fix: patch checks full target span + marker injection/integrity (algo #8)

core/evolution/evolve.py

@@ -19,6 +19,8 @@ from loguru import logger
 from core.evolution.patch import (
     APPENDIX_END,
     APPENDIX_START,
+    MOMENTUM_END,
+    MOMENTUM_START,
     append_to_appendix,
     apply_patch_with_report,
     extract_appendix_notes,
@@ -293,10 +295,39 @@ def _tool_protected_spans(text: str) -> list[str]:
 # =========================================================================
 
 
+def _check_marker_integrity(evolved: str) -> list[str]:
+    """校验 evolved 中冻结区 marker 的完整性（成对、至多一对、START 先于 END）。
+
+    进化写入可能破坏 appendix/momentum marker 配对，破坏后 append_to_appendix /
+    replace_momentum 等下游会静默误拼或抛错。此处集中拦截：任一 marker 对违反
+    「START 数==END 数、各至多一对、START 在 END 前」即整体 reject。
+
+    参数:
+        evolved: 改写后的全文。
+    返回:
+        错误信息列表（空列表表示 marker 完整）。
+    """
+    errors: list[str] = []
+    for name, start_m, end_m in (
+        ("APPENDIX", APPENDIX_START, APPENDIX_END),
+        ("MOMENTUM", MOMENTUM_START, MOMENTUM_END),
+    ):
+        s = evolved.count(start_m)
+        e = evolved.count(end_m)
+        if s != e:
+            errors.append(f"{name} marker 不配对：START={s} END={e}")
+        elif s > 1:
+            errors.append(f"{name} marker 出现多对（{s}），至多一对")
+        elif s == 1 and evolved.index(start_m) > evolved.index(end_m):
+            errors.append(f"{name} marker 顺序错误：END 出现在 START 之前")
+    return errors
+
+
 def validate_skill(original: str, evolved: str) -> ValidationResult:
     """校验 Skill 改写结果。
 
     检查项: frontmatter 三字段保留（name / description / task_type）、
+    marker 完整性（appendix/momentum 成对且至多一对、顺序正确）、
     长度比在 [0.3, 2.0]、代码块闭合。
 
     参数:
@@ -305,6 +336,11 @@ def validate_skill(original: str, evolved: str) -> ValidationResult:
 
     返回:
         ValidationResult 实例。
+
+    关键实现细节:
+        marker 完整性先于长度校验——长度校验经 _strip_protected_regions 调用
+        momentum_region_bounds，对损坏 marker 会抛 ValueError；故 marker 破坏时先
+        返回失败，避免异常穿透且明确 reject 该候选。
     """
     errors: list[str] = []
     orig_fm = _parse_frontmatter(original)
@@ -319,6 +355,10 @@ def validate_skill(original: str, evolved: str) -> ValidationResult:
                 errors.append(
                     f"frontmatter 字段 {key} 被修改: {orig_fm.get(key)!r} → {evol_fm.get(key)!r}"
                 )
+    marker_errors = _check_marker_integrity(evolved)
+    if marker_errors:
+        errors.extend(marker_errors)
+        return ValidationResult(passed=False, errors=errors)
     errors.extend(_check_length(original, evolved))
     errors.extend(_check_code_blocks(evolved))
     return ValidationResult(passed=len(errors) == 0, errors=errors)

core/evolution/patch.py

@@ -282,9 +282,34 @@ def _protected_ranges(content: str, spans: list[str]) -> list[tuple[int, int]]:
     return ranges
 
 
-def _in_ranges(pos: int, ranges: list[tuple[int, int]]) -> bool:
+def _span_overlaps_ranges(pos: int, length: int, ranges: list[tuple[int, int]]) -> bool:
-    """判断位置 pos 是否落在任意冻结区间内。"""
+    """判断 [pos, pos+length) 是否与任一冻结区间相交（不止起点）。
-    return any(start <= pos < end for start, end in ranges)
+
+    起点落在正文、末端伸入冻结区的 target 也须拦截，否则 replace/delete 会连带
+    改动冻结区（如破坏 appendix/momentum marker）。
+
+    参数:
+        pos: target 在正文中的起点。
+        length: target 长度。
+        ranges: 冻结区间 [start, end) 列表。
+    返回:
+        与任一区间相交返回 True。
+    """
+    end = pos + length
+    return any(start < end and pos < r_end for start, r_end in ranges)
+
+
+# 冻结区 marker 字面量：LLM 生成的 edit 不得注入这些字面量，否则破坏 marker 配对
+_MARKER_LITERALS = (APPENDIX_START, APPENDIX_END, MOMENTUM_START, MOMENTUM_END)
+
+
+def _edit_injects_marker(edit: dict) -> bool:
+    """判断 edit 的 target/content 是否含冻结区 marker 字面量（注入拦截）。"""
+    for key in ("target", "content"):
+        value = edit.get(key)
+        if isinstance(value, str) and any(m in value for m in _MARKER_LITERALS):
+            return True
+    return False
 
 
 def _append_at(content: str, ranges: list[tuple[int, int]]) -> int:
@@ -317,8 +342,8 @@ def _do_insert_after(
             _insert_at(content, _append_at(content, ranges), payload),
             "applied_insert_after_fallback",
         )
-    if _in_ranges(pos, ranges):
+    if _span_overlaps_ranges(pos, len(target), ranges):
-        logger.warning("insert_after 目标在冻结区，跳过 target={}", target[:80])
+        logger.warning("insert_after 目标跨入冻结区，跳过 target={}", target[:80])
         return content, "skipped_protected"
     at = pos + len(target)
     nl = content.find("\n", at)
@@ -340,8 +365,8 @@ def _do_replace_delete(
     if pos == -1:
         logger.warning("{} 锚点缺失，跳过 target={}", op, target[:80])
         return content, "skipped_target_not_found"
-    if _in_ranges(pos, ranges):
+    if _span_overlaps_ranges(pos, len(target), ranges):
-        logger.warning("{} 目标在冻结区，跳过 target={}", op, target[:80])
+        logger.warning("{} 目标跨入冻结区，跳过 target={}", op, target[:80])
         return content, "skipped_protected"
     new_content = content.replace(target, payload if op == "replace" else "", 1)
     return new_content, "applied_" + op
@@ -403,6 +428,17 @@ def apply_patch_with_report(
     reports: list[dict] = []
     for i, edit in enumerate(edits, 1):
         try:
+            if isinstance(edit, dict) and _edit_injects_marker(edit):
+                logger.warning("edit 含冻结区 marker 字面量，拒绝该 edit index={}", i)
+                report = {
+                    "op": str(edit.get("op", "")),
+                    "target": str(edit.get("target", "") or "")[:200],
+                    "content_preview": str(edit.get("content", "") or "")[:200],
+                    "status": "skipped_marker_injection",
+                }
+                report["index"] = i
+                reports.append(report)
+                continue
             ranges = _protected_ranges(content, spans)
             content, report = _apply_one(content, edit, ranges)
         except (KeyError, TypeError, ValueError, AttributeError) as exc:

tests/unit/test_evolve.py

@@ -309,6 +309,36 @@ class TestValidateSkill:
         result = validate_skill(orig, "no frontmatter body")
         assert not result.passed
 
+    def test_unpaired_appendix_marker_fails(self) -> None:
+        """evolved 出现孤立 APPENDIX_START（无配对 END）→ marker 完整性校验失败。"""
+        orig = "---\nname: a\ndescription: d\ntask_type: t\n---\nbody body body"
+        evol = f"---\nname: a\ndescription: d\ntask_type: t\n---\nbody {APPENDIX_START} body"
+        result = validate_skill(orig, evol)
+        assert not result.passed
+        assert any("marker" in e.lower() or "配对" in e for e in result.errors)
+
+    def test_duplicate_momentum_marker_fails(self) -> None:
+        """evolved 出现两对 MOMENTUM marker → 完整性校验失败（各至多一对）。"""
+        orig = "---\nname: a\ndescription: d\ntask_type: t\n---\nbody body body"
+        evol = (
+            "---\nname: a\ndescription: d\ntask_type: t\n---\n"
+            f"{MOMENTUM_START}x{MOMENTUM_END} mid {MOMENTUM_START}y{MOMENTUM_END}"
+        )
+        result = validate_skill(orig, evol)
+        assert not result.passed
+        assert any("marker" in e.lower() or "配对" in e for e in result.errors)
+
+    def test_paired_markers_pass(self) -> None:
+        """evolved 含成对 appendix+momentum marker（各一对，顺序正确）→ 校验通过。"""
+        body = "body " * 20
+        orig = f"---\nname: a\ndescription: d\ntask_type: t\n---\n{body}"
+        evol = (
+            f"---\nname: a\ndescription: d\ntask_type: t\n---\n{body}"
+            f"{APPENDIX_START}\n- n\n{APPENDIX_END}\n{MOMENTUM_START}\nm\n{MOMENTUM_END}"
+        )
+        result = validate_skill(orig, evol)
+        assert result.passed, result.errors
+
 
 class TestValidateSystem:
     """validate_system 测试。"""

tests/unit/test_patch.py

@@ -383,3 +383,68 @@ class TestApplyPatch:
         assert "line2" not in out
         assert "TAIL" in out
         assert all(r["status"].startswith("applied") for r in reports)
+
+
+class TestSpanOverlapProtection:
+    """冻结区跨度保护 + marker 注入拦截（算法 #8 加固）。"""
+
+    def test_replace_spanning_into_protected_marker_preserved(self) -> None:
+        """target 末端跨入含 marker 的 appendix 冻结区时被拦截，marker 不被破坏。"""
+        body = "正文最后一段。"
+        appendix = f"{APPENDIX_START}\n## 执行提醒\n- 规则A\n{APPENDIX_END}"
+        content = body + "\n\n" + appendix
+        # target 从正文末尾跨入 APPENDIX_START（含 marker 字面量）
+        target = "正文最后一段。\n\n" + APPENDIX_START
+        edits = [{"op": "delete", "target": target, "content": ""}]
+        new_content, report = apply_patch_with_report(
+            content, edits, protected_spans=[appendix]
+        )
+        # 无论经 marker 注入拦截还是跨度拦截，marker 都必须完整保留
+        assert APPENDIX_START in new_content and APPENDIX_END in new_content
+        assert report[0]["status"] in ("skipped_protected", "skipped_marker_injection")
+
+    def test_replace_spanning_into_frozen_section_is_skipped(self) -> None:
+        """target 起点在正文、末端伸入无 marker 的冻结区段 → 跨度拦截跳过。"""
+        body = "可改正文段落。"
+        frozen = "## 输出格式\n必须输出 JSON。"
+        content = body + "\n" + frozen
+        # target 从可改正文跨入冻结区段（不含任何 marker 字面量）
+        target = "可改正文段落。\n## 输出格式"
+        edits = [{"op": "replace", "target": target, "content": "破坏内容"}]
+        new_content, report = apply_patch_with_report(
+            content, edits, protected_spans=[frozen]
+        )
+        assert "破坏内容" not in new_content
+        assert "## 输出格式" in new_content
+        assert report[0]["status"] == "skipped_protected"
+
+    def test_insert_after_spanning_into_frozen_section_is_skipped(self) -> None:
+        """insert_after 的 target 末端伸入无 marker 冻结区段时跨度拦截跳过。"""
+        body = "可改正文。"
+        frozen = "## 输出格式\n固定内容。"
+        content = body + "\n" + frozen
+        target = "可改正文。\n## 输出格式"
+        edits = [{"op": "insert_after", "target": target, "content": "注入内容"}]
+        new_content, report = apply_patch_with_report(
+            content, edits, protected_spans=[frozen]
+        )
+        assert "注入内容" not in new_content
+        assert report[0]["status"] == "skipped_protected"
+
+    def test_edit_payload_with_marker_literal_rejected(self) -> None:
+        """edit payload 含 marker 字面量 → 拒绝该 edit。"""
+        edits = [
+            {"op": "append", "target": "", "content": f"注入 {APPENDIX_START} 破坏"}
+        ]
+        _, report = apply_patch_with_report("正文", edits, protected_spans=[])
+        assert any(
+            "marker" in str(s).lower() or "reject" in str(s).lower() for s in report[0].values()
+        )
+
+    def test_edit_target_with_marker_literal_rejected(self) -> None:
+        """edit target 含 momentum marker 字面量 → 拒绝该 edit。"""
+        edits = [
+            {"op": "replace", "target": f"{MOMENTUM_START}x", "content": "y"}
+        ]
+        _, report = apply_patch_with_report("正文", edits, protected_spans=[])
+        assert report[0]["status"] == "skipped_marker_injection"