on the search that found the document first, not on the position in the document.
+
+Resume with `--max-pages 2` and `--cursor`:
+
+| Invocation | New references | Pages so far | Charged bytes | `complete` |
+| --- | --- | --- | --- | --- |
+| 1 | 1 | 1 | 42,881 | false |
+| 2 | 1 | 2 | 144,887 | false |
+| 3 | 2 | 4 | 43,015 | false |
+| 4 | 0 | 4 | 0 | false |
+
+It never completes: 4 of the 9 references the one-shot run returns, no duplicates, and invocation 4 and every later
+one read nothing and return the same checkpoint (the runner stops there; the first run went on to 41 invocations
+with the same result). `stopped` is `null`, so the output does not say why it stopped.
+
+## What it showed
+
+1. **Large-document paging works in one shot.** A 145 KB document and six 43 KB ones are found in a single
+   invocation, each read once (1.00 of the bytes), and the 284 KB document is reported as `oversized` instead of
+   stalling the run.
+2. **BUG-1: a resumed run stalls** (fixed, see [After the fixes](#after-the-fixes-memo-0840-core-03770)). With `--max-pages 2` the units end `partial` and the checkpoint is returned,
+   but a resume makes no read and returns the same checkpoint, forever. Tried with `--max-pages 3` and `4`: the
+   same, at other points. A caller that follows `checkpoint` as the guide says loops without progress. A
+   likely cause, not confirmed, is the `huge` record or the `read-limited` state of `large` that the first
+   invocation reports (its units show `read-limited` with `validation_error`); the run was not repeated without
+   `huge`.
+3. **FIND-1: `around` for a document follows the search that found it first, not the document.** Core's guide says
+   the context is around "the first one that names a lane word, in document order". In `large` that would be p4;
+   the output is around p60, the paragraph of the key search. The three other matching paragraphs are not shown
+   and nothing in the item says there are more. Fixed, see [After the fixes](#after-the-fixes-memo-0840-core-03770).
+4. **FIND-2: the project name is searched as separate words.** `decoy-lantern` has no "Lantern Works" in it, only
+   "lantern" and "works" in one paragraph, and the name search returns it. The `--help` text calls each name "its
+   own literal search". For a project with a common-word name this gives false hits.
+
+## Not measured
+
+Timing beyond one wall-clock figure, memory, a store with many thousand documents, messages or mail next to the
+documents, a document between 256 KiB and a few MiB through other paths, other budgets (the CLI sets only
+`--max-pages`), whether `--max-pages` without `huge` completes, and any real data. All data here is invented.
+
+## After the fixes (Memo 0.84.0, Core 0.377.0)
+
+Run 2026-10-20 with the same runner and the same synthetic store. The two findings above were fixed in Core
+0.377.0 ([cli-core#1690](https://github.com/WireCatLabs/cli-core/pull/1690), resume;
+[cli-core#1694](https://github.com/WireCatLabs/cli-core/pull/1694), document `around` and the new `moreMatches`
+field), released in Memo 0.84.0 (`@wirecat/cli-memo` on npm). Setup: Memo built from the tag `v0.84.0`
+(`ca74d71`) with `@wirecat/cli-core` 0.377.0, same machine and Node. The runner gained one output field,
+`more_matches`, and nothing else; the full output is
+[`result-2026-10-20-memo-0.84.0.json`](result-2026-10-20-memo-0.84.0.json). `result.json` and `baseline.json` are
+unchanged.
+
+Default run, one invocation: 13 reads, 12 pages, `complete: true`, no checkpoint, 99 ms.
+
+| Check | Label (before the run) | Observed |
+| --- | --- | --- |
+| Documents found | `large`, `m1` to `m6`, `small` | all eight found |
+| Decoys not found | `decoy-lantern`, `decoy-garden` | `decoy-garden` not found; `decoy-lantern` found by the name search (expected, see below) |
+| `huge` (284,459 bytes) | listed under `oversized`, charged 0 | `oversized` lists it (284,515 bytes, name search); no item |
+| `large` found by | three searches | all three (`l1`, `l2`, `l3`) |
+| Charged bytes | at most 1.2 times the size of the documents read | 432,246 for 431,744 bytes of found documents: 1.00 |
+| Pages | name search at least 4 | name search 9, key search 1, alias search 2 |
+| Mediums and `small`: `around` | p28 to p32 and p1 to p3 | exactly as labelled, the hit paragraph marked `found` |
+| `large`: `around` | p2 to p6, around the first paragraph that names the project (p4) | **p2 to p6, found p4** (was p58 to p62) |
+| `large` with `--before-n 5 --after-n 5` | p1 to p9 | **p1 to p9, found p4** (was p55 to p65) |
+| Paragraphs that name the project, shown in `around` | 8 of 12 | 8 of 12, and now the right ones: `large` p4 is shown; p60, p150, p215 and the paragraph in `huge` (not read) are not |
+| `around` bytes (`usage.around_bytes`) | not labelled | 8,171 default; 17,483 with 5 and 5 |
+
+Checks that failed before and pass now: the `large` `around` position (both runs), and the resume. Everything else
+is unchanged.
+
+`moreMatches` in the default run: `large` 3 (p60, p150 and p215 are the other matching paragraphs); every other
+document 0. So the item now says that more paragraphs match than the `around` shows, though it still does not say
+which.
+
+Resume with `--max-pages 2` and `--cursor`:
+
+| Invocation | New references | Pages so far | Charged bytes | `complete` |
+| --- | --- | --- | --- | --- |
+| 1 | 1 | 1 | 42,881 | false |
+| 2 | 1 | 2 | 144,887 | false |
+| 3 | 2 | 4 | 43,015 | false |
+| 4 | 1 | 6 | 42,881 | false |
+| 5 | 2 | 8 | 85,762 | false |
+| 6 | 2 | 10 | 72,820 | false |
+| 7 | 0 | 12 | 0 | true |
+
+The resume completes in 7 invocations with the same 9 references as the one-shot run, no duplicates, 12 pages and
+432,246 charged bytes in total (the same as the one-shot run), and no invocation reads without progress.
+
+What is still open:
+
+- **`decoy-lantern` is found. This is expected behaviour, not a bug.** The owner decided (NEED-1579, option B) to
+  keep matching the words of a project name anywhere in one text, so "lantern" and "works" in one paragraph match
+  "Lantern Works". FIND-2 stays as a description of that behaviour, and a project with a common-word name will get
+  such hits. The baseline still labels the decoy as "not found"; that label is kept as written and is not
+  re-scored.
+- Only the first matching paragraph of a document is shown in `around`; `moreMatches` gives the count of the
+  others, not their position.
+- Everything under "Not measured" above is still not measured.
diff --git a/performance/projects/evidence-large-document/baseline.json b/performance/projects/evidence-large-document/baseline.json
new file mode 100644
index 0000000..bba0904
--- /dev/null
+++ b/performance/projects/evidence-large-document/baseline.json
@@ -0,0 +1,105 @@
+{
+  "frozen": "2026-10-20",
+  "written_before_first_run": true,
+  "request": {
+    "project": "LNW",
+    "source": "notes:<folder>",
+    "since_time": "60d",
+    "around": "default 2 before, 2 after"
+  },
+  "thresholds": { "page_bytes": 65536, "single_record_read_bytes": 262144, "default_read_budget_bytes": 2097152 },
+  "documents": {
+    "large": {
+      "approx_bytes": 150000,
+      "paragraphs": 220,
+      "label": "relevant",
+      "why": "above one page (64 KiB), below the 256 KiB single-record limit; four paragraphs name the project, far apart",
+      "hit_paragraphs": { "4": "name", "60": "key", "150": "alias", "215": "name" },
+      "around_default": [2, 3, 4, 5, 6],
+      "around_default_found": [4],
+      "around_five": [1, 2, 3, 4, 5, 6, 7, 8, 9]
+    },
+    "huge": {
+      "approx_bytes": 300000,
+      "paragraphs": 430,
+      "label": "relevant",
+      "why": "above 256 KiB: expected to be listed under oversized, not read",
+      "hit_paragraphs": { "10": "name" }
+    },
+    "m1": {
+      "approx_bytes": 45000,
+      "paragraphs": 65,
+      "label": "relevant",
+      "why": "medium document, one hit in the middle",
+      "hit_paragraphs": { "30": "name" },
+      "around_default": [28, 29, 30, 31, 32]
+    },
+    "m2": {
+      "approx_bytes": 45000,
+      "paragraphs": 65,
+      "label": "relevant",
+      "why": "medium document, one hit in the middle",
+      "hit_paragraphs": { "30": "name" },
+      "around_default": [28, 29, 30, 31, 32]
+    },
+    "m3": {
+      "approx_bytes": 45000,
+      "paragraphs": 65,
+      "label": "relevant",
+      "why": "medium document, one hit in the middle",
+      "hit_paragraphs": { "30": "name" },
+      "around_default": [28, 29, 30, 31, 32]
+    },
+    "m4": {
+      "approx_bytes": 45000,
+      "paragraphs": 65,
+      "label": "relevant",
+      "why": "medium document, one hit in the middle",
+      "hit_paragraphs": { "30": "name" },
+      "around_default": [28, 29, 30, 31, 32]
+    },
+    "m5": {
+      "approx_bytes": 45000,
+      "paragraphs": 65,
+      "label": "relevant",
+      "why": "medium document, one hit in the middle",
+      "hit_paragraphs": { "30": "name" },
+      "around_default": [28, 29, 30, 31, 32]
+    },
+    "m6": {
+      "approx_bytes": 45000,
+      "paragraphs": 65,
+      "label": "relevant",
+      "why": "medium document, one hit in the middle",
+      "hit_paragraphs": { "30": "name" },
+      "around_default": [28, 29, 30, 31, 32]
+    },
+    "small": {
+      "approx_bytes": 300,
+      "paragraphs": 3,
+      "label": "relevant",
+      "why": "tiny document that names the project only by its alias",
+      "hit_paragraphs": { "2": "alias" },
+      "around_default": [1, 2, 3]
+    },
+    "decoy-lantern": {
+      "approx_bytes": 30000,
+      "label": "irrelevant",
+      "why": "says lantern and works separately, never the project's name"
+    },
+    "decoy-garden": { "approx_bytes": 30000, "label": "irrelevant", "why": "another project's plan" }
+  },
+  "expect": {
+    "found": ["large", "m1", "m2", "m3", "m4", "m5", "m6", "small"],
+    "oversized": ["huge"],
+    "not_found": ["decoy-lantern", "decoy-garden"],
+    "paragraphs_relevant_total": 12,
+    "paragraphs_surfaced_by_around": "only those within 2 of the first hit paragraph of each document: large 1 of 4, each medium 1 of 1, small 1 of 1",
+    "large_lanes": ["name", "key", "alias"],
+    "pages_name_lane_at_least": 4,
+    "charged_bytes_per_document_size_max": 1.2,
+    "charged_bytes_for_oversized_document": 0,
+    "complete": true,
+    "resume_two_pages": { "completes": true, "duplicates": 0, "same_references_as_one_run": true }
+  }
+}
diff --git a/performance/projects/evidence-large-document/evidence-large-document.ts b/performance/projects/evidence-large-document/evidence-large-document.ts
new file mode 100644
index 0000000..fb36125
--- /dev/null
+++ b/performance/projects/evidence-large-document/evidence-large-document.ts
@@ -0,0 +1,239 @@
+// Project evidence on large imported documents, scored against baseline.json (written before the first run):
+// one document above the 64 KiB page, one above the 256 KiB single-record limit, six medium ones that need several
+// pages, and decoys. Measures which paragraphs are found, which show in `around`, pages, charged bytes, `oversized`
+// and `--max-pages` + `--cursor`. The frozen performance/projects/evidence test cannot reach any of this.
+//
+//   node evidence-large-document.ts <cli-memo checkout with dist/ and node_modules/>
+//
+// Run it with HOME, XDG_*, MESSAGING_STORE, WIRECAT_STATE_DIR and TMPDIR inside one fresh /var/tmp directory
+// (see README.md). Results: README.md. Nothing here opens the owner's store.
+import { execFileSync } from "node:child_process"
+import { existsSync, readFileSync, realpathSync } from "node:fs"
+import { join } from "node:path"
+import { pathToFileURL } from "node:url"
+
+const [memo] = process.argv.slice(2)
+if (!memo) throw new Error("usage: node evidence-large-document.ts <cli-memo checkout with dist/>")
+const baseline = JSON.parse(readFileSync(new URL("baseline.json", import.meta.url), "utf8"))
+
+const home = realpathSync(process.env.HOME ?? "")
+if (!home.startsWith("/var/tmp/")) throw new Error("run with HOME inside a fresh /var/tmp directory")
+for (const key of [
+  "XDG_CONFIG_HOME",
+  "XDG_DATA_HOME",
+  "XDG_CACHE_HOME",
+  "MESSAGING_STORE",
+  "WIRECAT_STATE_DIR",
+  "TMPDIR",
+])
+  if (!process.env[key]?.startsWith(home)) throw new Error(`${key} must stay inside HOME`)
+if (existsSync(process.env.MESSAGING_STORE as string)) throw new Error("refuse an existing store")
+const env = { ...process.env, PATH: process.env.PATH ?? "" }
+
+const filler = [
+  "The harbour crane was inspected on schedule and the report was filed without remarks.",
+  "Quarterly figures were compared with the plan and the difference was explained in the appendix.",
+  "A short note on vendor invoices: three were paid, two wait for a signature, one was disputed.",
+  "The meeting room was booked for Thursday and the agenda was sent to all attendees in advance.",
+  "Weather delays moved the delivery by two days, which the schedule absorbed without further changes.",
+  "Training material was revised after feedback and the new version replaces the earlier handout.",
+  "Open questions were collected in a list; each has an owner and a date for the next update.",
+]
+const paragraph = (index: number, hit?: string) => {
+  const id = `p${String(index).padStart(3, "0")}`
+  const body = Array.from({ length: 7 }, (_, i) => filler[(index + i) % filler.length]).join(" ")
+  return hit ? `${id} ${hit} ${body.slice(0, 420)}` : `${id} ${body}`
+}
+const document = (paragraphs: number, hits: Record<number, string> = {}) =>
+  Array.from({ length: paragraphs }, (_, i) => paragraph(i + 1, hits[i + 1])).join("\n\n")
+
+const NAME = "Lantern Works"
+const KEY = "LNW"
+const ALIAS = "Beacon Desk"
+const said = {
+  name: `${NAME} confirmed the review.`,
+  key: `${KEY} review is confirmed.`,
+  alias: `${ALIAS} confirmed the review.`,
+}
+const texts: Record<string, string> = {
+  large: document(220, { 4: said.name, 60: said.key, 150: said.alias, 215: said.name }),
+  huge: document(430, { 10: said.name }),
+  small: "p001 Notes.\n\np002 The Beacon Desk call is on Friday.\n\np003 Bring the folder.",
+  "decoy-lantern": document(45).replace(
+    "p020",
+    "p020 A lantern festival is held in the square and the town works late that night.",
+  ),
+  "decoy-garden": document(45).replace("p020", "p020 Garden plan: beds, water and compost for the spring."),
+}
+for (const key of ["m1", "m2", "m3", "m4", "m5", "m6"]) texts[key] = document(65, { 30: said.name })
+
+const { openStore } = await import(pathToFileURL(join(memo, "node_modules/@wirecat/cli-core/dist/store/index.js")).href)
+const store = await openStore({ env })
+const project = await store.knowledge.addProject({ name: NAME, key: KEY })
+await store.knowledge.addVocabularyAlias({ kind: "project", id: project.id }, ALIAS)
+const folder = await store.notes.addFolder({ name: "Large documents" })
+const keyOf: Record<string, string> = {}
+for (const [key, text] of Object.entries(texts)) {
+  const saved = await store.notes.saveFileNote({ folderId: folder.id, path: `${key}.md`, title: key, text })
+  keyOf[saved.note.ref] = key
+}
+await store.fillSearchIndex?.()
+await store.close()
+const size = (key: string) => Buffer.byteLength(texts[key] as string)
+
+const memoBin = join(memo, "dist/bin/memo.js")
+const guard = new URL("../../documents/drive-shared-store/guard.mjs", import.meta.url).pathname
+const evidence = (...args: string[]) =>
+  JSON.parse(
+    execFileSync(
+      process.execPath,
+      ["--import", guard, memoBin, "projects", "evidence", KEY, "--source", `notes:${folder.id}`, ...args, "--json"],
+      { env, encoding: "utf8", timeout: 120_000, maxBuffer: 16 * 1024 * 1024 },
+    ),
+  ).item
+
+type Item = {
+  ref: string
+  lanes: string[]
+  text: string
+  around?: { ref: string; text: string; found?: boolean }[] | null
+  moreMatches?: number
+  more_matches?: number
+}
+const marker = (text: string) => Number(/^p(\d{3})/.exec(text)?.[1] ?? Number.NaN)
+const summarise = (run: {
+  items: Item[]
+  oversized: { ref: string; bytes: number; lanes: string[] }[]
+  units: { state: string; pages: number; issue: string | null; [k: string]: unknown }[]
+  usage: Record<string, number>
+  totals: Record<string, number>
+  complete: boolean
+  checkpoint: string | null
+  stopped: string | null
+  matched_again: unknown[]
+}) => {
+  const found = run.items.map((item) => keyOf[item.ref] ?? `unknown:${item.ref}`)
+  const around = Object.fromEntries(
+    run.items.map((item) => [
+      keyOf[item.ref],
+      {
+        paragraphs: (item.around ?? []).map((one) => marker(one.text)),
+        found_marked: (item.around ?? []).filter((one) => one.found).map((one) => marker(one.text)),
+        in_other_document: (item.around ?? []).filter((one) => one.ref !== item.ref).length,
+      },
+    ]),
+  )
+  return {
+    found,
+    more_matches: Object.fromEntries(
+      run.items.map((item) => [keyOf[item.ref], item.moreMatches ?? item.more_matches ?? null]),
+    ),
+    lanes_per_document: Object.fromEntries(run.items.map((item) => [keyOf[item.ref], item.lanes])),
+    around,
+    oversized: run.oversized.map((one) => ({
+      document: keyOf[one.ref] ?? one.ref,
+      bytes: one.bytes,
+      lanes: one.lanes,
+    })),
+    units: run.units.map(({ state, pages, issue, ...rest }) => ({ ...rest, state, pages, issue })),
+    pages_total: run.units.reduce((sum, unit) => sum + unit.pages, 0),
+    usage: run.usage,
+    totals: run.totals,
+    complete: run.complete,
+    has_checkpoint: run.checkpoint !== null,
+    stopped: run.stopped,
+    matched_again: run.matched_again.length,
+  }
+}
+
+const expect = baseline.expect
+const docs = baseline.documents as Record<
+  string,
+  { hit_paragraphs?: Record<string, string>; around_default?: number[] }
+>
+const sameSet = (a: string[], b: string[]) => [...a].sort().join() === [...b].sort().join()
+
+const one = evidence()
+const base = summarise(one)
+const wide = summarise(evidence("--before-n", "5", "--after-n", "5"))
+const readable = expect.found as string[]
+const bytesRead = base.found.reduce((sum, key) => sum + size(key), 0)
+
+const aroundOk = Object.fromEntries(
+  readable.map((key) => [
+    key,
+    sameSet((base.around[key]?.paragraphs ?? []).map(String), (docs[key]?.around_default ?? []).map(String)),
+  ]),
+)
+const relevantParagraphs = Object.entries(docs).flatMap(([key, doc]) =>
+  Object.keys(doc.hit_paragraphs ?? {}).map((index) => ({ key, index: Number(index) })),
+)
+const surfaced = relevantParagraphs.filter(({ key, index }) => base.around[key]?.paragraphs.includes(index))
+
+const pages: { new_refs: number; pages: number; charged_bytes: number; complete: boolean }[] = []
+const refs: string[] = []
+let stalled = false
+let step = evidence("--max-pages", "2")
+for (let resumes = 0; ; resumes++) {
+  refs.push(...step.items.map((item: Item) => item.ref))
+  pages.push({
+    new_refs: step.items.length,
+    pages: step.units.reduce((sum: number, unit: { pages: number }) => sum + unit.pages, 0),
+    charged_bytes: step.usage.read_bytes,
+    complete: step.complete,
+  })
+  if (!step.checkpoint || resumes >= 40) break
+  if (resumes > 0 && step.usage.reads === 0 && step.items.length === 0) {
+    stalled = true
+    break
+  }
+  step = evidence("--max-pages", "2", "--cursor", step.checkpoint)
+}
+
+process.stdout.write(
+  `${JSON.stringify(
+    {
+      memo: execFileSync(process.execPath, [memoBin, "--version"], { env, encoding: "utf8" }).trim(),
+      node: process.version,
+      document_bytes: Object.fromEntries(Object.keys(texts).map((key) => [key, size(key)])),
+      default_run: {
+        ...base,
+        found_all_readable: sameSet(base.found, readable),
+        unexpected_found: base.found.filter((key) => !readable.includes(key)),
+        charged_bytes: base.usage.read_bytes,
+        found_document_bytes: bytesRead,
+        charged_per_document_byte: +(base.usage.read_bytes / bytesRead).toFixed(2),
+        oversized_ok: sameSet(
+          base.oversized.map((one) => one.document),
+          expect.oversized,
+        ),
+        around_matches_label: aroundOk,
+        relevant_paragraphs: relevantParagraphs.length,
+        relevant_paragraphs_in_around: surfaced.length,
+        relevant_paragraphs_not_surfaced: relevantParagraphs
+          .filter((one) => !surfaced.includes(one) && !base.oversized.some((over) => over.document === one.key))
+          .map(({ key, index }) => `${key}:p${index}`),
+      },
+      around_five_each_side: {
+        large: wide.around.large,
+        expected_large: baseline.documents.large.around_five,
+        around_bytes: wide.usage.around_bytes,
+      },
+      resume_two_pages: {
+        invocations: pages.length,
+        per_invocation: pages,
+        references: refs.length,
+        duplicates: refs.length - new Set(refs).size,
+        same_references_as_one_run: sameSet(
+          [...new Set(refs)].map((ref) => keyOf[ref] ?? ref),
+          base.found,
+        ),
+        cumulative_charged_bytes: pages.reduce((sum, one) => sum + one.charged_bytes, 0),
+        completed: pages.at(-1)?.complete ?? false,
+        stalled_without_a_read: stalled,
+      },
+    },
+    null,
+    2,
+  )}\n`,
+)
diff --git a/performance/projects/evidence-large-document/result-2026-10-20-memo-0.84.0.json b/performance/projects/evidence-large-document/result-2026-10-20-memo-0.84.0.json
new file mode 100644
index 0000000..7630c92
--- /dev/null
+++ b/performance/projects/evidence-large-document/result-2026-10-20-memo-0.84.0.json
@@ -0,0 +1,224 @@
+{
+  "memo": "0.84.0",
+  "node": "v24.19.0",
+  "document_bytes": {
+    "large": 144829,
+    "huge": 284459,
+    "small": 76,
+    "decoy-lantern": 29865,
+    "decoy-garden": 29841,
+    "m1": 42829,
+    "m2": 42829,
+    "m3": 42829,
+    "m4": 42829,
+    "m5": 42829,
+    "m6": 42829
+  },
+  "default_run": {
+    "found": ["m6", "large", "small", "m5", "m4", "m3", "m2", "m1", "decoy-lantern"],
+    "more_matches": {
+      "m6": 0,
+      "large": 3,
+      "small": 0,
+      "m5": 0,
+      "m4": 0,
+      "m3": 0,
+      "m2": 0,
+      "m1": 0,
+      "decoy-lantern": 0
+    },
+    "lanes_per_document": {
+      "m6": ["l1"],
+      "large": ["l2", "l3", "l1"],
+      "small": ["l3"],
+      "m5": ["l1"],
+      "m4": ["l1"],
+      "m3": ["l1"],
+      "m2": ["l1"],
+      "m1": ["l1"],
+      "decoy-lantern": ["l1"]
+    },
+    "around": {
+      "m6": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "large": {
+        "paragraphs": [2, 3, 4, 5, 6],
+        "found_marked": [4],
+        "in_other_document": 0
+      },
+      "small": {
+        "paragraphs": [1, 2, 3],
+        "found_marked": [2],
+        "in_other_document": 0
+      },
+      "m5": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "m4": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "m3": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "m2": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "m1": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "decoy-lantern": {
+        "paragraphs": [18, 19, 20, 21, 22],
+        "found_marked": [20],
+        "in_other_document": 0
+      }
+    },
+    "oversized": [
+      {
+        "document": "huge",
+        "bytes": 284515,
+        "lanes": ["l1"]
+      }
+    ],
+    "units": [
+      {
+        "lane": "l1",
+        "source": 0,
+        "after": 1787347295456,
+        "before": 1792531295456,
+        "state": "done",
+        "pages": 9,
+        "issue": null
+      },
+      {
+        "lane": "l2",
+        "source": 0,
+        "after": 1787347295456,
+        "before": 1792531295456,
+        "state": "done",
+        "pages": 1,
+        "issue": null
+      },
+      {
+        "lane": "l3",
+        "source": 0,
+        "after": 1787347295456,
+        "before": 1792531295456,
+        "state": "done",
+        "pages": 2,
+        "issue": null
+      }
+    ],
+    "pages_total": 12,
+    "usage": {
+      "reads": 13,
+      "read_bytes": 432246,
+      "around_bytes": 8171,
+      "items": 9,
+      "elapsed_ms": 102
+    },
+    "totals": {
+      "reads": 13,
+      "read_bytes": 432246,
+      "items": 9,
+      "invocations": 1
+    },
+    "complete": true,
+    "has_checkpoint": false,
+    "stopped": null,
+    "matched_again": 0,
+    "found_all_readable": false,
+    "unexpected_found": ["decoy-lantern"],
+    "charged_bytes": 432246,
+    "found_document_bytes": 431744,
+    "charged_per_document_byte": 1,
+    "oversized_ok": true,
+    "around_matches_label": {
+      "large": true,
+      "m1": true,
+      "m2": true,
+      "m3": true,
+      "m4": true,
+      "m5": true,
+      "m6": true,
+      "small": true
+    },
+    "relevant_paragraphs": 12,
+    "relevant_paragraphs_in_around": 8,
+    "relevant_paragraphs_not_surfaced": ["large:p60", "large:p150", "large:p215"]
+  },
+  "around_five_each_side": {
+    "large": {
+      "paragraphs": [1, 2, 3, 4, 5, 6, 7, 8, 9],
+      "found_marked": [4],
+      "in_other_document": 0
+    },
+    "expected_large": [1, 2, 3, 4, 5, 6, 7, 8, 9],
+    "around_bytes": 17483
+  },
+  "resume_two_pages": {
+    "invocations": 7,
+    "per_invocation": [
+      {
+        "new_refs": 1,
+        "pages": 1,
+        "charged_bytes": 42881,
+        "complete": false
+      },
+      {
+        "new_refs": 1,
+        "pages": 2,
+        "charged_bytes": 144887,
+        "complete": false
+      },
+      {
+        "new_refs": 2,
+        "pages": 4,
+        "charged_bytes": 43015,
+        "complete": false
+      },
+      {
+        "new_refs": 1,
+        "pages": 6,
+        "charged_bytes": 42881,
+        "complete": false
+      },
+      {
+        "new_refs": 2,
+        "pages": 8,
+        "charged_bytes": 85762,
+        "complete": false
+      },
+      {
+        "new_refs": 2,
+        "pages": 10,
+        "charged_bytes": 72820,
+        "complete": false
+      },
+      {
+        "new_refs": 0,
+        "pages": 12,
+        "charged_bytes": 0,
+        "complete": true
+      }
+    ],
+    "references": 9,
+    "duplicates": 0,
+    "same_references_as_one_run": true,
+    "cumulative_charged_bytes": 432246,
+    "completed": true,
+    "stalled_without_a_read": false
+  }
+}
diff --git a/performance/projects/evidence-large-document/result.json b/performance/projects/evidence-large-document/result.json
new file mode 100644
index 0000000..c1eac95
--- /dev/null
+++ b/performance/projects/evidence-large-document/result.json
@@ -0,0 +1,195 @@
+{
+  "memo": "0.82.0",
+  "node": "v24.19.0",
+  "document_bytes": {
+    "large": 144829,
+    "huge": 284459,
+    "small": 76,
+    "decoy-lantern": 29865,
+    "decoy-garden": 29841,
+    "m1": 42829,
+    "m2": 42829,
+    "m3": 42829,
+    "m4": 42829,
+    "m5": 42829,
+    "m6": 42829
+  },
+  "default_run": {
+    "found": ["m6", "large", "small", "m5", "m4", "m3", "m2", "m1", "decoy-lantern"],
+    "lanes_per_document": {
+      "m6": ["l1"],
+      "large": ["l2", "l3", "l1"],
+      "small": ["l3"],
+      "m5": ["l1"],
+      "m4": ["l1"],
+      "m3": ["l1"],
+      "m2": ["l1"],
+      "m1": ["l1"],
+      "decoy-lantern": ["l1"]
+    },
+    "around": {
+      "m6": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "large": {
+        "paragraphs": [58, 59, 60, 61, 62],
+        "found_marked": [60],
+        "in_other_document": 0
+      },
+      "small": {
+        "paragraphs": [1, 2, 3],
+        "found_marked": [2],
+        "in_other_document": 0
+      },
+      "m5": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "m4": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "m3": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "m2": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "m1": {
+        "paragraphs": [28, 29, 30, 31, 32],
+        "found_marked": [30],
+        "in_other_document": 0
+      },
+      "decoy-lantern": {
+        "paragraphs": [18, 19, 20, 21, 22],
+        "found_marked": [20],
+        "in_other_document": 0
+      }
+    },
+    "oversized": [
+      {
+        "document": "huge",
+        "bytes": 284515,
+        "lanes": ["l1"]
+      }
+    ],
+    "units": [
+      {
+        "lane": "l1",
+        "source": 0,
+        "after": 1787326090153,
+        "before": 1792510090153,
+        "state": "done",
+        "pages": 9,
+        "issue": null
+      },
+      {
+        "lane": "l2",
+        "source": 0,
+        "after": 1787326090153,
+        "before": 1792510090153,
+        "state": "done",
+        "pages": 1,
+        "issue": null
+      },
+      {
+        "lane": "l3",
+        "source": 0,
+        "after": 1787326090153,
+        "before": 1792510090153,
+        "state": "done",
+        "pages": 2,
+        "issue": null
+      }
+    ],
+    "pages_total": 12,
+    "usage": {
+      "reads": 13,
+      "read_bytes": 432246,
+      "around_bytes": 8171,
+      "items": 9,
+      "elapsed_ms": 104
+    },
+    "totals": {
+      "reads": 13,
+      "read_bytes": 432246,
+      "items": 9,
+      "invocations": 1
+    },
+    "complete": true,
+    "has_checkpoint": false,
+    "stopped": null,
+    "matched_again": 0,
+    "found_all_readable": false,
+    "unexpected_found": ["decoy-lantern"],
+    "charged_bytes": 432246,
+    "found_document_bytes": 431744,
+    "charged_per_document_byte": 1,
+    "oversized_ok": true,
+    "around_matches_label": {
+      "large": false,
+      "m1": true,
+      "m2": true,
+      "m3": true,
+      "m4": true,
+      "m5": true,
+      "m6": true,
+      "small": true
+    },
+    "relevant_paragraphs": 12,
+    "relevant_paragraphs_in_around": 8,
+    "relevant_paragraphs_not_surfaced": ["large:p4", "large:p150", "large:p215"]
+  },
+  "around_five_each_side": {
+    "large": {
+      "paragraphs": [55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65],
+      "found_marked": [60],
+      "in_other_document": 0
+    },
+    "expected_large": [1, 2, 3, 4, 5, 6, 7, 8, 9],
+    "around_bytes": 17887
+  },
+  "resume_two_pages": {
+    "invocations": 4,
+    "per_invocation": [
+      {
+        "new_refs": 1,
+        "pages": 1,
+        "charged_bytes": 42881,
+        "complete": false
+      },
+      {
+        "new_refs": 1,
+        "pages": 2,
+        "charged_bytes": 144887,
+        "complete": false
+      },
+      {
+        "new_refs": 2,
+        "pages": 4,
+        "charged_bytes": 43015,
+        "complete": false
+      },
+      {
+        "new_refs": 0,
+        "pages": 4,
+        "charged_bytes": 0,
+        "complete": false
+      }
+    ],
+    "references": 4,
+    "duplicates": 0,
+    "same_references_as_one_run": false,
+    "cumulative_charged_bytes": 230783,
+    "completed": false,
+    "stalled_without_a_read": true
+  }
+}