fix: four bugs — affected direction, hook root, glob fish/zsh, manifest drift (#1174 #1173 #1172 #1163)

#1174: affected.py load_graph now forces directed=True before
node_link_graph, matching the identical fix in serve.py and __main__.py.
Undirected graphs (directed:false in graph.json) were causing in_edges
to fall back to a direction-blind scan, missing true callers and
reporting false positives. Regression test added.

#1173: post-commit and post-checkout hook bodies now read
graphify-out/.graphify_root before calling _rebuild_code, falling back
to Path('.') if absent. A scoped build (graphify src/) no longer gets
silently expanded to the full repo on the next commit. Tests added.

#1172: Step 9 cleanup split into rm -f for fixed files and
find -maxdepth 1 -delete for the chunk glob. Under fish/zsh an
unmatched glob aborts the entire rm -f line, leaving temp files on disk.
Fixed in the three skillgen source fragments and regenerated.

#1163: detect_incremental type guard on stored mtime — if the manifest
contains a dict-valued mtime (schema drift from older versions), coerce
to None rather than propagating a non-numeric into comparisons.
Regression test added.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

graphify/affected.py

@@ -145,6 +145,9 @@ def load_graph(path: Path) -> nx.Graph:
     from networkx.readwrite import json_graph
 
     raw = json.loads(path.read_text(encoding="utf-8"))
+    # Force directed so stored caller→callee direction survives the round-trip;
+    # mirrors serve.py and __main__.py (#1174).
+    raw = {**raw, "directed": True}
     try:
         return json_graph.node_link_graph(raw, edges="links")
     except TypeError:

graphify/detect.py

@@ -1347,6 +1347,12 @@ def detect_incremental(
                     changed = True
                 else:
                     stored_mtime = stored.get("mtime")
+                    # Schema-drift guard (#1163): tolerate a nested {mtime: ...}
+                    # dict or any non-numeric value without crashing.
+                    if isinstance(stored_mtime, dict):
+                        stored_mtime = stored_mtime.get("mtime")
+                    if not isinstance(stored_mtime, (int, float)):
+                        stored_mtime = None
                     if stored_mtime is None or current_mtime != stored_mtime:
                         # mtime bumped — verify with content hash before re-extracting
                         changed = _md5_file(Path(f)) != stored_hash

graphify/hooks.py

@@ -98,7 +98,13 @@ try:
         signal.signal(signal.SIGALRM, lambda *_: (_ for _ in ()).throw(TimeoutError(f'graphify rebuild exceeded {_timeout}s')))
         signal.alarm(_timeout)
     _force = os.environ.get('GRAPHIFY_FORCE', '').lower() in ('1', 'true', 'yes')
-    _rebuild_code(Path('.'), changed_paths=changed, force=_force)
+    _root = Path('.')
+    _saved = Path('graphify-out/.graphify_root')
+    if _saved.exists():
+        _txt = _saved.read_text(encoding='utf-8').strip()
+        if _txt:
+            _root = Path(_txt)
+    _rebuild_code(_root, changed_paths=changed, force=_force)
 except TimeoutError as exc:
     print(f'[graphify hook] {exc}')
     sys.exit(1)
@@ -121,7 +127,13 @@ try:
     # post-checkout: branch switch can touch arbitrary files; full rebuild path
     # (no changed_paths) is correct here. The flock inside _rebuild_code still
     # prevents pile-ups when commit + checkout fire back-to-back.
-    _rebuild_code(Path('.'), force=_force)
+    _root = Path('.')
+    _saved = Path('graphify-out/.graphify_root')
+    if _saved.exists():
+        _txt = _saved.read_text(encoding='utf-8').strip()
+        if _txt:
+            _root = Path(_txt)
+    _rebuild_code(_root, force=_force)
 except TimeoutError as exc:
     print(f'[graphify] {exc}')
     sys.exit(1)

graphify/skill-aider.md

@@ -684,7 +684,7 @@ cost_path.write_text(json.dumps(cost, indent=2))
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f .graphify_detect.json .graphify_extract.json .graphify_ast.json .graphify_semantic.json .graphify_analysis.json .graphify_labels.json .graphify_chunk_*.json
+rm -f .graphify_detect.json .graphify_extract.json .graphify_ast.json .graphify_semantic.json .graphify_analysis.json .graphify_labels.json; find . -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-amp.md

@@ -520,7 +520,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-claw.md

@@ -523,7 +523,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-codex.md

@@ -520,7 +520,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-copilot.md

@@ -523,7 +523,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-devin.md

@@ -801,7 +801,7 @@ cost_path.write_text(json.dumps(cost, indent=2))
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_labels.json graphify-out/.graphify_chunk_*.json graphify-out/.graphify_incremental.json graphify-out/.graphify_transcripts.json graphify-out/.graphify_old.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_labels.json graphify-out/.graphify_incremental.json graphify-out/.graphify_transcripts.json graphify-out/.graphify_old.json; find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-droid.md

@@ -520,7 +520,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-kilo.md

@@ -523,7 +523,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-kiro.md

@@ -522,7 +522,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-opencode.md

@@ -515,7 +515,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-pi.md

@@ -522,7 +522,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-trae.md

@@ -521,7 +521,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-vscode.md

@@ -519,7 +519,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill-windows.md

@@ -545,7 +545,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

graphify/skill.md

@@ -523,7 +523,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tests/test_affected_cli.py

@@ -58,3 +58,37 @@ def test_affected_cli_relation_filter_limits_reverse_traversal(monkeypatch, tmp_
     assert "Relations: calls" in out
     assert "X()" in out
     assert "__init__.py" not in out
+
+
+def test_affected_cli_forces_directed_on_undirected_graph(monkeypatch, tmp_path, capsys):
+    """A graph persisted with directed=false must still recover caller->callee
+    direction (#1174): affected on the callee returns the caller, not the callee
+    or nothing. Without forcing directed=True, node_link_graph builds an
+    undirected Graph, predecessors() collapses, and the reverse traversal breaks.
+    """
+    graph = nx.DiGraph()
+    graph.add_node("A", label="caller_fn", source_file="a.py", source_location="L1")
+    graph.add_node("B", label="callee_fn", source_file="b.py", source_location="L2")
+    graph.add_edge("A", "B", relation="calls", context="call", confidence="EXTRACTED")
+
+    data = json_graph.node_link_data(graph, edges="links")
+    # Persist as undirected on disk to reproduce the bug condition.
+    data["directed"] = False
+    graph_path = tmp_path / "graph.json"
+    graph_path.write_text(json.dumps(data), encoding="utf-8")
+
+    monkeypatch.setattr(mainmod, "_check_skill_version", lambda _: None)
+    monkeypatch.setattr(
+        mainmod.sys,
+        "argv",
+        ["graphify", "affected", "B", "--relation", "calls", "--graph", str(graph_path)],
+    )
+
+    mainmod.main()
+
+    out = capsys.readouterr().out
+    # A (the caller) is affected by a change to B (the callee).
+    assert "caller_fn" in out
+    assert "calls" in out
+    # B is the query node, not an affected node, and the result is not empty.
+    assert "No affected nodes found." not in out

tests/test_detect.py

@@ -298,6 +298,43 @@ def test_detect_incremental_propagates_follow_symlinks(tmp_path, monkeypatch):
     assert second["new_total"] == 0
 
 
+def test_detect_incremental_survives_dict_valued_mtime(tmp_path, monkeypatch):
+    """A schema-drifted manifest whose entry stores mtime as a nested dict
+    (instead of a float) must not crash detect_incremental (#1163). The guard
+    coerces the bad mtime to None so the file is re-verified by content hash and
+    treated as new, rather than blowing up on the int/float comparison.
+    """
+    import json
+
+    monkeypatch.chdir(tmp_path)
+
+    src = tmp_path / "mod.py"
+    src.write_text("def f():\n    return 1\n", encoding="utf-8")
+
+    manifest_dir = tmp_path / "graphify-out"
+    manifest_dir.mkdir()
+    manifest_path = str(manifest_dir / "manifest.json")
+
+    # Drifted entry: a non-empty ast_hash (so the dict branch reaches the mtime
+    # comparison) with mtime stored as a dict rather than a float. Absolute key
+    # so it matches detect's absolute file paths without re-anchoring.
+    drifted = {
+        str(src.resolve()): {
+            "mtime": {"mtime": 123.0},
+            "ast_hash": "deadbeef" * 4,
+            "semantic_hash": "cafebabe" * 4,
+        }
+    }
+    Path(manifest_path).write_text(json.dumps(drifted), encoding="utf-8")
+
+    # Must not raise (pre-fix: TypeError comparing float and dict).
+    result = detect_incremental(tmp_path, manifest_path)
+
+    # The drifted file is re-classified as new rather than silently skipped.
+    assert any("mod.py" in f for f in result["new_files"]["code"])
+    assert not any("mod.py" in f for f in result["unchanged_files"]["code"])
+
+
 def test_classify_video_extensions():
     """Video and audio file extensions should classify as VIDEO."""
     from graphify.detect import FileType

tests/test_hooks.py

@@ -295,6 +295,28 @@ def test_rebuild_bodies_are_shell_quote_safe():
         assert "'''" not in body  # would terminate the launcher's _src literal
 
 
+@pytest.mark.parametrize(
+    "name,body",
+    [("post-commit", _REBUILD_BODY_COMMIT), ("post-checkout", _REBUILD_BODY_CHECKOUT)],
+)
+def test_rebuild_bodies_read_graphify_root(name, body):
+    """The rebuild must honour the persisted scan root rather than hardcoding the
+    repo top (#1173). Both bodies read graphify-out/.graphify_root and pass the
+    recovered root to _rebuild_code instead of the bare Path('.')."""
+    assert "graphify-out/.graphify_root" in body, f"{name} ignores .graphify_root (#1173)"
+    # The recovered root is what gets rebuilt, not a hardcoded cwd.
+    assert "_rebuild_code(_root" in body, f"{name} does not pass the recovered root"
+    # Quote-safe inside the shell-double-quoted launcher: single quotes only.
+    assert "read_text(encoding='utf-8')" in body, f"{name} root read is not single-quoted"
+
+
+def test_rebuild_bodies_with_graphify_root_are_valid_python():
+    """The .graphify_root snippet must parse so a quoting slip can't ship a hook
+    that crashes the moment git fires it (#1173)."""
+    for body in (_REBUILD_BODY_COMMIT, _REBUILD_BODY_CHECKOUT):
+        ast.parse(body)
+
+
 def test_detached_launch_targets_graphify_python():
     """The launcher must run via the resolved $GRAPHIFY_PYTHON, not a bare
     `python`, so it uses the same interpreter the detection block selected."""

tests/test_skillgen.py

@@ -455,12 +455,13 @@ def test_monolith_roundtrip_passes_for_aider_and_devin():
         assert problems == [], f"[{key}]\n" + "\n".join(problems)
 
 
-def test_monoliths_change_only_the_enum_and_the_description():
-    """The rendered monolith differs from v8 on exactly the enum + description lines.
+def test_monoliths_change_only_the_enum_description_and_chunk_cleanup():
+    """The rendered monolith differs from v8 on exactly the allowed lines.
 
-    Two changes are now in play for the monoliths: the file_type enum unified to
-    the six-value superset (the prose guidance line + the schema line) and the
-    frontmatter description unified across all platforms. Nothing else may differ.
+    Three changes are now in play for the monoliths: the file_type enum unified to
+    the six-value superset (the prose guidance line + the schema line), the
+    frontmatter description unified across all platforms, and the shell-agnostic
+    chunk-cleanup rewrite (#1172). Nothing else may differ.
     """
     platforms = gen.load_platforms()
     for key in ("aider", "devin"):
@@ -468,11 +469,12 @@ def test_monoliths_change_only_the_enum_and_the_description():
         original = gen._normalise(gen._git_show(platforms[key].roundtrip_ref)).splitlines()
         assert len(rendered) == len(original), f"[{key}] line count changed"
         diff_idx = [i for i, (r, o) in enumerate(zip(rendered, original)) if r != o]
-        # Exactly three lines change: the prose enum guidance, the schema line,
-        # and the frontmatter description.
-        assert len(diff_idx) == 3, f"[{key}] expected 3 changed lines, got {len(diff_idx)}"
+        # Exactly four lines change: the prose enum guidance, the schema line,
+        # the frontmatter description, and the chunk-cleanup rewrite.
+        assert len(diff_idx) == 4, f"[{key}] expected 4 changed lines, got {len(diff_idx)}"
         enum_changes = 0
         desc_changes = 0
+        cleanup_changes = 0
         for i in diff_idx:
             line = rendered[i]
             if gen.ENUM_VALUES in line or gen.ENUM_PROSE in line:
@@ -482,12 +484,18 @@ def test_monoliths_change_only_the_enum_and_the_description():
                 assert UNIFIED_DESCRIPTION in line, (
                     f"[{key}] description line is not the unified text: {line!r}"
                 )
+            elif gen._is_chunk_cleanup_line(line):
+                cleanup_changes += 1
+                # The unmatched-glob abort is fixed: the rm no longer carries the
+                # bare chunk glob, and a find ... -delete sweeps the chunks.
+                assert ".graphify_chunk_*.json" not in line.split("find", 1)[0]
             else:
                 raise AssertionError(
-                    f"[{key}] changed line {i} is neither enum nor description: {line!r}"
+                    f"[{key}] changed line {i} is none of enum/description/cleanup: {line!r}"
                 )
         assert enum_changes == 2, f"[{key}] expected 2 enum line changes, got {enum_changes}"
         assert desc_changes == 1, f"[{key}] expected 1 description change, got {desc_changes}"
+        assert cleanup_changes == 1, f"[{key}] expected 1 cleanup change, got {cleanup_changes}"
         # The six-value superset replaced the five-value enum in both files.
         assert any(gen.ENUM_VALUES in line for line in rendered)
 

tools/skillgen/expected/graphify__skill-aider.md

@@ -684,7 +684,7 @@ cost_path.write_text(json.dumps(cost, indent=2))
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f .graphify_detect.json .graphify_extract.json .graphify_ast.json .graphify_semantic.json .graphify_analysis.json .graphify_labels.json .graphify_chunk_*.json
+rm -f .graphify_detect.json .graphify_extract.json .graphify_ast.json .graphify_semantic.json .graphify_analysis.json .graphify_labels.json; find . -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-amp.md

@@ -520,7 +520,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-claw.md

@@ -523,7 +523,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-codex.md

@@ -520,7 +520,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-copilot.md

@@ -523,7 +523,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-devin.md

@@ -801,7 +801,7 @@ cost_path.write_text(json.dumps(cost, indent=2))
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_labels.json graphify-out/.graphify_chunk_*.json graphify-out/.graphify_incremental.json graphify-out/.graphify_transcripts.json graphify-out/.graphify_old.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_labels.json graphify-out/.graphify_incremental.json graphify-out/.graphify_transcripts.json graphify-out/.graphify_old.json; find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-droid.md

@@ -520,7 +520,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-kilo.md

@@ -523,7 +523,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-kiro.md

@@ -522,7 +522,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-opencode.md

@@ -515,7 +515,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-pi.md

@@ -522,7 +522,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-trae.md

@@ -521,7 +521,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-vscode.md

@@ -519,7 +519,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill-windows.md

@@ -545,7 +545,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/expected/graphify__skill.md

@@ -523,7 +523,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/fragments/core/aider.md

@@ -684,7 +684,7 @@ cost_path.write_text(json.dumps(cost, indent=2))
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f .graphify_detect.json .graphify_extract.json .graphify_ast.json .graphify_semantic.json .graphify_analysis.json .graphify_labels.json .graphify_chunk_*.json
+rm -f .graphify_detect.json .graphify_extract.json .graphify_ast.json .graphify_semantic.json .graphify_analysis.json .graphify_labels.json; find . -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/fragments/core/core.md

@@ -457,7 +457,8 @@ cost_path.write_text(json.dumps(cost, indent=2, ensure_ascii=False), encoding=\"
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_chunk_*.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json
+find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/fragments/core/devin.md

@@ -801,7 +801,7 @@ cost_path.write_text(json.dumps(cost, indent=2))
 print(f'This run: {input_tok:,} input tokens, {output_tok:,} output tokens')
 print(f'All time: {cost[\"total_input_tokens\"]:,} input, {cost[\"total_output_tokens\"]:,} output ({len(cost[\"runs\"])} runs)')
 "
-rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_labels.json graphify-out/.graphify_chunk_*.json graphify-out/.graphify_incremental.json graphify-out/.graphify_transcripts.json graphify-out/.graphify_old.json
+rm -f graphify-out/.graphify_detect.json graphify-out/.graphify_extract.json graphify-out/.graphify_ast.json graphify-out/.graphify_semantic.json graphify-out/.graphify_analysis.json graphify-out/.graphify_labels.json graphify-out/.graphify_incremental.json graphify-out/.graphify_transcripts.json graphify-out/.graphify_old.json; find graphify-out -maxdepth 1 -name '.graphify_chunk_*.json' -delete 2>/dev/null
 rm -f graphify-out/.needs_update 2>/dev/null || true
 ```
 

tools/skillgen/gen.py

@@ -705,6 +705,19 @@ def _is_frontmatter_description_line(line: str) -> bool:
     return line.lstrip().startswith("description:")
 
 
+def _is_chunk_cleanup_line(line: str) -> bool:
+    """Whether a line is the Step 9 chunk-file cleanup ``rm -f`` command.
+
+    The bare glob ``.graphify_chunk_*.json`` in the v8 cleanup line aborts the
+    whole ``rm`` under fish/zsh when no chunk files exist (no-match is a hard
+    error there, unlike bash). The fix (graphify #1172) drops the glob from the
+    ``rm`` and deletes the chunk files with ``find ... -delete`` instead. That
+    rewrite touches the single cleanup line in place (no line added or removed),
+    so it joins the enum and description unifications as an allowed monolith diff.
+    """
+    return line.lstrip().startswith("rm -f") and "find " in line and "-name '.graphify_chunk_" in line
+
+
 def monolith_roundtrip(platform: Platform) -> list[str]:
     """Assert a monolith renders diff-clean vs its v8 blob modulo allowed changes.
 
@@ -736,8 +749,9 @@ def monolith_roundtrip(platform: Platform) -> list[str]:
     for i, (r, o) in enumerate(zip(rendered_lines, original_lines), start=1):
         if r == o:
             continue
-        # The permitted diffs are the enum unification and the unified description.
-        if _is_enum_line(r) or _is_frontmatter_description_line(r):
+        # The permitted diffs are the enum unification, the unified description,
+        # and the shell-agnostic chunk-cleanup rewrite (#1172).
+        if _is_enum_line(r) or _is_frontmatter_description_line(r) or _is_chunk_cleanup_line(r):
             continue
         problems.append(
             f"[{platform.key}] line {i} differs and is not an enum or description unification:\n"