feat(migration): 建立 Skill 迁移工作流与追踪工具

2026-08-25 10:46:09 +08:00
parent 795c331706
commit 0e7eb1eb93
10 changed files with 1063 additions and 3 deletions
@@ -0,0 +1,79 @@
+#!/usr/bin/env python3
+"""检查 CraftKit Skill 的基础结构、引用和敏感内容。"""
+
+from __future__ import annotations
+
+import argparse
+import json
+import re
+from pathlib import Path
+
+NAME_RE = re.compile(r"^[a-z0-9]+(?:-[a-z0-9]+)*$")
+LINK_RE = re.compile(r"\[[^]]+]\((?!https?://|#)([^)]+)\)")
+SECRET_RE = re.compile(
+    r"(?i)(api[_-]?key|password|private[_-]?key|access[_-]?token)\s*[:=]\s*[^\s]+"
+)
+
+
+def frontmatter(text: str) -> dict[str, str]:
+    """解析当前校验所需的简单顶层 frontmatter 字段。"""
+    if not text.startswith("---\n"):
+        return {}
+    end = text.find("\n---", 4)
+    if end < 0:
+        return {}
+    values: dict[str, str] = {}
+    for line in text[4:end].splitlines():
+        if ":" in line and not line.startswith((" ", "\t")):
+            key, value = line.split(":", 1)
+            values[key.strip()] = value.strip().strip("'\"")
+    return values
+
+
+def check_skill(root: Path, forbidden: list[str]) -> dict:
+    """返回结构化问题列表，不修改被检查目录。"""
+    issues: list[dict[str, str]] = []
+    skill_md = root / "SKILL.md"
+    if not skill_md.is_file():
+        return {"skill": root.name, "issues": [{"code": "missing-skill-md", "path": "SKILL.md"}]}
+    text = skill_md.read_text(encoding="utf-8-sig")
+    metadata = frontmatter(text)
+    name = metadata.get("name", "")
+    if not NAME_RE.fullmatch(name):
+        issues.append({"code": "invalid-name", "path": "SKILL.md"})
+    if name != root.name:
+        issues.append({"code": "name-path-mismatch", "path": "SKILL.md"})
+    if not metadata.get("description"):
+        issues.append({"code": "missing-description", "path": "SKILL.md"})
+
+    for path in sorted(item for item in root.rglob("*") if item.is_file()):
+        relative = path.relative_to(root).as_posix()
+        content = path.read_text(encoding="utf-8-sig", errors="replace")
+        if "[TODO:" in content:
+            issues.append({"code": "todo-placeholder", "path": relative})
+        if SECRET_RE.search(content):
+            issues.append({"code": "possible-secret", "path": relative})
+        lowered = content.casefold()
+        for term in forbidden:
+            if term.casefold() in lowered:
+                issues.append({"code": f"forbidden:{term}", "path": relative})
+        if path.suffix.lower() == ".md":
+            for link in LINK_RE.findall(content):
+                target = link.split("#", 1)[0]
+                if target and not (path.parent / target).resolve().is_file():
+                    issues.append({"code": "broken-link", "path": f"{relative}:{link}"})
+    return {"skill": root.name, "issues": issues}
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="检查 CraftKit Skill")
+    parser.add_argument("skill", nargs="+", type=Path)
+    parser.add_argument("--forbid", action="append", default=[])
+    args = parser.parse_args()
+    results = [check_skill(path.resolve(), args.forbid) for path in args.skill]
+    print(json.dumps({"results": results}, ensure_ascii=False, indent=2))
+    raise SystemExit(1 if any(item["issues"] for item in results) else 0)
+
+
+if __name__ == "__main__":
+    main()