fix(skillgen): pass root= to extract() in Step 3 Part A (#3727)

extract() uses cache_root as the fallback anchor for source_file and node ids
when root is not given (#1941). Part A called it with cache_root alone. That is
harmless while the cache sits in the scan root, but the two parameters look
alike and the block gets adapted: pointing only the cache somewhere else
re-anchored every code node. Same node count, every id renamed, every cached
edge into code left dangling - the node-count shrink guard cannot see it.

Pass root=Path('INPUT_PATH') next to cache_root, as the CLI already does, and
say in the block which of the two anchors ids. The rendered call is behaviour
preserving as written; it only stops being a trap when the cache is relocated.

The aider/devin monoliths call extract(code_files) with neither argument and
are left untouched.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

graphify/skill-agents.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-amp.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-claw.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-codex.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-copilot.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-droid.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-kilo.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-kiro.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-opencode.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-pi.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-trae.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-vscode.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

graphify/skill-windows.md

@@ -218,7 +218,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding="utf-8")
     print(f'AST: {len(result["nodes"])} nodes, {len(result["edges"])} edges')
 else:

graphify/skill.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tests/test_skill_part_a_explicit_root.py

@@ -0,0 +1,78 @@
+"""Step 3 Part A passes root= to extract(), so the cache location never anchors ids (#3727).
+
+``extract(paths, cache_root=...)`` uses ``cache_root`` as the fallback anchor for
+``source_file`` and node ids when ``root`` is not given (#1941). Part A called it
+with ``cache_root`` alone. That is harmless while the cache sits in the scan root,
+but the two parameters look alike, and anyone adapting the block to a cache that
+lives elsewhere re-anchored every code node: same node count, every id renamed,
+and every cached edge into code left dangling. A node-count guard cannot see it.
+"""
+from __future__ import annotations
+
+import sys
+from pathlib import Path
+
+import pytest
+
+# tests/ -> repo root is one parent up; put it on the path so tools.skillgen
+# imports regardless of pytest's import mode.
+REPO_ROOT = Path(__file__).resolve().parent.parent
+if str(REPO_ROOT) not in sys.path:
+    sys.path.insert(0, str(REPO_ROOT))
+
+from graphify.extract import extract  # noqa: E402
+from tools.skillgen import gen  # noqa: E402
+
+_CALL = "extract(code_files, cache_root="
+
+
+def _split_hosts() -> list[gen.Platform]:
+    return [p for p in gen.load_platforms().values() if p.bucket != "monolith"]
+
+
+def _part_a_calls(platform: gen.Platform) -> list[str]:
+    body = gen.render(platform)[0].content
+    return [line.strip() for line in body.splitlines() if _CALL in line]
+
+
+def test_split_hosts_are_discovered():
+    """The parametrized guard below must not pass vacuously on an empty list."""
+    assert len(_split_hosts()) >= 10, [p.key for p in _split_hosts()]
+
+
+@pytest.mark.parametrize("platform", _split_hosts(), ids=lambda p: p.key)
+def test_part_a_passes_an_explicit_root(platform: gen.Platform):
+    calls = _part_a_calls(platform)
+    assert calls, f"[{platform.key}] Part A extract() call not found"
+    for call in calls:
+        assert "root=Path('INPUT_PATH')" in call.replace("cache_root=Path('INPUT_PATH')", ""), (
+            f"[{platform.key}] Part A calls extract() with cache_root but no root= (#3727): {call!r}"
+        )
+
+
+def test_relocating_only_the_cache_does_not_rename_part_a_nodes(tmp_path: Path):
+    """Run the shipped call twice, moving nothing but the cache: ids must not change."""
+    scan_root = tmp_path / "proj"
+    (scan_root / "pkg").mkdir(parents=True)
+    module = scan_root / "pkg" / "mod.py"
+    module.write_text("def hello():\n    return 1\n", encoding="utf-8")
+    elsewhere = tmp_path / "cache-elsewhere"
+    elsewhere.mkdir()
+
+    (call,) = _part_a_calls(gen.load_platforms()["claude"])
+    expression = call.removeprefix("result = ")
+    in_place = expression.replace("Path('INPUT_PATH')", f"Path({str(scan_root)!r})")
+    relocated = expression.replace(
+        "cache_root=Path('INPUT_PATH')", f"cache_root=Path({str(elsewhere)!r})"
+    ).replace("Path('INPUT_PATH')", f"Path({str(scan_root)!r})")
+    assert in_place != relocated
+
+    scope = {"extract": extract, "Path": Path, "code_files": [module]}
+    ids_in_place = sorted(n["id"] for n in eval(in_place, dict(scope))["nodes"])
+    ids_relocated = sorted(n["id"] for n in eval(relocated, dict(scope))["nodes"])
+
+    assert ids_in_place, "expected AST nodes for the module"
+    assert ids_relocated == ids_in_place, (
+        "moving only the cache renamed the code nodes: "
+        f"{ids_in_place} -> {ids_relocated}"
+    )

tools/skillgen/expected/graphify__skill-agents.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-amp.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-claw.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-codex.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-copilot.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-droid.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-kilo.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-kiro.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-opencode.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-pi.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-trae.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-vscode.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/expected/graphify__skill-windows.md

@@ -218,7 +218,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding="utf-8")
     print(f'AST: {len(result["nodes"])} nodes, {len(result["edges"])} edges')
 else:

tools/skillgen/expected/graphify__skill.md

@@ -190,7 +190,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else:

tools/skillgen/fragments/core/core.md

@@ -143,7 +143,10 @@ for f in detect.get('files', {}).get('code', []):
     code_files.extend(collect_files(Path(f)) if Path(f).is_dir() else [Path(f)])
 
 if code_files:
-    result = extract(code_files, cache_root=Path('INPUT_PATH'))
+    # root= anchors node ids and source_file on the scan root; cache_root only says
+    # where graphify-out/cache lives. Pass both: without root= the anchor falls back
+    # to cache_root, so pointing the cache elsewhere would rename every node (#3727).
+    result = extract(code_files, cache_root=Path('INPUT_PATH'), root=Path('INPUT_PATH'))
     Path('graphify-out/.graphify_ast.json').write_text(json.dumps(result, indent=2, ensure_ascii=False), encoding=\"utf-8\")
     print(f'AST: {len(result[\"nodes\"])} nodes, {len(result[\"edges\"])} edges')
 else: