feat(shapes): a component file is a candidate shape in its own right (milestone 439 step 5)

A single-file component defines `.card`, `.title` and a `Props`, never
anything named after itself — so the ledger could not hold "StatusChip is
canon" and could not say a new card was built where one already existed.

- coverage.is_file_unit: a file that RENDERS (its markup names classes, or it
  opens a <script>/<template>/<style> block) and defines nothing named after
  its stem gets one `file` row. Structural, not a framework list: Svelte and
  Vue components qualify, a TSX component already has its function's row, a
  module is accounted for by its definitions. Emitting every file would have
  put every module of every project in the todo at once.
- No body fingerprint on file rows, so an edit never re-asks a judgment.
- `file` is its own form and family; derive grouping skips it (`index`,
  `+page` repeat by convention). Divergence buckets it on its own, so a new
  component where a component canon dominates is a fair question.
- mark_canonicals: a snippet recorded at a path with no symbol makes that
  file row canonical; it still covers no definition inside.
- The write hooks apply the same test before noting a new file for the
  end-of-turn question.

Plugin version minted.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

plugin/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "scribe",
   "description": "Scribe for Claude Code: connects the scribe MCP server, adds the hooks that deliver live project state and relevant records at the right moment, ships the shared client-neutral Scribe skills (using-scribe, writing-plans, reporting-back, systematic-debugging, verification, brainstorming, reusing-code, shape-accounting), and syncs your saved Scribe Processes as skills (/scribe:sync).",
-  "version": "2026.10.01.1242",
+  "version": "2026.10.01.1245",
   "author": {
     "name": "Bryan Van Deusen"
   },

plugin/hooks/scribe_after_write.sh

@@ -141,7 +141,7 @@ while IFS= read -r rel_path; do
   names=$(printf '%s' "$code" | scribe_defs | sort -u | head -12 || true)
   # What this write defined, for the end-of-turn question (milestone 439) —
   # recorded before the arms below, which skip a write that defines nothing.
-  printf '%s\n' "$names" | scribe_written_append "$safe_sid" "$rel_path" "$fresh"
+  printf '%s\n' "$names" | scribe_written_append "$safe_sid" "$rel_path" "$fresh" "$code"
   # Nothing DEFINED in what was written (prose, data, a call-site edit) →
   # nothing to say; the arms are about shapes.
   [ -n "$names" ] || continue

plugin/hooks/scribe_defs.sh

@@ -1163,22 +1163,32 @@ SCRIBE_LEDGER_DIRS="scribe-priorart scribe-autoinject"
 # end of the turn (scribe_shape_check.sh). Its own file — one file, one
 # question (lesson #4226) — under the prior-art directory, so the session
 # clear above already sweeps it. Read `kind<TAB>name` lines on stdin (the
-# scribe_defs format). A NEW file adds one `file` line named by its stem: a
-# component, module or package file is a candidate in its own right, in any
-# language, without a framework rule saying so.
+# scribe_defs format).
+#
+# A NEW file that is a unit in its own right adds one `file` line named by its
+# stem — the same test the server's sync applies (coverage.is_file_unit): it
+# RENDERS (markup names classes, or it opens a <script>/<template>/<style>
+# block) and nothing inside it is named after it. That is a component in any
+# framework, without a list of frameworks.
 #   $1 session id (already made filename-safe)   $2 repo-relative path
 #   $3 "new" when the file did not exist before this write
+#   $4 the code written (to test whether the file renders)
 scribe_written_append() {
-  local sid="$1" rel="$2" fresh="${3:-}" dir stem
+  local sid="$1" rel="$2" fresh="${3:-}" code="${4:-}" dir stem names
   if [ -z "$sid" ] || [ -z "$rel" ]; then cat >/dev/null; return 0; fi
   dir="${TMPDIR:-/tmp}/scribe-priorart"
   mkdir -p "$dir" 2>/dev/null || true
+  names=$(cat)
   {
     if [ "$fresh" = "new" ]; then
       stem=${rel##*/}; stem=${stem%.*}
-      [ -n "$stem" ] && printf '%s\tfile\t%s\n' "$rel" "$stem"
+      if [ -n "$stem" ] \
+        && ! printf '%s\n' "$names" | awk -F'\t' -v s="$stem" '$1 == "sym" && $2 == s { f = 1 } END { exit !f }' \
+        && printf '%s\n' "$code" | grep -qE '^[[:space:]]*<(script|template|style)([[:space:]>]|$)|(class|className)[[:space:]]*='; then
+        printf '%s\tfile\t%s\n' "$rel" "$stem"
       fi
-    awk -F'\t' -v p="$rel" 'NF >= 2 && ($1 == "sym" || $1 == "css") && $2 != "" { printf "%s\t%s\t%s\n", p, $1, $2 }'
+    fi
+    printf '%s\n' "$names" | awk -F'\t' -v p="$rel" 'NF >= 2 && ($1 == "sym" || $1 == "css") && $2 != "" { printf "%s\t%s\t%s\n", p, $1, $2 }'
   } >> "$dir/${sid}.written.ids" 2>/dev/null || true
   return 0
 }

plugin/hooks/scribe_prior_art.sh

@@ -202,12 +202,13 @@ rule_exclude_q=""
 if [ -n "$session_id" ]; then
   safe_sid=$(printf '%s' "$session_id" | tr -c 'A-Za-z0-9._-' '_')
   # What this write defined, for the end-of-turn question (milestone 439). A
-  # Write onto a path that does not exist yet creates a file — a candidate in
-  # its own right. Written before any server call: the ask at the end of the
-  # turn must not depend on this write's hint having been answered.
+  # Write onto a path that does not exist yet creates a file, which may be a
+  # unit in its own right (see scribe_written_append). Written before any
+  # server call: the ask at the end of the turn must not depend on this
+  # write's hint having been answered.
   fresh=""
   [ -e "$file_path" ] || fresh="new"
-  printf '%s\n' "$shapes" | scribe_written_append "$safe_sid" "$rel_path" "$fresh"
+  printf '%s\n' "$shapes" | scribe_written_append "$safe_sid" "$rel_path" "$fresh" "$code"
   idfile="$state_dir/${safe_sid}.ids"
   syncfile="$state_dir/${safe_sid}.sync.ids"
   derivefile="$state_dir/${safe_sid}.derive.ids"

src/scribe/services/coverage.py

@@ -597,6 +597,42 @@ def scannable(path: str) -> bool:
     return not path.lower().endswith(_SKIP_SUFFIXES)
 
 
+# --- the file as a unit (milestone 439) ---------------------------------------
+#
+# A component is the thing people reuse in a UI, and in a single-file
+# component nothing inside is named after it: CatalogueBookCard.svelte defines
+# `.card`, `.title` and a `Props`, never `CatalogueBookCard`. A ledger of
+# definitions alone therefore could never say "a new card was built where
+# CatalogueBookCard is canon". So such a file gets a row of its own.
+#
+# WHICH FILES, decided by structure rather than by a framework list: a file
+# that RENDERS — its markup names classes, or it opens with a <script>,
+# <template> or <style> block — and that DEFINES NOTHING NAMED AFTER ITSELF.
+# A TSX `function CatalogueBookCard` already has its row; a module of helpers
+# is accounted for by its helpers. Emitting a row for every file would put
+# every module in every project into the todo at once, which is a flood, not
+# a question. No body fingerprint either: every edit would otherwise ask for
+# the file's judgment to be re-confirmed.
+_RENDER_BLOCK_RE = re.compile(r"^\s*<(?:script|template|style)\b", re.IGNORECASE | re.MULTILINE)
+
+
+def file_stem(path: str) -> str:
+    """`web/src/lib/StatusChip.svelte` → `StatusChip`."""
+    return path.rsplit("/", 1)[-1].rsplit(".", 1)[0]
+
+
+def is_file_unit(
+    path: str, text: str, defs: list[Definition], refs: dict[str, int] | None = None,
+) -> bool:
+    """Is this file a candidate shape in its own right? See the block above."""
+    stem = file_stem(path)
+    if not stem:
+        return False
+    if any(d.name == stem for d in defs if d.kind == "sym"):
+        return False
+    return bool(refs) or _RENDER_BLOCK_RE.search(text) is not None
+
+
 class ArchiveShape(NamedTuple):
     """A definition located in a repo archive — what the sync upserts and
     the proposer matches. The leading (path, kind, name) triple is the
@@ -658,6 +694,11 @@ def scan_archive(blob: bytes) -> ArchiveScan:
                 continue
             defs = extract_definitions(text)
             scoped = scoped_definitions(path, text, defs)
+            refs = class_references(path, text)
+            if is_file_unit(path, text, defs, refs):
+                shapes.append(ArchiveShape(
+                    path, "file", file_stem(path), f"file {path}", "", "",
+                ))
             shapes.extend(
                 ArchiveShape(
                     path, d.kind, d.name, d.signature, d.body_sha, d.body,
@@ -665,7 +706,6 @@ def scan_archive(blob: bytes) -> ArchiveScan:
                 )
                 for d in defs
             )
-            refs = class_references(path, text)
             if refs:
                 references[path] = refs
     return ArchiveScan(shapes, references)

src/scribe/services/shape_ledger.py

@@ -408,6 +408,11 @@ async def mark_canonicals(
     fall back to unclassified. Agent judgments are never overwritten.
     """
     usable = [(nid, p, s) for nid, p, s in recorded if (s or "").strip()]
+    # A location naming a FILE and no symbol is a whole-file record: it makes
+    # no claim about a definition inside (`location_covers`), but it is
+    # exactly the claim a `file` row asks about (milestone 439).
+    whole_files = {(p or "").strip(): nid for nid, p, s in recorded
+                   if (p or "").strip() and not (s or "").strip()}
     now = datetime.now(timezone.utc)
     async with async_session() as session:
         rows = (
@@ -426,6 +431,8 @@ async def mark_canonicals(
                 ),
                 None,
             )
+            if covering is None and row.kind == "file":
+                covering = whole_files.get(row.path)
             if covering is not None and row.status in _MECHANICAL_TODO:
                 await _judge(session, row, status="canonical", snippet_id=covering,
                              by="mechanical", reason=None, at=now)
@@ -991,6 +998,10 @@ def shape_form(signature: str, kind: str = "sym") -> str:
     """
     if kind == "css":
         return "css"
+    if kind == "file":
+        # A whole file (milestone 439): its own family, comparable only with
+        # other files — a component is never told to build from a helper.
+        return "file"
     sig = (signature or "").strip()
     if not sig:
         return FORM_UNKNOWN
@@ -1085,6 +1096,8 @@ def shape_family(form: str) -> str:
         return "value"
     if form == "css":
         return "css"
+    if form == "file":
+        return "file"
     return ""
 
 
@@ -2040,6 +2053,10 @@ async def apply_derive_groups(project_id: int) -> int:
                     CodeShape.status.in_(_MECHANICAL_TODO),
                     CodeShape.vanished_at.is_(None),
                     CodeShape.proposed_snippet_id.is_(None),
+                    # A file row is named by its stem, and `index`, `+page`
+                    # and `__init__` repeat by convention, not by copying —
+                    # a derive family of them would be pure noise.
+                    CodeShape.kind != "file",
                 )
             )
         ).scalars().all()

tests/test_integration_shape_classify.py

@@ -1231,3 +1231,31 @@ async def test_unjudged_is_what_nobody_read(seeded):
     assert got[1]["evidence"]["stamped"] == {"snippet_id": sid, "score": 0.7}
     assert got[2]["evidence"] == {}
     assert await unjudged_shapes(pid, []) == []
+
+
+@pytest.mark.integration
+async def test_a_whole_file_snippet_makes_its_file_row_canonical(seeded):
+    """Milestone 439: a snippet recorded at a component's path, with no
+    symbol, is the canon for that file row — the claim a whole-file record
+    makes and a definition row never could. It still covers no definition
+    inside the file."""
+    from scribe.services import snippets as snippets_svc
+    from scribe.services.coverage import _recorded_locations
+    from scribe.services.shape_ledger import mark_canonicals
+
+    owner, pid = seeded["owner"], seeded["pid"]
+    await sync_repo_shapes(pid, REPO, SHAPES + [
+        ("web/lib/StatusChip.svelte", "file", "StatusChip"),
+        ("web/lib/StatusChip.svelte", "css", "chip"),
+    ], seen_marker="c1")
+    chip = await snippets_svc.create_snippet(
+        owner, name="StatusChip — the one status pill",
+        code="<span class=\"chip\">{label}</span>", language="svelte",
+        repo="Widget", path="web/lib/StatusChip.svelte", project_id=pid,
+    )
+    await mark_canonicals(pid, await _recorded_locations(owner, pid))
+    rows, _ = await list_project_shapes(owner, pid, path="web/lib/StatusChip.svelte")
+    by = {(r.kind, r.symbol): r for r in rows}
+    assert by[("file", "StatusChip")].status == "canonical"
+    assert by[("file", "StatusChip")].snippet_id == int(chip.id)
+    assert by[("css", "chip")].status == "unclassified"

tests/test_pattern_coverage.py

@@ -310,6 +310,7 @@ def test_scan_archive_returns_definitions_and_references_from_one_walk():
     scan = scan_archive(_tarball(tree))
     assert isinstance(scan, ArchiveScan)
     assert [(d.path, d.kind, d.name) for d in scan.definitions] == TREE_SHAPES + [
+        ("web/Card.vue", "file", "Card"),     # a component is a unit (milestone 439)
         ("web/Card.vue", "css", "card"),
     ]
     # Only files whose markup names a class appear; the .py/.css files don't.
@@ -886,3 +887,23 @@ async def test_binding_ref_is_the_branch_the_ledger_follows(seeded):
     coverage = await compute_coverage(uid, pid, selector=_selector(_tarball(TREE)))
     assert coverage["repos"][0]["ref"] == "main"
 
+
+
+@pytest.mark.parametrize("path,text,unit", [
+    # A single-file component: nothing inside carries its name.
+    ("web/lib/StatusChip.svelte", "<script>\nfunction tone() {\n  return 1;\n}\n</script>\n<span class=\"chip\"></span>\n", True),
+    ("web/views/Card.vue", "<template><div class=\"card\"/></template>\n", True),
+    # A TSX component already has its row: the function named after the file.
+    ("web/Card.tsx", "export function Card() {\n  return <div className=\"card\" />;\n}\n", False),
+    # A JSX file whose markup names classes but whose component is named otherwise.
+    ("web/index.jsx", "function App() {\n  return <div className=\"app\" />;\n}\n", True),
+    # Modules render nothing: their definitions account for them.
+    ("internal/api/upnext.go", "func stageOf(x int) int {\n\treturn x\n}\n", False),
+    ("src/app/util.py", "def helper():\n    pass\n", False),
+    ("web/button.css", ".btn {\n  color: red;\n}\n", False),
+])
+def test_a_file_is_a_unit_when_it_renders_and_nothing_inside_carries_its_name(path, text, unit):
+    """Milestone 439: one structural test, not a framework list."""
+    from scribe.services.coverage import class_references, extract_definitions, is_file_unit
+
+    assert is_file_unit(path, text, extract_definitions(text), class_references(path, text)) is unit