chore(consistency): sync task status from DB [auto]

Updated by fix-consistency on 2026-05-15:
  - update .custodian-brief.md for repo-scoping

src/repo_scoping/scope/__init__.py

@@ -0,0 +1,4 @@
+from repo_registry.scope.generator import ScopeGenerator
+from repo_registry.scope.validator import ScopeValidator
+
+__all__ = ["ScopeGenerator", "ScopeValidator"]

src/repo_scoping/scope/generator.py

@@ -0,0 +1,323 @@
+from __future__ import annotations
+
+import re
+from dataclasses import asdict
+
+from repo_registry.core.service import RegistryService
+from repo_registry.storage.sqlite import NotFoundError
+
+
+SCOPE_SECTIONS = [
+    "One-liner",
+    "Core Idea",
+    "In Scope",
+    "Out of Scope",
+    "Relevant When",
+    "Not Relevant When",
+    "Current State",
+    "How It Fits",
+    "Terminology",
+    "Related / Overlapping",
+    "Getting Oriented",
+    "Provided Capabilities",
+    "Notes",
+]
+
+
+NEEDS_INPUT = "<!-- needs curator input -->"
+
+
+class ScopeGenerator:
+    """Render SCOPE.md from approved repository characteristics."""
+
+    def __init__(self, service: RegistryService) -> None:
+        self.service = service
+
+    def generate(self, repo_slug: str) -> str:
+        repository = self._repository_by_slug(repo_slug)
+        ability_map = asdict(self.service.ability_map(repository.id))
+        facts = [asdict(fact) for fact in self.service.list_observed_facts(repository.id)]
+        sections = {
+            "One-liner": self._one_liner(ability_map),
+            "Core Idea": self._core_idea(ability_map),
+            "In Scope": self._in_scope(ability_map),
+            "Out of Scope": self._curator_stub(),
+            "Relevant When": self._relevant_when(ability_map),
+            "Not Relevant When": self._curator_stub(),
+            "Current State": self._current_state(repository.status, facts),
+            "How It Fits": self._how_it_fits(ability_map),
+            "Terminology": self._terminology(ability_map, facts),
+            "Related / Overlapping": self._curator_stub(),
+            "Getting Oriented": self._getting_oriented(ability_map, facts),
+            "Provided Capabilities": self._provided_capabilities(ability_map),
+            "Notes": self._curator_stub(),
+        }
+        lines = [
+            "# SCOPE",
+            "",
+            "> This file helps you quickly understand what this repository is about,",
+            "> when it is relevant, and when it is not.",
+            "> It was generated from approved repo-scoping characteristics.",
+            "",
+            "---",
+            "",
+        ]
+        for section in SCOPE_SECTIONS:
+            lines.extend([f"## {section}", "", sections[section].rstrip(), "", "---", ""])
+        return "\n".join(lines).rstrip() + "\n"
+
+    def _repository_by_slug(self, repo_slug: str):
+        wanted = self._slug(repo_slug)
+        for repository in self.service.list_repositories():
+            candidates = {
+                self._slug(repository.name),
+                self._slug(repository.url.rstrip("/").rsplit("/", 1)[-1].removesuffix(".git")),
+            }
+            if wanted in candidates:
+                return repository
+        raise NotFoundError(f"repository slug {repo_slug!r} was not found")
+
+    def _one_liner(self, ability_map: dict) -> str:
+        scope = ability_map["scope"]
+        description = self._sentence(scope.get("description", ""))
+        if description:
+            return description
+        return f"{scope['name']} defines the repository scope for {ability_map['repository']['name']}."
+
+    def _core_idea(self, ability_map: dict) -> str:
+        scope = ability_map["scope"]
+        abilities = ability_map.get("abilities", [])
+        lines = [scope.get("description") or self._one_liner(ability_map)]
+        if abilities:
+            lines.append("")
+            lines.append("Approved abilities:")
+            lines.extend(
+                f"- {ability['name']} — {ability.get('description') or 'Approved repository ability.'}"
+                for ability in abilities[:5]
+            )
+        else:
+            lines.extend(["", NEEDS_INPUT])
+        return "\n".join(lines)
+
+    def _in_scope(self, ability_map: dict) -> str:
+        abilities = ability_map.get("abilities", [])
+        if not abilities:
+            return self._curator_stub()
+        lines = []
+        for ability in abilities:
+            capabilities = ", ".join(
+                capability["name"] for capability in ability.get("capabilities", [])[:4]
+            )
+            suffix = f" Includes {capabilities}." if capabilities else ""
+            lines.append(
+                f"- {ability['name']} — {ability.get('description') or 'Approved ability.'}{suffix}"
+            )
+        return "\n".join(lines)
+
+    def _relevant_when(self, ability_map: dict) -> str:
+        features = [
+            feature
+            for feature in self._features(ability_map)
+            if self._is_usecase_feature(feature)
+        ]
+        if not features:
+            features = self._features(ability_map)[:5]
+        if not features:
+            return self._curator_stub()
+        lines = [
+            f"- You need {feature['name']} ({feature.get('primary_class') or feature.get('type', 'feature')})."
+            for feature in features
+        ]
+        if not any(self._is_usecase_feature(feature) for feature in features):
+            lines.append(NEEDS_INPUT)
+        return "\n".join(lines)
+
+    def _current_state(self, status: str, facts: list[dict]) -> str:
+        kinds = self._facts_by_kind(facts)
+        languages = self._fact_names(kinds.get("language", []))
+        frameworks = self._fact_names(kinds.get("framework", []))
+        tests = kinds.get("test", [])
+        interfaces = kinds.get("interface", [])
+        manifests = kinds.get("manifest", [])
+        implementation = "substantial" if interfaces or manifests else "partial"
+        if not facts:
+            implementation = "unknown"
+        lines = [
+            f"- Status: {status}",
+            f"- Implementation: {implementation}",
+            "- Stability: evolving",
+            "- Usage: internal",
+            f"- Languages: {', '.join(languages) if languages else 'unknown'}",
+            f"- Frameworks: {', '.join(frameworks) if frameworks else 'none detected'}",
+            f"- Tests observed: {len(tests)}",
+            f"- Interfaces observed: {len(interfaces)}",
+            f"- Manifests observed: {len(manifests)}",
+        ]
+        if not facts:
+            lines.append(NEEDS_INPUT)
+        return "\n".join(lines)
+
+    def _how_it_fits(self, ability_map: dict) -> str:
+        evidence = [
+            item
+            for capability in self._capabilities(ability_map)
+            for item in capability.get("evidence", [])
+        ]
+        if not evidence:
+            return "\n".join(
+                [
+                    "- Upstream dependencies: " + NEEDS_INPUT,
+                    "- Downstream consumers: " + NEEDS_INPUT,
+                    "- Often used with: " + NEEDS_INPUT,
+                ]
+            )
+        refs = ", ".join(
+            sorted({item.get("reference", "") for item in evidence if item.get("reference")})[:8]
+        )
+        return "\n".join(
+            [
+                f"- Supported by evidence references: {refs or 'available evidence'}",
+                "- Upstream dependencies: " + NEEDS_INPUT,
+                "- Downstream consumers: " + NEEDS_INPUT,
+                "- Often used with: " + NEEDS_INPUT,
+            ]
+        )
+
+    def _terminology(self, ability_map: dict, facts: list[dict]) -> str:
+        terms = set()
+        for item in [ability_map["scope"], *ability_map.get("abilities", [])]:
+            terms.add(item.get("name", ""))
+            terms.add(item.get("primary_class", ""))
+            terms.update(item.get("attributes", []))
+        for capability in self._capabilities(ability_map):
+            terms.add(capability.get("name", ""))
+            terms.add(capability.get("primary_class", ""))
+            terms.update(capability.get("attributes", []))
+        for fact in facts:
+            if fact.get("kind") in {"framework", "llm_provider", "provider_registry"}:
+                terms.add(fact.get("name", ""))
+        visible = [term for term in sorted(terms) if term]
+        if not visible:
+            return self._curator_stub()
+        return "\n".join(
+            [
+                "- Preferred terms: " + ", ".join(visible[:12]),
+                "- Also known as: " + NEEDS_INPUT,
+                "- Potentially confusing terms: " + NEEDS_INPUT,
+            ]
+        )
+
+    def _getting_oriented(self, ability_map: dict, facts: list[dict]) -> str:
+        paths = self._source_paths(ability_map, facts)
+        if not paths:
+            return self._curator_stub()
+        return "\n".join(
+            [
+                f"- Start with: {paths[0]}",
+                f"- Key files / directories: {', '.join(paths[:8])}",
+                f"- Entry points: {', '.join(paths[:5])}",
+            ]
+        )
+
+    def _provided_capabilities(self, ability_map: dict) -> str:
+        capabilities = self._capabilities(ability_map)
+        if not capabilities:
+            return f"<!-- No approved capabilities yet. -->\n{NEEDS_INPUT}"
+        blocks = []
+        for capability in capabilities:
+            keywords = self._keywords_for_capability(capability)
+            blocks.append(
+                "\n".join(
+                    [
+                        "```capability",
+                        f"type: {self._capability_type(capability.get('primary_class', 'other'))}",
+                        f"title: {capability['name']}",
+                        "description: >",
+                        f"  {capability.get('description') or 'Approved repository capability.'}",
+                        f"keywords: [{', '.join(keywords)}]",
+                        "```",
+                    ]
+                )
+            )
+        return "\n\n".join(blocks)
+
+    def _capabilities(self, ability_map: dict) -> list[dict]:
+        return [
+            capability
+            for ability in ability_map.get("abilities", [])
+            for capability in ability.get("capabilities", [])
+        ]
+
+    def _features(self, ability_map: dict) -> list[dict]:
+        return [
+            feature
+            for capability in self._capabilities(ability_map)
+            for feature in capability.get("features", [])
+        ]
+
+    def _is_usecase_feature(self, feature: dict) -> bool:
+        labels = {str(feature.get("primary_class", "")).lower()}
+        labels.update(str(item).lower() for item in feature.get("attributes", []))
+        return bool(labels & {"business-usecase", "usecase", "workflow", "review"})
+
+    def _keywords_for_capability(self, capability: dict) -> list[str]:
+        keywords = [capability.get("primary_class", "")]
+        keywords.extend(capability.get("attributes", []))
+        for feature in capability.get("features", []):
+            keywords.append(feature.get("primary_class", ""))
+            keywords.extend(feature.get("attributes", []))
+        return [self._keyword(item) for item in self._unique(keywords)[:8] if item]
+
+    def _capability_type(self, primary_class: str) -> str:
+        normalized = primary_class.lower()
+        if normalized in {"api", "infrastructure", "data", "security", "documentation"}:
+            return normalized
+        if normalized in {"interface", "integration", "llm-integration"}:
+            return "api"
+        if normalized in {"storage", "repository-structure"}:
+            return "data"
+        return "other"
+
+    def _facts_by_kind(self, facts: list[dict]) -> dict[str, list[dict]]:
+        grouped: dict[str, list[dict]] = {}
+        for fact in facts:
+            grouped.setdefault(fact.get("kind", ""), []).append(fact)
+        return grouped
+
+    def _fact_names(self, facts: list[dict]) -> list[str]:
+        return self._unique([fact.get("name", "") for fact in facts])
+
+    def _source_paths(self, ability_map: dict, facts: list[dict]) -> list[str]:
+        paths = [fact.get("path", "") for fact in facts if fact.get("path")]
+        for feature in self._features(ability_map):
+            paths.append(feature.get("location", ""))
+            for source_ref in feature.get("source_refs", []):
+                paths.append(source_ref.get("path", ""))
+        return self._unique(paths)
+
+    def _curator_stub(self) -> str:
+        return f"- {NEEDS_INPUT}"
+
+    def _sentence(self, text: str) -> str:
+        cleaned = re.sub(r"\s+", " ", text.strip())
+        if not cleaned:
+            return ""
+        return re.split(r"(?<=[.!?])\s+", cleaned, maxsplit=1)[0]
+
+    def _slug(self, value: str) -> str:
+        return re.sub(r"[^a-z0-9]+", "-", value.lower()).strip("-")
+
+    def _keyword(self, value: str) -> str:
+        return self._slug(value) or "other"
+
+    def _unique(self, values: list[str]) -> list[str]:
+        result: list[str] = []
+        seen: set[str] = set()
+        for value in values:
+            item = str(value).strip()
+            key = item.lower()
+            if not item or key in seen:
+                continue
+            seen.add(key)
+            result.append(item)
+        return result

src/repo_scoping/scope/validator.py

@@ -0,0 +1,184 @@
+from __future__ import annotations
+
+import re
+from dataclasses import dataclass
+from pathlib import Path
+
+from repo_registry.scope.generator import SCOPE_SECTIONS, ScopeGenerator
+
+
+@dataclass(frozen=True)
+class ScopeDiffSection:
+    section: str
+    status: str
+    current_text: str | None
+    proposed_text: str | None
+
+
+@dataclass(frozen=True)
+class ScopeDiff:
+    sections: list[ScopeDiffSection]
+
+    @property
+    def needs_update(self) -> bool:
+        return any(section.status != "ok" for section in self.sections)
+
+
+@dataclass(frozen=True)
+class ScopeValidationIssue:
+    check: str
+    severity: str
+    message: str
+
+
+@dataclass(frozen=True)
+class ValidationResult:
+    issues: list[ScopeValidationIssue]
+
+    @property
+    def ok(self) -> bool:
+        return not any(issue.severity == "error" for issue in self.issues)
+
+
+class ScopeValidator:
+    """Validate and diff SCOPE.md files."""
+
+    def __init__(self, generator: ScopeGenerator | None = None) -> None:
+        self.generator = generator
+
+    def diff(self, repo_slug: str, existing_path: Path) -> ScopeDiff:
+        if self.generator is None:
+            raise ValueError("ScopeValidator.diff requires a ScopeGenerator")
+        current = existing_path.read_text(encoding="utf-8") if existing_path.exists() else ""
+        proposed = self.generator.generate(repo_slug)
+        current_sections = self._parse_sections(current)
+        proposed_sections = self._parse_sections(proposed)
+        sections: list[ScopeDiffSection] = []
+        for section in SCOPE_SECTIONS:
+            current_text = current_sections.get(section)
+            proposed_text = proposed_sections.get(section, "")
+            if current_text is None:
+                status = "missing"
+            elif self._normalize(current_text) == self._normalize(proposed_text):
+                status = "ok"
+            else:
+                status = "stale"
+            sections.append(
+                ScopeDiffSection(
+                    section=section,
+                    status=status,
+                    current_text=current_text,
+                    proposed_text=proposed_text,
+                )
+            )
+        return ScopeDiff(sections=sections)
+
+    def validate(self, path: Path) -> ValidationResult:
+        issues: list[ScopeValidationIssue] = []
+        if not path.exists():
+            return ValidationResult(
+                issues=[
+                    ScopeValidationIssue(
+                        check="C5a",
+                        severity="error",
+                        message="SCOPE.md is missing.",
+                    )
+                ]
+            )
+        content = path.read_text(encoding="utf-8")
+        sections = self._parse_sections(content)
+        missing = [section for section in SCOPE_SECTIONS if section not in sections]
+        if missing:
+            severity = "warn" if missing == ["Provided Capabilities"] else "error"
+            issues.append(
+                ScopeValidationIssue(
+                    check="C5b",
+                    severity=severity,
+                    message=f"Missing SCOPE.md section(s): {', '.join(missing)}.",
+                )
+            )
+        ordered = self._heading_order(content)
+        expected_order = [section for section in SCOPE_SECTIONS if section in sections]
+        if ordered[: len(expected_order)] != expected_order:
+            issues.append(
+                ScopeValidationIssue(
+                    check="C5b",
+                    severity="warn",
+                    message="SCOPE.md sections are not in canonical order.",
+                )
+            )
+        capabilities = sections.get("Provided Capabilities")
+        if capabilities is None:
+            issues.append(
+                ScopeValidationIssue(
+                    check="C5c",
+                    severity="warn",
+                    message="Provided Capabilities section is missing.",
+                )
+            )
+        elif "```capability" in capabilities:
+            for index, block in enumerate(self._capability_blocks(capabilities), start=1):
+                keys = self._capability_keys(block)
+                missing_keys = {"type", "title"} - keys
+                if missing_keys:
+                    issues.append(
+                        ScopeValidationIssue(
+                            check="C5c",
+                            severity="warn",
+                            message=(
+                                f"Capability block {index} is missing required field(s): "
+                                f"{', '.join(sorted(missing_keys))}."
+                            ),
+                        )
+                    )
+        elif "No approved capabilities yet" not in capabilities:
+            issues.append(
+                ScopeValidationIssue(
+                    check="C5c",
+                    severity="warn",
+                    message=(
+                        "Provided Capabilities has no capability blocks or explicit "
+                        "empty-state note."
+                    ),
+                )
+            )
+        return ValidationResult(issues=issues)
+
+    def _parse_sections(self, content: str) -> dict[str, str]:
+        matches = list(re.finditer(r"^##\s+(.+?)\s*$", content, re.MULTILINE))
+        sections: dict[str, str] = {}
+        for index, match in enumerate(matches):
+            title = match.group(1).strip()
+            start = match.end()
+            end = matches[index + 1].start() if index + 1 < len(matches) else len(content)
+            body = content[start:end]
+            body = re.sub(r"\n---\s*$", "", body.strip())
+            sections[title] = body.strip()
+        return sections
+
+    def _heading_order(self, content: str) -> list[str]:
+        return [
+            match.group(1).strip()
+            for match in re.finditer(r"^##\s+(.+?)\s*$", content, re.MULTILINE)
+            if match.group(1).strip() in SCOPE_SECTIONS
+        ]
+
+    def _normalize(self, value: str | None) -> str:
+        if value is None:
+            return ""
+        without_comments = re.sub(r"<!--.*?-->", "", value, flags=re.DOTALL)
+        without_markdown = re.sub(r"[`*_>#-]+", " ", without_comments)
+        return re.sub(r"\s+", " ", without_markdown).strip().lower()
+
+    def _capability_blocks(self, content: str) -> list[str]:
+        return re.findall(
+            r"```capability\s*(.*?)```",
+            content,
+            flags=re.DOTALL | re.IGNORECASE,
+        )
+
+    def _capability_keys(self, block: str) -> set[str]:
+        return {
+            match.group(1)
+            for match in re.finditer(r"^([A-Za-z_][A-Za-z0-9_-]*):", block, re.MULTILINE)
+        }