fix(wiki): first-occurrence wins when heading normalization collides pre-existing duplicates

Review of PR #130 found a real bug in the new normalizeHeading matching:
idx was built with idx[normalizeHeading(s.heading)] = i, plain map
assignment while iterating secsA — last write wins. A page that already
carries two on-disk sections differing only by case or whitespace (real,
pre-existing corpus drift, the exact pattern #106 itself cites) never
collided in idx before this normalization existed, since it keyed on the
exact string. After normalization they collide, and every future update
silently, deterministically routes to whichever duplicate happens to
sort last in the file — the first one is frozen forever, never
receiving an update again, with nothing to say so. Same bug class
parseSections's own doc comment already warns about (two keys sharing
one map slot means one loses lookup visibility), reintroduced here.

Fixed by keeping only the first occurrence's index when a collision
happens while building idx — deterministic, and matches "earliest is
canonical," the same reasoning behind every other on-disk-wins default
in this file.

New test reproduces the exact review finding with a positional
assertion (the fresh bullet must land adjacent to the first
occurrence's own content, not the second) — a Contains-only version of
this test would have passed even with the bug present, since nothing is
literally deleted, a bullet just silently routes to the wrong of two
still-present sections.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

README.md

@@ -104,6 +104,24 @@ Two fields are **generator-owned** and do get refreshed on a revise:
 `last_updated` and `source_hash`. Everything else on disk wins. The boundary is
 an explicit allowlist, so a field nobody has thought about stays human-owned.
 
+### Disagreement is reported, not resolved
+
+When the fresh extraction produces a *different version* of a section that
+already exists, the on-disk version still wins — but the run now says so:
+
+```
+wiki/sources/shape-up.md: source now disagrees on section "Summary" — on-disk version kept
+```
+
+That line lands in `log.md` with the rest of the run's warnings. Without it, "the
+source is unchanged" and "the source changed its mind and we ignored it" look
+identical from the outside.
+
+This is deliberately **textual** disagreement, not semantic contradiction: cheap,
+deterministic, no model on the write path. Deciding whether a disagreement is a
+genuine contradiction stays with the Judge-gated `brain-enrich` pass, which is
+approval-gated by design. Detection here, adjudication there.
+
 ## Re-ingesting an unchanged source is skipped
 
 Ingest stores `source_hash` — sha256 of the ingested **body**, frontmatter

ingestion/internal/pipeline/pipeline.go

@@ -62,13 +62,20 @@ func Run(ctx context.Context, cfg Config, brainDir, content, source string, dryR
 		}}, nil
 	}
 
+	// The source's own page path is deterministic (sourceSlug, computed above)
+	// even before extraction runs, unlike a concept/entity page's path, which
+	// only exists once the model has produced a title. So page-aware
+	// extraction (#106) is scoped to the source page: it's the one page every
+	// chunk's prompt can be given in advance.
+	existingPage := readSourcePage(brainDir, sourceSlug)
+
 	chunks := Chunk(content, cfg.ChunkSize)
 
 	var allRaw []RawPage
 	var allWarnings []string
 
 	for _, chunk := range chunks {
-		userPrompt := BuildPrompt(schema, source, chunk, inventory)
+		userPrompt := BuildPrompt(schema, source, chunk, inventory, existingPage)
 		output, err := cfg.Complete(ctx, systemPrompt, userPrompt)
 		if err != nil {
 			return Result{}, fmt.Errorf("LLM call: %w", err)
@@ -117,10 +124,18 @@ func buildAndWrite(rawPages []RawPage, sourceSlug, date, brainDir, source string
 			if err := os.MkdirAll(filepath.Dir(dest), 0o755); err != nil {
 				return Result{}, fmt.Errorf("mkdir for %s: %w", page.Path, err)
 			}
-			content, err := reviseOrCreate(dest, page)
+			content, conflicts, err := reviseOrCreate(dest, page)
 			if err != nil {
 				return Result{}, err
 			}
+			for _, heading := range conflicts {
+				// Reported, not resolved. The on-disk version stands; whether
+				// the disagreement is a genuine contradiction is the
+				// Judge-gated enrich pass's call, not the write path's.
+				warnings = append(warnings, fmt.Sprintf(
+					"%s: source now disagrees on section %q — on-disk version kept",
+					page.Path, heading))
+			}
 			if err := os.WriteFile(dest, []byte(content), 0o644); err != nil {
 				return Result{}, fmt.Errorf("write %s: %w", page.Path, err)
 			}
@@ -160,6 +175,16 @@ func hashBody(content string) string {
 	return hex.EncodeToString(sum[:])
 }
 
+// readSourcePage returns the full on-disk content of the source page at
+// wiki/sources/<sourceSlug>.md, or "" when it doesn't exist yet.
+func readSourcePage(brainDir, sourceSlug string) string {
+	raw, err := os.ReadFile(filepath.Join(brainDir, "wiki", "sources", sourceSlug+".md"))
+	if err != nil {
+		return ""
+	}
+	return string(raw)
+}
+
 // storedSourceHash reads source_hash from the source page a previous ingest
 // wrote, or "" when there is no such page or no such field.
 //
@@ -167,11 +192,7 @@ func hashBody(content string) string {
 // skip costs one extraction, while a wrong skip loses an update silently — and
 // silence is the failure mode this repo keeps paying for.
 func storedSourceHash(brainDir, sourceSlug string) string {
-	raw, err := os.ReadFile(filepath.Join(brainDir, "wiki", "sources", sourceSlug+".md"))
-	if err != nil {
-		return ""
-	}
-	for _, line := range strings.Split(string(raw), "\n") {
+	for _, line := range strings.Split(readSourcePage(brainDir, sourceSlug), "\n") {
 		if key, value, found := strings.Cut(line, ":"); found && strings.TrimSpace(key) == "source_hash" {
 			return strings.TrimSpace(value)
 		}
@@ -198,15 +219,16 @@ func storedSourceHash(brainDir, sourceSlug string) string {
 // its first-ingest date on a revised page. Left deliberately — the alternative
 // is this function editing frontmatter it does not own, and no lint reads that
 // field today.
-func reviseOrCreate(dest string, page wiki.Page) (string, error) {
+func reviseOrCreate(dest string, page wiki.Page) (string, []string, error) {
 	existing, err := os.ReadFile(dest)
 	switch {
 	case os.IsNotExist(err):
-		return page.Content, nil
+		return page.Content, nil, nil
 	case err != nil:
-		return "", fmt.Errorf("read existing %s: %w", page.Path, err)
+		return "", nil, fmt.Errorf("read existing %s: %w", page.Path, err)
 	}
-	return wiki.Merge(wiki.Page{Path: page.Path, Content: string(existing)}, page).Content, nil
+	merged, conflicts := wiki.MergeReport(wiki.Page{Path: page.Path, Content: string(existing)}, page)
+	return merged.Content, conflicts, nil
 }
 
 // mergeAll deduplicates pages by path, merging content from later occurrences.

ingestion/internal/pipeline/pipeline_test.go

@@ -8,6 +8,7 @@ import (
 	"net/http/httptest"
 	"os"
 	"path/filepath"
+	"strings"
 	"testing"
 	"time"
 
@@ -276,6 +277,104 @@ func TestRun_ReExtractsWhenTheSourceBodyChanged(t *testing.T) {
 	assert.Equal(t, 2, calls, "changed content must be re-extracted")
 }
 
+// --- #106: page-aware extraction — the model sees the page it is revising --
+
+func TestRun_ExistingSourcePageIsIncludedInTheExtractionPrompt(t *testing.T) {
+	// #103 made the merge asymmetric (on-disk wins); #106 is the other half —
+	// the extraction itself gets to see what's already there instead of
+	// producing a page blind and letting the merge reconcile it afterwards.
+	// This only proves the existing page reaches the prompt; the byte-identical
+	// guarantee for an echoed-back revision is a separate test below.
+	dir := t.TempDir()
+	for _, sub := range []string{"wiki/concepts", "wiki/entities", "wiki/sources"} {
+		require.NoError(t, os.MkdirAll(filepath.Join(dir, sub), 0o755))
+	}
+
+	seeded := "---\ntitle: 'Shape Up'\n---\n\n## Summary\n\nA book about shaping product work.\n\n## Key Claims\n\n- Appetite beats estimation.\n"
+	page := filepath.Join(dir, "wiki", "sources", "shape-up.md")
+	require.NoError(t, os.WriteFile(page, []byte(seeded), 0o644))
+
+	var gotPrompt string
+	cfg := Config{Complete: func(_ context.Context, _, user string) (string, error) {
+		gotPrompt = user
+		return mustJSON([]RawPage{{
+			Title: "Shape Up", Type: "source", Subtype: "book",
+			Content: "## Summary\n\nA book about shaping product work.\n",
+		}}), nil
+	}}
+
+	// Deliberately does not mention "Appetite beats estimation." — that phrase
+	// can only reach the prompt via the existing on-disk page, not the source
+	// content, so its presence proves the page was threaded through.
+	content := "A short editorial update to the book's listing. No new claims here."
+	_, err := Run(context.Background(), cfg, dir, content, "shape-up", false)
+	require.NoError(t, err)
+
+	assert.Contains(t, gotPrompt, "Appetite beats estimation.",
+		"the existing page's content must be visible to the extraction prompt")
+	assert.Contains(t, gotPrompt, "revis", "the prompt must ask for a revision, not a fresh page")
+}
+
+func TestRun_NewSourceHasNoExistingPageInThePrompt(t *testing.T) {
+	// The other side of the same behavior: a source with no page on disk yet
+	// must not get an "existing page" section injected into its prompt at all.
+	dir := t.TempDir()
+	for _, sub := range []string{"wiki/concepts", "wiki/entities", "wiki/sources"} {
+		require.NoError(t, os.MkdirAll(filepath.Join(dir, sub), 0o755))
+	}
+
+	var gotPrompt string
+	cfg := Config{Complete: func(_ context.Context, _, user string) (string, error) {
+		gotPrompt = user
+		return mustJSON([]RawPage{{
+			Title: "New Thing", Type: "source", Subtype: "article",
+			Content: "## Summary\n\nBrand new.\n",
+		}}), nil
+	}}
+
+	_, err := Run(context.Background(), cfg, dir, "Brand new content.", "new-thing", false)
+	require.NoError(t, err)
+
+	assert.NotContains(t, gotPrompt, "Current page",
+		"a source with no page on disk yet must not get a revision section in its prompt")
+}
+
+func TestRun_RevisionEchoingExistingPageStaysByteIdentical(t *testing.T) {
+	// The issue's explicit risk: a model given the current page will sometimes
+	// echo it back verbatim. That must not duplicate content or otherwise
+	// perturb the page — the deterministic merge is still the backstop.
+	dir := t.TempDir()
+	for _, sub := range []string{"wiki/concepts", "wiki/entities", "wiki/sources"} {
+		require.NoError(t, os.MkdirAll(filepath.Join(dir, sub), 0o755))
+	}
+
+	seeded := "---\ntitle: 'Shape Up'\n---\n\n## Summary\n\nA book about shaping product work.\n\n## Key Claims\n\n- Appetite beats estimation.\n"
+	page := filepath.Join(dir, "wiki", "sources", "shape-up.md")
+	require.NoError(t, os.WriteFile(page, []byte(seeded), 0o644))
+
+	cfg := Config{Complete: func(_ context.Context, _, _ string) (string, error) {
+		return mustJSON([]RawPage{{
+			Title: "Shape Up", Type: "source", Subtype: "book",
+			Content: "## Summary\n\nA book about shaping product work.\n\n## Key Claims\n\n- Appetite beats estimation.\n",
+		}}), nil
+	}}
+
+	content := "A short editorial update to the book's listing. No new claims here."
+	result, err := Run(context.Background(), cfg, dir, content, "shape-up", false)
+	require.NoError(t, err)
+	require.False(t, result.Skipped)
+
+	got, err := os.ReadFile(page)
+	require.NoError(t, err)
+
+	want := "---\ntitle: 'Shape Up'\nlast_updated: " + time.Now().UTC().Format("2006-01-02") +
+		"\nsource_hash: " + hashBody(content) + "\n---\n\n" +
+		"## Summary\n\nA book about shaping product work.\n\n" +
+		"## Key Claims\n\n- Appetite beats estimation.\n"
+	assert.Equal(t, want, string(got),
+		"a revision pass that echoes the existing page back must not duplicate or alter its content")
+}
+
 func TestRun_FrontmatterOnlyChangeIsNotDrift(t *testing.T) {
 	// The hash covers the body, so re-fetching a page whose metadata moved
 	// (a re-crawl stamping a new date) is not treated as the source changing.
@@ -293,3 +392,34 @@ func TestRun_FrontmatterOnlyChangeIsNotDrift(t *testing.T) {
 
 	assert.Equal(t, 1, calls, "only the body counts as the source")
 }
+
+func TestRunRaw_ReportsWhenTheSourceDisagreesWithWhatIsOnDisk(t *testing.T) {
+	// The on-disk version still wins — that is what makes unattended ingest
+	// safe. But the run must say that the source now claims something else,
+	// or a changed source is indistinguishable from an unchanged one.
+	dir := t.TempDir()
+	page := filepath.Join(dir, "wiki", "sources", "shape-up.md")
+	require.NoError(t, os.MkdirAll(filepath.Dir(page), 0o755))
+	require.NoError(t, os.WriteFile(page, []byte(`---
+title: 'Shape Up'
+---
+
+## Summary
+
+Appetite is fixed, scope is variable.
+`), 0o644))
+
+	res, err := RunRaw(dir, "shape-up", []RawPage{{
+		Title: "Shape Up", Type: "source", Subtype: "book", Domain: "product",
+		Content: "## Summary\n\nAppetite is negotiable and scope is fixed.\n",
+	}}, false)
+	require.NoError(t, err)
+
+	got, err := os.ReadFile(page)
+	require.NoError(t, err)
+	assert.Contains(t, string(got), "Appetite is fixed", "on-disk version wins")
+
+	joined := strings.Join(res.Warnings, " ")
+	assert.Contains(t, joined, "Summary", "the disagreeing section must be named")
+	assert.Contains(t, joined, "shape-up", "and the page it happened on")
+}

ingestion/internal/pipeline/prompt.go

@@ -22,8 +22,18 @@ Each element must have exactly these fields:
 Wikilinks in content: [[Display Name]] — just the display name, no slug, no pipe separator.
 Only link to pages listed in the inventory or pages you are creating in this response.`
 
-// BuildPrompt constructs the user prompt for a single chunk.
-func BuildPrompt(schema, source, content string, inventory map[wiki.PageType][]wiki.Entry) string {
+// BuildPrompt constructs the user prompt for a single chunk. existingPage is
+// the full current content (frontmatter + body) of the source's own wiki page
+// when one already exists on disk, or "" for a source being ingested for the
+// first time. When non-empty, the model is asked to revise that page rather
+// than write a fresh one — see hyperguild#106. The deterministic merge in
+// wiki.Merge remains the backstop regardless of what the model returns: an
+// on-disk section the model doesn't touch survives untouched, and a section
+// it does touch only takes effect where Merge's strategy allows the fresh
+// side to change anything (bullet union, append) — never for an ordinary
+// section, so the model proposing a revision is never how a page gets
+// overwritten.
+func BuildPrompt(schema, source, content string, inventory map[wiki.PageType][]wiki.Entry, existingPage string) string {
 	var sb strings.Builder
 
 	fmt.Fprintf(&sb, "Today's date is %s.\n\n", time.Now().UTC().Format("2006-01-02"))
@@ -49,6 +59,18 @@ func BuildPrompt(schema, source, content string, inventory map[wiki.PageType][]w
 		sb.WriteString("\n")
 	}
 
+	if existingPage != "" {
+		sb.WriteString("## Current page for this source — revise it, don't restart\n\n")
+		sb.WriteString("This source has already been ingested and has the wiki page shown below.\n")
+		sb.WriteString("You are revising that page, not writing a fresh one:\n")
+		sb.WriteString("  - Add sections or claims the source below supports that this page doesn't have yet.\n")
+		sb.WriteString("  - Strengthen a claim where the source gives more detail, evidence, or nuance than what's already there.\n")
+		sb.WriteString("  - Leave unchanged whatever the source doesn't affect — reuse the existing wording verbatim rather than paraphrasing it, so an unchanged claim is recognized as unchanged.\n")
+		sb.WriteString("  - A merge step keeps this page's on-disk content for any section you don't return, so it's fine to omit a section you aren't revising.\n\n")
+		sb.WriteString(existingPage)
+		sb.WriteString("\n\n")
+	}
+
 	sb.WriteString("## Non-negotiable rules\n\n")
 	sb.WriteString("1. Output ONLY a valid JSON array — no prose, no fences.\n")
 	sb.WriteString("2. Fields: title, type, subtype (if applicable), domain (if applicable), content.\n")

ingestion/internal/wiki/merge.go

@@ -37,32 +37,100 @@ type section struct {
 // Sections are merged by strategy: bullet sections union unique lines,
 // append sections concatenate, all others keep a's version. Sections in b
 // not present in a are appended.
+// Merge discards any disagreement silently; prefer MergeReport where the
+// caller can act on it.
 func Merge(a, b Page) Page {
+	page, _ := MergeReport(a, b)
+	return page
+}
+
+// MergeReport is Merge, plus the headings of sections where b carried a
+// different version of a section a already had — and lost.
+//
+// That loss is the property which makes ingest safe to run unattended (#103):
+// a regenerating writer can add to a note but never rewrite it. But the
+// discarded text is also a signal, and until now it went in the bin: the source
+// now says something different about a section we already hold. Nobody could
+// tell "the source is unchanged" from "the source changed its mind and we
+// ignored it".
+//
+// Deliberately NOT a semantic contradiction check. This reports textual
+// disagreement, which is cheap, deterministic, and needs no model on the write
+// path. Deciding whether a disagreement is a genuine contradiction stays with
+// the Judge-gated enrich pass that already exists (brain-gardener#175), which
+// is approval-gated by design — detection here, adjudication there.
+func MergeReport(a, b Page) (Page, []string) {
 	fmA, preambleA, secsA := parseSections(a.Content)
 	fmB, _, secsB := parseSections(b.Content)
 
+	var conflicts []string
+
+	// First occurrence wins on a collision — a page can already carry two
+	// on-disk sections that only differ by case or whitespace (the exact
+	// corpus drift this normalization exists to catch going forward, found
+	// pre-existing in review of PR #130): a plain last-write assignment
+	// here would silently and deterministically route every future update
+	// to whichever duplicate happens to sort last in the file, freezing
+	// the other one forever with nothing to say so.
 	idx := make(map[string]int, len(secsA))
 	for i, s := range secsA {
-		idx[s.heading] = i
+		key := normalizeHeading(s.heading)
+		if _, exists := idx[key]; !exists {
+			idx[key] = i
+		}
 	}
 
 	for _, sB := range secsB {
-		i, exists := idx[sB.heading]
+		key := normalizeHeading(sB.heading)
+		i, exists := idx[key]
 		if !exists {
-			idx[sB.heading] = len(secsA)
+			idx[key] = len(secsA)
 			secsA = append(secsA, sB)
 			continue
 		}
 		sA := secsA[i]
 		switch {
-		case bulletSections[sB.heading]:
+		case bulletSectionsNorm[key]:
 			secsA[i].content = mergeBullets(sA.content, sB.content)
-		case appendSections[sB.heading]:
+		case appendSectionsNorm[key]:
 			secsA[i].content = appendNew(sA.content, sB.content)
+		default:
+			// a wins — but say that b disagreed, rather than dropping it mute.
+			if canonicalBlock(sA.content) != canonicalBlock(sB.content) {
+				conflicts = append(conflicts, sB.heading)
+			}
 		}
 	}
 
-	return Page{Path: a.Path, Content: rebuildContent(refreshOwned(fmA, fmB), preambleA, secsA)}
+	return Page{Path: a.Path, Content: rebuildContent(refreshOwned(fmA, fmB), preambleA, secsA)}, conflicts
+}
+
+// normalizeHeading returns the key Merge uses to decide "is this the same
+// section" — trimmed and case-folded, so "Key Claims" and "Key claims" match
+// (hyperguild#106: heading casing drift must not fork a section). It is used
+// only for that comparison; the on-disk section's original heading text
+// (secsA[i].heading, untouched here) is what actually gets written out, so a
+// human's heading casing is never silently rewritten by a merge.
+func normalizeHeading(h string) string {
+	return strings.ToLower(strings.TrimSpace(h))
+}
+
+// bulletSectionsNorm and appendSectionsNorm are bulletSections/appendSections
+// keyed by normalizeHeading instead of exact text, so a section's merge
+// strategy (union bullets, append) is chosen by the same normalized identity
+// Merge uses to match it — otherwise a heading that matched an on-disk
+// section (case-insensitively) could still fail the exact-string strategy
+// lookup and silently fall through to "keep a's version," dropping b's
+// content instead of unioning or appending it.
+var bulletSectionsNorm = normalizeSet(bulletSections)
+var appendSectionsNorm = normalizeSet(appendSections)
+
+func normalizeSet(names map[string]bool) map[string]bool {
+	out := make(map[string]bool, len(names))
+	for k := range names {
+		out[normalizeHeading(k)] = true
+	}
+	return out
 }
 
 // generatorOwned names the frontmatter fields the ingest pipeline may refresh
@@ -301,6 +369,21 @@ func canonicalLine(line string) string {
 	return strings.Join(strings.Fields(line), " ")
 }
 
+// canonicalBlock applies canonicalLine per line of a multi-line section body
+// — the same whitespace-insensitivity appendNew/mergeBullets already give a
+// single line, extended to a whole section for MergeReport's disagreement
+// check (found in review, PR #129: comparing raw strings.TrimSpace only
+// tolerated leading/trailing whitespace, not a re-extraction's mid-line
+// spacing, which every other comparison in this file already tolerates).
+func canonicalBlock(block string) string {
+	lines := strings.Split(block, "\n")
+	out := make([]string, len(lines))
+	for i, l := range lines {
+		out[i] = canonicalLine(l)
+	}
+	return strings.Join(out, "\n")
+}
+
 // appendNew concatenates the lines of b that a does not already carry.
 //
 // Append sections used to concatenate wholesale, which is correct exactly once.

ingestion/internal/wiki/merge_test.go

@@ -305,6 +305,68 @@ Rewritten by the model.
 	assert.Equal(t, want, got)
 }
 
+// --- hyperguild#106: heading casing drift must not fork a section ----------
+// Page-aware extraction asks the model to revise the page it's given, but
+// nothing guarantees the model reproduces a heading's exact on-disk casing
+// ("Key Claims" vs "Key claims"). Without normalizing the match key, that
+// drift silently creates a near-duplicate section instead of updating the
+// original — the failure mode #106 exists to close.
+
+func TestMerge_HeadingCaseDriftMatchesSameSection(t *testing.T) {
+	a := Page{Path: "p.md", Content: "---\ntitle: X\n---\n\n## Key Claims\n\n- Appetite beats estimation.\n"}
+	b := Page{Path: "p.md", Content: "---\ntitle: X\n---\n\n## Key claims\n\n- New pattern holds under load.\n"}
+
+	got := Merge(a, b).Content
+
+	assert.Equal(t, 1, strings.Count(got, "## Key"),
+		"casing drift in the heading must not create a second section")
+	assert.Contains(t, got, "Appetite beats estimation.")
+	assert.Contains(t, got, "New pattern holds under load.",
+		"the differently-cased section's content must still be unioned in")
+	assert.Contains(t, got, "## Key Claims",
+		"the on-disk heading's original casing must be preserved in the output")
+	assert.NotContains(t, got, "## Key claims",
+		"the fresh extraction's heading casing must not overwrite the on-disk display text")
+}
+
+func TestMerge_HeadingWhitespaceDriftMatchesSameSection(t *testing.T) {
+	a := Page{Path: "p.md", Content: "---\ntitle: X\n---\n\n## Open Questions \n\n- Does appetite scale?\n"}
+	b := Page{Path: "p.md", Content: "---\ntitle: X\n---\n\n##  Open Questions\n\n- What about remote teams?\n"}
+
+	got := Merge(a, b).Content
+
+	assert.Equal(t, 1, strings.Count(got, "## Open Questions"),
+		"surrounding whitespace drift in the heading must not create a second section")
+	assert.Contains(t, got, "Does appetite scale?")
+	assert.Contains(t, got, "What about remote teams?")
+}
+
+func TestMerge_PreExistingCaseDuplicateHeadingsRouteToTheFirstOccurrence(t *testing.T) {
+	// Found in review of PR #130: normalizeHeading's idx map is built by
+	// iterating secsA with idx[normalizeHeading(s.heading)] = i -- last
+	// write wins. Before heading normalization existed, two on-disk
+	// sections differing only by case (real, pre-existing corpus drift --
+	// the exact pattern #106 itself cites) never collided, since idx keyed
+	// on the exact string. After normalization they collide, and every
+	// fresh update silently, deterministically routes to whichever
+	// duplicate happens to appear LAST in the file -- the first one is
+	// frozen forever, never receiving updates again, with nothing to say
+	// so. A Contains-only check doesn't catch this: nothing is literally
+	// deleted, a bullet just lands under the wrong of two still-present
+	// headings. Assert it lands under the FIRST occurrence instead —
+	// deterministic and matches "earliest is canonical," same reasoning
+	// as every other on-disk-wins default in this file.
+	a := Page{Path: "p.md", Content: "---\ntitle: X\n---\n\n## Key Claims\n\n- claim one\n\n## KEY CLAIMS\n\n- claim two\n"}
+	b := Page{Path: "p.md", Content: "---\ntitle: X\n---\n\n## Key Claims\n\n- claim one\n- fresh claim\n"}
+
+	got := Merge(a, b).Content
+
+	assert.Contains(t, got, "## Key Claims\n\n- claim one\n- fresh claim\n",
+		"the fresh bullet must land in the first-occurrence section, not silently route to the last duplicate")
+	assert.Contains(t, got, "## KEY CLAIMS\n\n- claim two\n",
+		"the second duplicate is untouched by this update, not merged into or overwritten")
+}
+
 func TestRefreshOwned_CommentLinesSurviveUnmangled(t *testing.T) {
 	// Third review round on PR #126: parseFrontmatterBlock dropped the
 	// sibling api package's comment guard (strings.HasPrefix(trimmed, "#")).
@@ -332,3 +394,85 @@ func TestRefreshOwned_IndentedListItemsWithColonsAreNotMisparsedAsKeys(t *testin
 	got := refreshOwned(a, b)
 	assert.Equal(t, "---\naliases:\n  - 'Chapter 3: The Reckoning'\ntitle: 'X'\nsource_hash: bbbb\n---\n", got)
 }
+
+func TestMerge_ReportsSectionsWhereTheFreshVersionWasDiscarded(t *testing.T) {
+	// The asymmetry that makes ingest safe (on-disk wins) also throws away a
+	// real signal: the source now says something different about a section we
+	// already have. Today that disagreement is silently discarded, so nobody
+	// can tell "the source is unchanged" from "the source changed its mind and
+	// we ignored it".
+	existing := Page{Path: "wiki/sources/x.md", Content: `---
+title: 'X'
+---
+
+## Summary
+
+Appetite is fixed, scope is variable.
+
+## Method
+
+Six-week cycles.
+`}
+	fresh := Page{Path: "wiki/sources/x.md", Content: `---
+title: 'X'
+---
+
+## Summary
+
+Appetite is negotiable and scope is fixed.
+
+## Method
+
+Six-week cycles.
+`}
+
+	got, conflicts := MergeReport(existing, fresh)
+
+	assert.Contains(t, got.Content, "Appetite is fixed", "on-disk version still wins")
+	assert.NotContains(t, got.Content, "negotiable")
+	assert.Equal(t, []string{"Summary"}, conflicts,
+		"only the section that actually disagreed is reported")
+}
+
+func TestMerge_ReportsNothingWhenTheFreshVersionAgrees(t *testing.T) {
+	page := Page{Path: "wiki/sources/x.md", Content: `---
+title: 'X'
+---
+
+## Summary
+
+Appetite is fixed, scope is variable.
+`}
+
+	_, conflicts := MergeReport(page, page)
+	assert.Empty(t, conflicts, "identical content is not a disagreement")
+}
+
+func TestMerge_ReportsNothingForInternalWhitespaceVariantsOnly(t *testing.T) {
+	// Found in review of PR #129: the conflict comparison used raw
+	// strings.TrimSpace, which only tolerates LEADING/TRAILING whitespace —
+	// unlike appendNew/mergeBullets, which already canonicalize internal
+	// whitespace runs too (#103: a re-extraction's spacing isn't guaranteed
+	// byte-stable). A mid-line double-space from re-extraction would
+	// false-positive as a "disagreement" here even though every other
+	// whitespace-sensitive comparison in this file already tolerates it.
+	existing := Page{Path: "wiki/sources/x.md", Content: `---
+title: 'X'
+---
+
+## Summary
+
+Appetite  is fixed,   scope is variable.
+`}
+	fresh := Page{Path: "wiki/sources/x.md", Content: `---
+title: 'X'
+---
+
+## Summary
+
+Appetite is fixed, scope is variable.
+`}
+
+	_, conflicts := MergeReport(existing, fresh)
+	assert.Empty(t, conflicts, "internal whitespace differences alone are not a disagreement")
+}