Drop internal plan references from code comments

The pack cited internal planning-doc items (plan P3-7, decision D1, plan
risk 9.2) that aren't public, leaving outside contributors with pointers
they can't follow. Each comment already states its reason; only the
dangling citation is removed.

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

tests/test_api.py

@@ -160,7 +160,7 @@ def test_401_flags_needs_login(monkeypatch):
 
 def test_401_with_env_token_hints_at_expiry(monkeypatch):
     """
-    Plan P0-5: the env token never refreshes itself, so a headless install's
+    The env token never refreshes itself, so a headless install's
     401 hours later needs to say the likely cause — an expired STYLEREF_TOKEN.
     """
     monkeypatch.setenv("STYLEREF_TOKEN", "stale-token")
@@ -213,7 +213,7 @@ def test_get_style_spec_reads_the_canonical_url_header(monkeypatch):
     assert url == "https://styleref.io/share/real-slug"
 
 
-# ── extraction stays out of this client (plan P0-6 / decision D1) ────────────
+# ── extraction stays out of this client ──────────────────────────────────────
 
 
 def test_client_has_no_extraction_endpoints():
@@ -306,7 +306,7 @@ def test_client_tag_header_is_sent(monkeypatch):
     assert captured.get("X-styleref-client") == "comfyui"
 
 
-# ── save (P5-6) ──────────────────────────────────────────────────────────────
+# ── save ─────────────────────────────────────────────────────────────────────
 
 
 def test_save_style_posts_to_the_save_endpoint(monkeypatch):
@@ -322,7 +322,7 @@ def test_save_style_posts_to_the_save_endpoint(monkeypatch):
     assert seen["url"].endswith("/styles/%40ada%2FWarm%20Editorial/save")
 
 
-# ── conditional GET (P5-8) ───────────────────────────────────────────────────
+# ── conditional GET ──────────────────────────────────────────────────────────
 
 
 def test_304_is_success_not_error(monkeypatch):

tests/test_auth.py

@@ -138,7 +138,7 @@ def test_dead_session_degrades_to_anonymous(tmp_path, monkeypatch):
     assert auth.bearer_token() is None
 
 
-# ── headless detection (plan risk 9.2) ───────────────────────────────────────
+# ── headless detection ───────────────────────────────────────────────────────
 
 
 def test_ssh_session_is_headless():
@@ -174,7 +174,7 @@ def test_headless_help_leads_with_the_credentials_file_then_the_env_var():
     assert help_text.index("credentials.json") < help_text.index("STYLEREF_TOKEN")
 
 
-# ── OAuth callback page (plan P0-4) ──────────────────────────────────────────
+# ── OAuth callback page ──────────────────────────────────────────────────────
 
 
 class _FakeWFile:

tests/test_nodes.py

@@ -36,12 +36,12 @@ def clear_cache():
     load_node._GENERATION.clear()
 
 
-# ── pack registration (plan P0-6) ────────────────────────────────────────────
+# ── pack registration ────────────────────────────────────────────────────────
 
 
 def test_pack_registers_exactly_the_five_nodes():
     """
-    Extraction moved to the web app (decision D1): the pack registers
+    Extraction moved to the web app: the pack registers
     Load/Apply/Facets/ReferenceImages/Login and nothing else — an Extract node
     here means the removal regressed.
     """
@@ -65,7 +65,7 @@ def test_pack_registers_exactly_the_five_nodes():
 
 def test_targets_are_exactly_the_web_apps_copy_list_in_order():
     """
-    Owner decision 2026-07-22 (supersedes D2): the target list mirrors the web
+    Owner decision 2026-07-22: the target list mirrors the web
     app's copy-box format list exactly — same entries, same order — so the
     style means the same thing here as on styleref.io. Source of truth:
     PROMPT_FORMAT_OPTIONS in src/components/ui/PromptFormatDropdown.tsx.
@@ -246,7 +246,7 @@ def test_apply_rejects_an_unloaded_style():
         node.apply(make_style(ref="", name="(none)"), "subject", "flux")
 
 
-# ── Apply: node-body feedback (plan P0-2 / P1-8) ─────────────────────────────
+# ── Apply: node-body feedback ────────────────────────────────────────────────
 
 
 def test_apply_returns_ui_text_and_unchanged_result_tuple(monkeypatch):
@@ -370,7 +370,7 @@ def test_private_refs_are_tagged_as_library(monkeypatch):
 
 def test_ungenerated_style_error_reaches_the_user_verbatim(monkeypatch):
     """
-    Plan P0-7: when the server refuses a never-generated own-style ref (409),
+    When the server refuses a never-generated own-style ref (409),
     its message — which tells the user to Generate on styleref.io — must pass
     through unreworded.
     """
@@ -390,7 +390,7 @@ def test_ungenerated_style_error_reaches_the_user_verbatim(monkeypatch):
 
 
 def test_load_returns_ui_text_and_result_tuple(monkeypatch):
-    """P0-2: the load confirmation renders in the node body."""
+    """The load confirmation renders in the node body."""
     monkeypatch.setattr(
         api, "get_style_spec", lambda ref, etag=None: ({"name": "Warm", "sections": {}}, None, None)
     )
@@ -422,7 +422,7 @@ def test_load_never_fetches_style_text(monkeypatch):
 
 
 def test_load_is_an_output_node():
-    """A search-only Load must be queueable alone and still display (P0-2)."""
+    """A search-only Load must be queueable alone and still display."""
     assert load_node.StyleRefLoad.OUTPUT_NODE is True
 
 
@@ -456,7 +456,7 @@ def test_search_failure_does_not_sink_the_run(monkeypatch):
 
 
 def test_search_category_is_forwarded(monkeypatch):
-    """The vanilla category combo (P3-4) must reach the API; `any` must not."""
+    """The vanilla category combo must reach the API; `any` must not."""
     seen = {}
 
     def fake_search(query, category=None, sort=None, limit=12):
@@ -474,7 +474,7 @@ def test_search_category_is_forwarded(monkeypatch):
     assert seen["category"] is None
 
 
-# ── IS_CHANGED contracts (plan P0-1's regression class) ──────────────────────
+# ── IS_CHANGED contracts ─────────────────────────────────────────────────────
 
 
 def test_load_is_changed_is_stable_when_cached():
@@ -497,13 +497,13 @@ def test_login_is_always_dirty():
 def test_apply_has_no_is_changed_override():
     """
     Apply's dirtiness must come from its inputs alone. An IS_CHANGED override
-    returning NaN here is the credit-burn/api-hammer class of bug (plan P0-1) —
+    returning NaN here is the credit-burn/api-hammer class of bug —
     the node would re-run on every queue.
     """
     assert "IS_CHANGED" not in vars(apply_node.StyleRefApply)
 
 
-# ── Apply: sections filter (P5-3) ────────────────────────────────────────────
+# ── Apply: sections filter ───────────────────────────────────────────────────
 
 
 def test_sections_filter_reaches_the_api_and_the_cache_key(monkeypatch):
@@ -527,7 +527,7 @@ def test_sections_filter_reaches_the_api_and_the_cache_key(monkeypatch):
     assert calls[-1] is None
 
 
-# ── Facets: full schema coverage (P5-2 / D4) ─────────────────────────────────
+# ── Facets: full schema coverage ─────────────────────────────────────────────
 
 
 def test_facets_outputs_are_exactly_the_schema_sections_plus_custom_items():
@@ -636,7 +636,7 @@ def test_reference_images_errors_clearly_without_images():
         ri.StyleRefReferenceImages().fetch(make_style(ref="x", name="Bare"))
 
 
-# ── Load: save_to_library (P5-6) ─────────────────────────────────────────────
+# ── Load: save_to_library ────────────────────────────────────────────────────
 
 
 def test_save_to_library_saves_once_per_session(monkeypatch):
@@ -690,7 +690,7 @@ def test_own_styles_are_not_re_saved(monkeypatch):
     assert "Already your own style" in out["ui"]["text"][0]
 
 
-# ── Load: ETag revalidation (P5-8) ───────────────────────────────────────────
+# ── Load: ETag revalidation ──────────────────────────────────────────────────
 
 
 def test_cache_off_revalidates_with_the_stored_etag(monkeypatch):

tests/test_style.py

@@ -128,7 +128,7 @@ def test_negatives_are_normalized_comma_terms(style):
 
 def test_negatives_cover_artifacts_and_avoided_colors():
     """
-    Plan P1-9: guardrails carry more than avoided_visuals — artifacts to avoid
+    Guardrails carry more than avoided_visuals — artifacts to avoid
     and avoided colors are user-authored constraints and belong in the negative.
     """
     style = make_style(
@@ -241,7 +241,7 @@ def test_is_style_rejects_arbitrary_dicts():
     assert is_style(make_style(ref="x", name="ok")) is True
 
 
-# ── dimensions (P5-2: width/height from the style) ───────────────────────────
+# ── dimensions (width/height from the style) ─────────────────────────────────
 
 
 def _fmt_style(aspect: str | None = None, resolution: str | None = None):

tests/test_workflows.py

@@ -4,8 +4,8 @@ Workflow template tests.
 A broken template is only discovered when a user drags it onto a canvas and it
 fails silently, so the structural invariants are checked here instead: every
 link points at a node that exists, at a slot that exists, carrying a type both
-ends agree on — and the layout invariants of plan P2-1/P2-2: no overlapping
-nodes, instructions in visible Note nodes, titles short enough to render.
+ends agree on — plus the layout invariants: no overlapping nodes, instructions
+in visible Note nodes, titles short enough to render.
 """
 
 from __future__ import annotations
@@ -36,7 +36,7 @@ def _load(path: str) -> dict:
 
 
 def _note_text(graph: dict) -> str:
-    """All visible Note-node text — the instruction channel (plan P2-2)."""
+    """All visible Note-node text — the instruction channel."""
     return "\n".join(
         str(n["widgets_values"][0])
         for n in graph["nodes"]
@@ -50,7 +50,7 @@ def test_every_declared_workflow_is_committed():
 
 
 def test_extraction_template_is_gone():
-    """Plan P0-6: extraction lives on the web; template 03 tells that story."""
+    """Extraction lives on the web; template 03 tells that story."""
     assert "03-extract-and-reuse.json" not in {os.path.basename(p) for p in FILES}
     assert "03-your-own-style.json" in {os.path.basename(p) for p in FILES}
 
@@ -78,7 +78,7 @@ def test_committed_file_matches_the_generator(path):
 
 @pytest.mark.parametrize("path", FILES, ids=[_relpath(p) for p in FILES])
 def test_only_the_shipped_styleref_nodes_appear(path):
-    """Plan P0-6: a template referencing a removed node fails on the canvas."""
+    """A template referencing a removed node fails on the canvas."""
     allowed = {
         "StyleRefLoad",
         "StyleRefApply",
@@ -128,7 +128,7 @@ def test_link_ids_are_unique_and_registered_on_both_ends(path):
 
 @pytest.mark.parametrize("path", FILES, ids=[_relpath(p) for p in FILES])
 def test_no_nodes_overlap_at_default_sizes(path):
-    """Plan P2-1: overlapping nodes read as a broken template — layout is math,
+    """Overlapping nodes read as a broken template — layout is math,
     not guesses, so this must hold for every template."""
     nodes = _load(path)["nodes"]
     boxes = [(n["id"], *n["pos"], *n["size"]) for n in nodes]
@@ -142,7 +142,7 @@ def test_no_nodes_overlap_at_default_sizes(path):
 def test_every_workflow_explains_itself_in_a_visible_note(path):
     """
     Templates are documentation — and ComfyUI never renders extra.styleref, so
-    the instructions must live in a visible core Note node (plan P2-2).
+    the instructions must live in a visible core Note node.
     """
     graph = _load(path)
     assert len(_note_text(graph)) > 60, "no substantial Note node found"
@@ -168,14 +168,14 @@ def test_auth_requirements_are_stated(path):
     needs_auth = any(n["type"] == "StyleRefLogin" for n in graph["nodes"])
     if needs_auth:
         assert "sign-in" in notes or "sign in" in notes
-        # The sign-in path ships in the template (plan P2-6), not just in prose.
+        # The sign-in path ships in the template, not just in prose.
         assert any(n["type"] == "StyleRefLogin" for n in graph["nodes"])
 
 
 @pytest.mark.parametrize("path", FILES, ids=[_relpath(p) for p in FILES])
 def test_seeds_randomize(path):
     """
-    Plan P2-4: a fixed seed makes the second queue a silent cache no-op —
+    A fixed seed makes the second queue a silent cache no-op —
     deadly in a demo.
     """
     for node in _load(path)["nodes"]:
@@ -190,7 +190,7 @@ _MODEL_DIRECTORIES = {"checkpoints", "diffusion_models", "text_encoders", "vae"}
 
 @pytest.mark.parametrize("path", FILES, ids=[_relpath(p) for p in FILES])
 def test_model_loaders_carry_download_metadata(path):
-    """Plan P2-7: a missing model should be a guided download, not a red error —
+    """A missing model should be a guided download, not a red error —
     for every loader type (checkpoints and the Z-Image UNET/CLIP/VAE stack)."""
     graph = _load(path)
     loader_files = {
@@ -288,7 +288,7 @@ def test_zimage_templates_have_no_negative_encoder(name):
 
 
 def test_consistency_template_shows_one_style_many_subjects():
-    """Plan P2-5: three Apply nodes with three different subjects, one Load."""
+    """Three Apply nodes with three different subjects, one Load."""
     graph = _load(os.path.join(WORKFLOW_DIR, "02-consistency-grid.json"))
     applies = [n for n in graph["nodes"] if n["type"] == "StyleRefApply"]
     loads = [n for n in graph["nodes"] if n["type"] == "StyleRefLoad"]
@@ -299,7 +299,7 @@ def test_consistency_template_shows_one_style_many_subjects():
 
 
 def test_prompt_previews_exist_where_promised():
-    """Plan P2-3: templates 01 and 04 make the compiled prompts visible."""
+    """Templates 01 and 04 make the compiled prompts visible."""
     for name in ("01-quickstart.json", "04-facets.json"):
         graph = _load(os.path.join(WORKFLOW_DIR, name))
         assert any(n["type"] == "PreviewAny" for n in graph["nodes"]), name