localization: pin UTF-8 explicitly on every read_text/write_text call

src/socrates120x/audit/checks.py

@@ -136,7 +136,7 @@ def _load_ignore_list(project: Path, section: str) -> set[str]:
     if not config_path.is_file():
         return set()
     try:
-        data = json.loads(config_path.read_text())
+        data = json.loads(config_path.read_text(encoding="utf-8"))
     except (OSError, ValueError):
         return set()
     if not isinstance(data, dict):
@@ -198,7 +198,7 @@ def run(self, project: Path) -> list[Finding]:
             path = project / adapter
             if not path.is_file():
                 continue
-            text = path.read_text(errors="replace")
+            text = path.read_text(errors="replace", encoding="utf-8")
             if "AGENTS.md" not in text:
                 findings.append(Finding(
                     check=self.name,
@@ -229,7 +229,7 @@ def run(self, project: Path) -> list[Finding]:
             acc = sprint / "acceptance.md"
             if not acc.is_file():
                 continue
-            for line_no, line in enumerate(acc.read_text(errors="replace").splitlines(), 1):
+            for line_no, line in enumerate(acc.read_text(errors="replace", encoding="utf-8").splitlines(), 1):
                 lower = line.lower()
                 for weasel in WEASEL_WORDS:
                     if weasel.lower() in lower:
@@ -258,7 +258,7 @@ def run(self, project: Path) -> list[Finding]:
         state = project / "planning" / "STATE.md"
         if not state.is_file():
             return []  # Already flagged by RequiredFilesCheck.
-        text = state.read_text(errors="replace")
+        text = state.read_text(errors="replace", encoding="utf-8")
         m = self._DATE.search(text)
         if not m:
             return [Finding(
@@ -294,7 +294,7 @@ def run(self, project: Path) -> list[Finding]:
         risks = project / "planning" / "RISKS.md"
         if not risks.is_file():
             return []
-        lower = risks.read_text(errors="replace").lower()
+        lower = risks.read_text(errors="replace", encoding="utf-8").lower()
         if not any(phrase.lower() in lower for phrase in ALWAYS_ON_RISK_PHRASES):
             return [Finding(
                 check=self.name,
@@ -318,7 +318,7 @@ def run(self, project: Path) -> list[Finding]:
         domain = project / "planning" / "DOMAIN.md"
         if not domain.is_file():
             return []
-        terms = self._extract_terms(domain.read_text(errors="replace"))
+        terms = self._extract_terms(domain.read_text(errors="replace", encoding="utf-8"))
         if not terms:
             return []
 
@@ -366,15 +366,15 @@ def _concatenate_other_files(self, project: Path) -> str:
         for rel in candidates:
             path = project / rel
             if path.is_file():
-                parts.append(path.read_text(errors="replace"))
+                parts.append(path.read_text(errors="replace", encoding="utf-8"))
         # Also include sprint files.
         sprints = project / "planning" / "sprints"
         if sprints.is_dir():
             for sprint in sprints.iterdir():
                 if not sprint.is_dir():
                     continue
                 for f in sprint.glob("*.md"):
-                    parts.append(f.read_text(errors="replace"))
+                    parts.append(f.read_text(errors="replace", encoding="utf-8"))
         return "\n".join(parts)
 
 

src/socrates120x/audit/companyos_checks.py

@@ -101,7 +101,7 @@ def run(self, project: Path) -> list[Finding]:
         for pattern in sorted(patterns.glob("*.md")):
             if pattern.name == "README.md":
                 continue
-            body = pattern.read_text(errors="replace")
+            body = pattern.read_text(errors="replace", encoding="utf-8")
             m = self._SOURCE_LINE.search(body)
             if not m:
                 continue
@@ -137,7 +137,7 @@ def run(self, project: Path) -> list[Finding]:
         if not proposals.is_file() or not builds.is_dir():
             return []
         build_names = {p.name for p in builds.iterdir() if p.is_dir()}
-        body = proposals.read_text(errors="replace")
+        body = proposals.read_text(errors="replace", encoding="utf-8")
         findings: list[Finding] = []
         seen: set[str] = set()
         for m in self._SLUG.finditer(body):
@@ -167,7 +167,7 @@ def _build_client_reference(build: Path) -> str | None:
     if answers.is_file():
         import json
         try:
-            data = json.loads(answers.read_text())
+            data = json.loads(answers.read_text(encoding="utf-8"))
             if isinstance(data, dict):
                 v = data.get("client")
                 if isinstance(v, str) and v.strip():
@@ -176,7 +176,7 @@ def _build_client_reference(build: Path) -> str | None:
             pass
     agents = build / "AGENTS.md"
     if agents.is_file():
-        text = agents.read_text(errors="replace")
+        text = agents.read_text(errors="replace", encoding="utf-8")
         m = re.search(r"Client:\s*\*\*([^*\n]+)\*\*", text)
         if m:
             return m.group(1).strip()

src/socrates120x/companyos.py

@@ -57,7 +57,7 @@ def scaffold_companyos(target: Path, *, overwrite: bool = False) -> list[Path]:
     for rel, body in files.items():
         path = target / rel
         path.parent.mkdir(parents=True, exist_ok=True)
-        path.write_text(body)
+        path.write_text(body, encoding="utf-8")
         written.append(path)
     return written
 

src/socrates120x/decide.py

@@ -39,9 +39,9 @@ def record_decision(project: Path, text: str) -> int:
     today = _dt.date.today().isoformat()
     bullet = f"- **{cleaned} ({today})**"
 
-    body = decisions_path.read_text()
+    body = decisions_path.read_text(encoding="utf-8")
     new_body = _insert_decision(body, bullet)
-    decisions_path.write_text(new_body)
+    decisions_path.write_text(new_body, encoding="utf-8")
     print(f"Appended to {decisions_path}:")
     print(f"  {bullet}")
     return 0

src/socrates120x/extract.py

@@ -138,7 +138,7 @@ def run_extract(
 
     slug = _sanitize_slug(iv.answers.get("pattern_slug", "untitled"))
     target = target_dir / f"CANDIDATE-{slug}.md"
-    target.write_text(render_pattern(iv.answers, project=project))
+    target.write_text(render_pattern(iv.answers, project=project), encoding="utf-8")
     print(f"\nPattern candidate written to: {target}")
     print()
     print("Next steps:")

src/socrates120x/interview.py

@@ -194,10 +194,10 @@ class Interview:
 
     def load(self) -> None:
         if self.answers_path.exists() and self.resume:
-            self.answers = json.loads(self.answers_path.read_text())
+            self.answers = json.loads(self.answers_path.read_text(encoding="utf-8"))
 
     def save(self) -> None:
-        self.answers_path.write_text(json.dumps(self.answers, indent=2) + "\n")
+        self.answers_path.write_text(json.dumps(self.answers, indent=2) + "\n", encoding="utf-8")
 
     def run(
         self,

src/socrates120x/journal.py

@@ -32,7 +32,7 @@ def create_or_open_entry(project: Path, *, show: bool = False, list_all: bool =
     entry = journal_dir / f"{today}.md"
     is_new = not entry.exists()
     if is_new:
-        entry.write_text(_template(today))
+        entry.write_text(_template(today), encoding="utf-8")
         print(f"Created {entry}")
 
     cmd = editor_command()
@@ -93,5 +93,5 @@ def _show_latest(journal_dir: Path) -> int:
     if not entries:
         print("(no journal entries yet)")
         return 0
-    print(entries[0].read_text())
+    print(entries[0].read_text(encoding="utf-8"))
     return 0

src/socrates120x/onboard.py

@@ -35,7 +35,7 @@ def _load_answers_json(project: Path) -> dict[str, Any] | None:
     if not path.is_file():
         return None
     try:
-        data = json.loads(path.read_text())
+        data = json.loads(path.read_text(encoding="utf-8"))
         if isinstance(data, dict):
             return data
     except (OSError, ValueError):
@@ -79,12 +79,12 @@ def _synthesize_from_markdown(project: Path) -> str:
     today = _dt.date.today().isoformat()
     name = project.name
 
-    readme = (project / "README.md").read_text(errors="replace") if (project / "README.md").is_file() else ""
-    agents = (project / "AGENTS.md").read_text(errors="replace") if (project / "AGENTS.md").is_file() else ""
-    state = (project / "planning" / "STATE.md").read_text(errors="replace") if (project / "planning" / "STATE.md").is_file() else ""
-    decisions = (project / "planning" / "DECISIONS.md").read_text(errors="replace") if (project / "planning" / "DECISIONS.md").is_file() else ""
-    risks = (project / "planning" / "RISKS.md").read_text(errors="replace") if (project / "planning" / "RISKS.md").is_file() else ""
-    questions = (project / "planning" / "QUESTIONS.md").read_text(errors="replace") if (project / "planning" / "QUESTIONS.md").is_file() else ""
+    readme = (project / "README.md").read_text(errors="replace", encoding="utf-8") if (project / "README.md").is_file() else ""
+    agents = (project / "AGENTS.md").read_text(errors="replace", encoding="utf-8") if (project / "AGENTS.md").is_file() else ""
+    state = (project / "planning" / "STATE.md").read_text(errors="replace", encoding="utf-8") if (project / "planning" / "STATE.md").is_file() else ""
+    decisions = (project / "planning" / "DECISIONS.md").read_text(errors="replace", encoding="utf-8") if (project / "planning" / "DECISIONS.md").is_file() else ""
+    risks = (project / "planning" / "RISKS.md").read_text(errors="replace", encoding="utf-8") if (project / "planning" / "RISKS.md").is_file() else ""
+    questions = (project / "planning" / "QUESTIONS.md").read_text(errors="replace", encoding="utf-8") if (project / "planning" / "QUESTIONS.md").is_file() else ""
 
     tagline = _extract_tagline(readme, agents)
     client = _extract_field(agents, "Client") or _extract_field(readme, "Client")
@@ -177,7 +177,7 @@ def write_welcome(project: Path) -> Path:
     """Write WELCOME.md into the project root and return its path."""
     body = synthesize_welcome(project)
     target = project / "WELCOME.md"
-    target.write_text(body)
+    target.write_text(body, encoding="utf-8")
     return target
 
 

src/socrates120x/pack.py

@@ -119,7 +119,7 @@ def write_pack(
         format=format,
     )
     target = project / f".socrates-architect-pack.{FORMAT_EXTENSIONS[format]}"
-    target.write_text(body)
+    target.write_text(body, encoding="utf-8")
     return target
 
 
@@ -233,7 +233,7 @@ def _kit_sections(kit: Path) -> list[_Section]:
         path = kit / name
         if not path.is_file():
             continue
-        text = path.read_text(errors="replace").strip()
+        text = path.read_text(errors="replace", encoding="utf-8").strip()
         if not text:
             continue
         out.append(_Section(
@@ -253,7 +253,7 @@ def _file_section(path: Path, *, rel_display: str, label: str) -> _Section:
             path=rel_display,
             kind="missing",
         )
-    text = path.read_text(errors="replace").strip()
+    text = path.read_text(errors="replace", encoding="utf-8").strip()
     if not text:
         return _Section(
             label=f"{label}  (`{rel_display}`)",

src/socrates120x/patterns.py

@@ -85,7 +85,7 @@ def review_patterns(companyos_root: Path, *, use_cache: bool = True) -> PatternR
             candidates += 1
         else:
             promoted += 1
-        body = pattern.read_text(errors="replace")
+        body = pattern.read_text(errors="replace", encoding="utf-8")
         source = _extract_source_project(body)
         extracted = _extract_extracted_date(body)
 
@@ -215,7 +215,7 @@ def _slug_in_project(slug: str, project_dir: Path) -> bool:
     needle = slug.lower()
     for f in project_dir.rglob("*.md"):
         try:
-            text = f.read_text(errors="replace").lower()
+            text = f.read_text(errors="replace", encoding="utf-8").lower()
         except OSError:
             continue
         if needle in text:
@@ -245,7 +245,7 @@ def _load_usage_cache(patterns_dir: Path) -> dict[str, Any] | None:
     if not cache_path.is_file():
         return None
     try:
-        data = json.loads(cache_path.read_text())
+        data = json.loads(cache_path.read_text(encoding="utf-8"))
     except (OSError, ValueError):
         return None
     if not isinstance(data, dict):
@@ -261,7 +261,7 @@ def _save_usage_cache(patterns_dir: Path, payload: dict[str, Any]) -> None:
     if not patterns_dir.is_dir():
         return
     with contextlib.suppress(OSError):
-        (patterns_dir / USAGE_CACHE_FILENAME).write_text(json.dumps(payload, indent=2))
+        (patterns_dir / USAGE_CACHE_FILENAME).write_text(json.dumps(payload, indent=2), encoding="utf-8")
 
 
 def format_pattern_report(report: PatternReport, *, use_color: bool | None = None) -> str:

src/socrates120x/prompting.py

@@ -160,7 +160,7 @@ def _ask_multiline_editor(output_fn: OutputFn, q: Question) -> str:
     try:
         output_fn(f"   (opening {editor[0]} — save & quit to submit)")
         subprocess.run([*editor, str(tmp_path)], check=True)
-        raw = tmp_path.read_text()
+        raw = tmp_path.read_text(encoding="utf-8")
     finally:
         with contextlib.suppress(OSError):
             tmp_path.unlink()

src/socrates120x/render.py

@@ -42,7 +42,7 @@ def render_all(target: Path, answers: dict[str, Any]) -> list[Path]:
     for rel, body in files.items():
         path = target / rel
         path.parent.mkdir(parents=True, exist_ok=True)
-        path.write_text(body)
+        path.write_text(body, encoding="utf-8")
         written.append(path)
     return written
 

src/socrates120x/ship.py

@@ -137,7 +137,7 @@ def _extract_check(project: Path) -> ShipFinding:
         if not loc.is_dir():
             continue
         for f in loc.glob("CANDIDATE-*.md"):
-            text = f.read_text(errors="replace")
+            text = f.read_text(errors="replace", encoding="utf-8")
             if f"`{project.name}`" in text:
                 found = True
                 break
@@ -180,7 +180,7 @@ def _state_check(project: Path) -> ShipFinding:
     answers_path = project / ".socrates-answers.json"
     if answers_path.is_file():
         try:
-            data = json.loads(answers_path.read_text())
+            data = json.loads(answers_path.read_text(encoding="utf-8"))
         except (OSError, ValueError):
             data = None
         if isinstance(data, dict) and data.get("state_next"):
@@ -189,7 +189,7 @@ def _state_check(project: Path) -> ShipFinding:
             pass
     # Fall back to recency check via the embedded date.
     import re
-    m = re.search(r"Last updated:\s*(\d{4}-\d{2}-\d{2})", state.read_text(errors="replace"))
+    m = re.search(r"Last updated:\s*(\d{4}-\d{2}-\d{2})", state.read_text(errors="replace", encoding="utf-8"))
     if not m:
         return ShipFinding(
             name="state",

src/socrates120x/status.py

@@ -134,7 +134,7 @@ def _extract_tagline(project: Path) -> str:
     answers_path = project / ".socrates-answers.json"
     if answers_path.is_file():
         try:
-            data = json.loads(answers_path.read_text())
+            data = json.loads(answers_path.read_text(encoding="utf-8"))
             if isinstance(data, dict):
                 t = data.get("tagline")
                 if isinstance(t, str):
@@ -143,7 +143,7 @@ def _extract_tagline(project: Path) -> str:
             pass
     agents = project / "AGENTS.md"
     if agents.is_file():
-        m = re.search(r"\*\*[^*]+\*\*\s+—\s+(.+)", agents.read_text(errors="replace"))
+        m = re.search(r"\*\*[^*]+\*\*\s+—\s+(.+)", agents.read_text(errors="replace", encoding="utf-8"))
         if m:
             return m.group(1).strip()
     return ""
@@ -171,7 +171,7 @@ def _state_age_days(project: Path) -> int | None:
     state = project / "planning" / "STATE.md"
     if not state.is_file():
         return None
-    m = _DATE.search(state.read_text(errors="replace"))
+    m = _DATE.search(state.read_text(errors="replace", encoding="utf-8"))
     if not m:
         return None
     try:
@@ -208,9 +208,9 @@ def _has_extract(project: Path) -> bool:
         sibling = parent.parent / "patterns"
         if sibling.is_dir():
             for f in sibling.glob("CANDIDATE-*.md"):
-                if f"Source project | `{project.name}`" in f.read_text(errors="replace"):
+                if f"Source project | `{project.name}`" in f.read_text(errors="replace", encoding="utf-8"):
                     return True
-                if f"`{project.name}`" in f.read_text(errors="replace"):
+                if f"`{project.name}`" in f.read_text(errors="replace", encoding="utf-8"):
                     return True
     # 3) Or has an in-progress extract answers file.
     return (project / ".socrates-extract-answers.json").is_file()

src/socrates120x/timeline.py

@@ -91,7 +91,7 @@ def _journal_events(project: Path) -> list[TimelineEvent]:
             d = _dt.date.fromisoformat(entry.stem)
         except ValueError:
             continue
-        first_line = _first_real_line(entry.read_text(errors="replace"))
+        first_line = _first_real_line(entry.read_text(errors="replace", encoding="utf-8"))
         events.append(TimelineEvent(
             date=d,
             kind=EventKind.JOURNAL,
@@ -119,7 +119,7 @@ def _sprint_events(project: Path) -> list[TimelineEvent]:
         req = sprint / "requirements.md"
         detail = ""
         if req.is_file():
-            detail = _extract_goal(req.read_text(errors="replace"))
+            detail = _extract_goal(req.read_text(errors="replace", encoding="utf-8"))
         events.append(TimelineEvent(
             date=mtime,
             kind=EventKind.SPRINT,
@@ -137,7 +137,7 @@ def _decision_events(project: Path) -> list[TimelineEvent]:
     if not decisions_file.is_file():
         return []
     events: list[TimelineEvent] = []
-    for line in decisions_file.read_text(errors="replace").splitlines():
+    for line in decisions_file.read_text(errors="replace", encoding="utf-8").splitlines():
         stripped = line.lstrip()
         if not stripped.startswith("- "):
             continue