feat(diag): 加 glm.pro52 空响应定层探针(交生产跑)(bump 0.58.48)

失败面板 empty 簇(近7天15)主角从 opus48 换成 glm.pro52(11/15):assistant 轮
既无 tool_calls 又无正文,被 loop 当正常收尾静默 done。要判 glm 是①真吐空、
②工具意图漏成正文(narrated)、还是③瞬态。

scripts/diag_glm_empty_probe.py:build_agent(resume=True) 复原真实 system prompt +
全工具 schema + 完整多轮上下文(空响应发生在深层,单条 user 消息复现不了 —— 与
2a1bc25d narrated 探针的关键差别),打 glm.pro52,stream/non-stream 各一发,
三态判定 EMPTY/NARRATED/STRUCTURED。默认锁 task 35744bea(近7天 empty 5 次
最可复现),argv 可换 task/user/档做对照。

本机 import litellm 卡 ~20min,只能交生产机跑:
  .venv/Scripts/python.exe scripts/diag_glm_empty_probe.py
输出写 scripts/_glm_empty_probe_*.txt。定层后再决处置。

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

PROGRESS.md

@@ -2,7 +2,7 @@
 
 > 配合 `DESIGN.md`。本文件只记 phase 状态、决策偏差、文件量、下一步。每条 1-2 句:做了啥 + 关键判断;细节查 `git log` / `git diff` / `DESIGN §7.9`。
 
-最后更新:2026-07-20(文件预览背景撤回毛玻璃、改回半透明压暗遮罩,bump 0.58.46)
+最后更新:2026-07-20(加 glm.pro52 空响应定层探针,交生产跑,bump 0.58.48)
 
 ---
 
@@ -23,6 +23,8 @@
 
 ### 2026-07
 
+- **07-20 / 0.58.48 / glm.pro52 空响应定层探针**:失败面板 empty 簇(15)主角从 opus48 换成 `glm.pro52`(11/15)。加 `scripts/diag_glm_empty_probe.py`——`build_agent(resume=True)` 复原真实 system prompt + 全工具 schema + **完整多轮上下文**(空响应发生在深层,单条 user 消息复现不了,这是与 2a1bc25d narrated 探针的关键差别),打 glm.pro52,三态判定 EMPTY(真吐空)/NARRATED(工具意图漏成正文)/STRUCTURED。默认锁 task 35744bea(近7天 empty 5 次最可复现)。本机 import litellm 卡 ~20min,交生产机 `.venv/Scripts/python.exe scripts/diag_glm_empty_probe.py` 跑,输出写 `scripts/_glm_empty_probe_*.txt`。定层后再决处置。
+- **07-20 / 0.58.47 / 工具失败面板 6 簇巡检 + svg 质量门签名诚实化**:扫近7天失败聚集逐簇定根因。①**write/edit malformed(45/32)**=100% DeepSeek v4 wire 把并行 tool_call 的 arguments 拼成 `{...}{...}`,salvage(0.58.24)复盘提前到 07-20:命中率稳 ~85%(post-launch salvaged 252/malformed 44)、残差涉事 task 全在跑无终态失败=落非流式重试自愈,**决策保持现状不再写代码**(面板绝对数被上线前 07-13/14 尾巴撑高,要看 salvaged/malformed 比)。②**shell/exit "Font issues"(20)**=`svg_quality_checker` 质量门按设计 exit-1,但 `toolfail._classify` 取 `[exit]` 前尾行当签名,而尾行永远是通用 tip「4. Font issues:…」→ 所有门失败(typography/alignment/spec_lock drift)误标成 Font issues 误导排查。修:checker 在 errors>0 时末尾多打确定性 `gate_verdict_line()`「[GATE FAIL] svg_quality_checker: N error file(s) / M total; top issues: …」作为最后一行 stdout(放 main() 里 export 之后),toolfail 现有取尾行逻辑自动拾取诚实签名、toolfail 零改动;`_categorize_issue` 加 `spec_lock/typography` 类避免真错落 "Other"。补 3 测试(GateVerdictLineTests),39+3 全绿。③**run_python 预检(31)**=pysyntax 按设计拦中文正文硬拼进 .py(引号/缩进崩),非平台 bug=提示词治理待办。④**empty(15)**=主角换成 `glm.pro52`(11/15,非旧 opus48 案),需生产跑 narrated 探针定层(本机 import litellm 卡,task e7c2f50b 备用)。⑤**web_fetch(6)**=外部 403 拒抓为主,非平台锅。
 - **07-20 / 0.58.45 / glob 绝对路径崩溃(工具失败聚集 #2)**:巡检近7天工具失败聚集,`glob` 排第2(3次/3task/3用户、近24h 仍在发)。根因:`tools/fs.py` GlobTool 直接 `base.glob(pattern)`,模型传绝对 pattern(沙箱 cwd 就是 `/workspace`,自然写 `/workspace/**/*.py`)时 pathlib 抛 `NotImplementedError: Non-relative patterns are unsupported`,整个调用崩返 `[Error executing glob]`。改:绝对 pattern 先 `_split_abs_glob` 拆成「存在的静态前缀目录 + 相对 glob 串」(左起吃掉不含 `*?[` 的组件当 base)再匹配;补首字符 `/`\`\\` 判定(POSIX 前导斜杠在 Windows 宿主 `is_absolute()` 为 False,沙箱是 Linux,两处对齐)。顺手删掉原 `if "**"`/`else` 两分支代码完全相同的死重构。`/sandbox/tools/` 是 Dockerfile `COPY tools/` 拷贝,**需重建沙箱镜像**才生效。
 - **07-20 / 0.58.45 / PPT 页脚漂移治本(工具失败聚集 #1,A+C)**:`shell/exit` 排第1(近7天16条 exit-1、近24h 6条,9task/7用户),实测拆解 15/16 是**逐页手写绝对坐标偏离 `spec_lock.layout_grid` 锁定值 2–16px 的「想对齐没对齐」**(footer_y 近失 27行为最大头,margin_x/content_top 各7行,content_bottom 溢出12行),字体/spec_lock 只是 WARN 不 fail。关键发现:`executor-base.md §180-185` 早有「HARD—checker-enforced」的 snap 纪律却仍被漂移,加 prose 无用,治本须机械化。**A**:spec_lock_reference 新增 `## footer` 锁定片段——x/y 用 `margin_x`/`footer_y` 预填好、页码 `text-anchor="end"` 在 `canvas_w−margin_x`、留 `{PAGE}` 占位,executor 每页**逐字粘贴只换页码**,把 footer_y 这个跨页漂移第一大头从「每页手算」收敛成「一次编写」(executor-base 加对应硬规则,SKILL.md 同步);冒烟验证锁定片段过质检 errors=0、无 footer_y/margin_x 报错。**C**:`svg_quality_checker._ALIGN_TOL` 2→3px(2px 对手写实测过严),同步刷 SKILL/spec_lock_reference/executor-base 的「2–15px/2–12px」表述与测试值(margin drift 用例 60/63/66→60/64/68,30 测试全绿)。content_bottom 溢出(装太多项,另一性质)与 grid-snap 后处理(B)本轮未做。
 - **07-19 / 0.58.43**:文件预览遮罩改整屏、消除"上暗下亮"割裂——旧实现用 `#file-preview-modal { bottom: var(--preview-bottom-inset) }` 把遮罩顶到 chat-form 之上,好让预览时能边看边打字,代价是底部整条(输入区+左栏入口+右栏+状态)全不压暗,与上方压暗区形成一条硬边界的通亮带,很割裂。改为遮罩铺满整屏(`bottom`→`padding-bottom`,仍让居中卡片停在输入区之上不遮挡),底部随整屏统一压暗;仅 chat-form 由新 `body.fp-open` 类 `position:relative;z-index:95`(压过遮罩 z90,层叠上下文已核:`#app`/`#pane-mid` 均不建 context)+ 上向阴影,上浮成"悬浮活动条",继续可打字。预览时左栏/右栏/状态点击落到遮罩=关预览(符合"背景不可用"预期)。改 `web/static/dev.html` + `web/static/js/preview.js`(open/close 加/去 `fp-open`)。

core/__init__.py

@@ -1,3 +1,3 @@
 # zcbot 版本号单一事实源:web/app.py 的 FastAPI version、/healthz 返回、前端展示都引这里。
 # 改版本只动这一行。
-__version__ = "0.58.46"
+__version__ = "0.58.48"

scripts/diag_glm_empty_probe.py

@@ -0,0 +1,134 @@
+"""定层 glm.pro52 空响应(失败面板 kind=empty_response 主角,11/15/近7天)。
+
+背景:assistant 轮既无 tool_calls 又无正文,被 run loop 当正常收尾静默 done。
+与 opus48 的「tool_use 漏成正文」(narrated,见 diag_narrated_toolcall_2a1bc25d)
+是不同现象——要判 glm 到底是①真吐空(tc=0 且 content=0)、②漏成正文(narrated)、
+还是③正常(能回结构化 tool_calls,线上那次只是瞬态)。
+
+做法:build_agent(resume=True) 复原真实 system prompt + 全套工具 schema + **完整多轮
+上下文**(空响应发生在深层,单条 user 消息复现不了 —— 这是与 2a1bc25d 探针的关键差别),
+直接打 glm.pro52,stream / non-stream 各打一发。只读 DB,不跑 loop 不写库。
+输出 ASCII(Windows GBK 安全),完整正文写文件。
+
+用法(生产机跑,本机 import litellm 卡 ~20min):
+  .venv/Scripts/python.exe scripts/diag_glm_empty_probe.py [task_id] [user_id] [model_profile]
+默认 = 35744bea(近7天 empty 5 次、最可复现)。model_profile 缺省用 task 自身档(glm.pro52),
+传第三参可换档做对照(如 deepseek_v4.pro 看是否 glm 独有)。
+"""
+import os
+import sys
+from pathlib import Path
+from uuid import UUID
+
+ROOT = Path(__file__).resolve().parent.parent
+env = ROOT / ".env"
+if env.exists():
+    for line in env.read_text(encoding="utf-8").splitlines():
+        line = line.strip()
+        if line and not line.startswith("#") and "=" in line:
+            k, v = line.split("=", 1)
+            os.environ.setdefault(k.strip(), v.strip().strip('"').strip("'"))
+os.environ.setdefault("LITELLM_LOCAL_MODEL_COST_MAP", "True")
+
+sys.path.insert(0, str(ROOT))
+import litellm  # noqa: E402
+from core.agent_builder import build_agent  # noqa: E402
+
+# 默认锁定 35744bea(近7天 empty 5 次、最可复现);可 argv 覆盖。
+TASK_ID = sys.argv[1] if len(sys.argv) > 1 else "35744bea-49c0-41c8-86be-8d0c5c05bd32"
+USER_ID = UUID(sys.argv[2]) if len(sys.argv) > 2 else UUID("3cc07a22-f667-47f3-9617-e7d8a1a60cf0")
+MODEL_OVERRIDE = sys.argv[3] if len(sys.argv) > 3 else None
+
+_tag = (MODEL_OVERRIDE or "self").replace(".", "_").replace("/", "_")
+OUT = open(ROOT / "scripts" / f"_glm_empty_probe_{TASK_ID[:8]}_{_tag}.txt", "w", encoding="utf-8")
+
+
+def w(*a):
+    print(*a, file=OUT)
+
+
+agent, session, sid, tstate, wd = build_agent(
+    user_id=USER_ID, session_id=TASK_ID, resume=True,
+    model_name=MODEL_OVERRIDE,
+)
+schemas = agent.executor.schemas()
+caps = agent.caps
+llm = agent.llm
+
+# 复原 loop 下一轮真会送的完整消息(system + 全部历史)—— 空响应就发生在这个上下文上。
+messages = list(session.messages)
+sys_txt = next((m["content"] for m in messages if m.get("role") == "system"), "")
+roles = [m.get("role") for m in messages]
+
+w("[SETUP]")
+w(f"  task={TASK_ID} model_id={caps.model_id} profile={caps.family}.{caps.variant}")
+w(f"  api_base={llm.api_base} temp={caps.optimal_temperature} "
+  f"parallel_tools={caps.parallel_tools}")
+w(f"  system_prompt chars={len(sys_txt)}  tools={len(schemas)}")
+w(f"  messages count={len(messages)}  last_role={roles[-1] if roles else None}")
+w(f"  role tail={roles[-6:]}")
+w("")
+
+kwargs_base = dict(
+    model=caps.model_id,
+    api_base=llm.api_base,
+    api_key=llm.api_key,
+    messages=messages,
+    tools=schemas,
+    temperature=caps.optimal_temperature,
+    timeout=300,
+)
+if caps.parallel_tools:
+    kwargs_base["parallel_tool_calls"] = True
+
+stdout_lines = []
+
+
+def probe(tag, stream):
+    w(f"========== {tag} ==========")
+    try:
+        if stream:
+            chunks = list(litellm.completion(
+                **kwargs_base, stream=True,
+                stream_options={"include_usage": True}))
+            resp = litellm.stream_chunk_builder(chunks)
+        else:
+            resp = litellm.completion(**kwargs_base)
+    except Exception as e:
+        w(f"[FAIL] {type(e).__name__}: {str(e)[:300]}")
+        stdout_lines.append(f"{tag}: FAIL {type(e).__name__}")
+        return
+    msg = resp.choices[0].message
+    tc = msg.tool_calls or []
+    content = msg.content or ""
+    fr = resp.choices[0].finish_reason
+    u = resp.usage
+    w(f"  finish_reason={fr}")
+    w(f"  usage prompt={u.prompt_tokens} completion={u.completion_tokens}")
+    w(f"  tool_calls count={len(tc)}")
+    for i, t in enumerate(tc):
+        w(f"    [{i}] {t.function.name}({(t.function.arguments or '')[:200]})")
+    w(f"  content chars={len(content)}")
+    w("  content:")
+    w(content[:2000])
+    w("")
+    # 三态判定:EMPTY(真吐空)/ NARRATED(工具意图漏成正文)/ STRUCTURED / plain-text
+    narrated = (not tc) and ("`" in content) and ("\n- " in content or "No result received" in content)
+    if tc:
+        verdict = "STRUCTURED tool_calls"
+    elif not content.strip():
+        verdict = "EMPTY (tc=0 content=0)  <-- 复现空响应"
+    elif narrated:
+        verdict = "NARRATED-as-text"
+    else:
+        verdict = "plain-text-no-tools"
+    stdout_lines.append(f"{tag}: {verdict} (finish={fr}, tc={len(tc)}, content={len(content)}c)")
+
+
+probe("NON-STREAM", stream=False)
+probe("STREAM", stream=True)
+
+OUT.close()
+for ln in stdout_lines:
+    print(ln)
+print(f"full -> scripts/_glm_empty_probe_{TASK_ID[:8]}_{_tag}.txt")

skills/ppt/scripts/svg_quality_checker.py

@@ -20,7 +20,7 @@ import re
 import json
 import html
 from pathlib import Path
-from typing import List, Dict, Tuple
+from typing import List, Dict, Tuple, Optional
 from collections import Counter, defaultdict
 from xml.etree import ElementTree as ET
 
@@ -1095,6 +1095,8 @@ class SVGQualityChecker:
             return 'Alignment/grid'
         elif error_msg.startswith('Geometry:'):
             return 'Geometry'
+        elif 'spec_lock' in error_msg or 'typography' in error_msg:
+            return 'spec_lock/typography'
         elif 'font' in error_msg.lower():
             return 'Font issues'
         else:
@@ -2325,6 +2327,26 @@ class SVGQualityChecker:
                   "(<path>/<polygon>/<polyline>/<circle>/<image>) — verify dense content "
                   "pages aren't just text + boxes.")
 
+    def gate_verdict_line(self) -> Optional[str]:
+        """One deterministic line summarizing the gate outcome, meant to be the
+        LAST stdout line on failure.
+
+        Failure-cluster tooling (core/toolfail) signatures a non-zero shell exit
+        by the last content line before ``[exit N]``. Without this, that line is
+        always the generic trailing tip ``4. Font issues: ...`` — so every kind
+        of quality-gate failure (typography px, alignment drift, spec_lock, ...)
+        collapses onto one misleading "Font issues" cluster. This gives it a
+        stable, honest signature instead. ``None`` when there are no errors.
+        """
+        if self.summary['errors'] <= 0:
+            return None
+        top = sorted(self.issue_types.items(), key=lambda kv: (-kv[1], kv[0]))[:3]
+        cats = ", ".join(k for k, _ in top) or "unspecified"
+        return (
+            f"[GATE FAIL] svg_quality_checker: {self.summary['errors']} error "
+            f"file(s) / {self.summary['total']} total; top issues: {cats}"
+        )
+
     def print_summary(self):
         """Print check summary"""
         print("=" * 80)
@@ -2576,6 +2598,12 @@ def main() -> None:
                 output_file = sys.argv[idx + 1]
         checker.export_report(output_file)
 
+    # Deterministic gate verdict as the LAST stdout line on failure (after any
+    # --export [REPORT] line) so core/toolfail signatures this cluster honestly.
+    verdict = checker.gate_verdict_line()
+    if verdict:
+        print(verdict)
+
     # Return exit code
     if checker.summary['errors'] > 0:
         sys.exit(1)

tests/test_svg_alignment_check.py

@@ -331,5 +331,48 @@ class SpecContractTests(unittest.TestCase):
         self.assertFalse(any("CJK" in e for e in r["errors"]))
 
 
+class GateVerdictLineTests(unittest.TestCase):
+    """gate_verdict_line() 给 core/toolfail 一条稳定诚实的失败签名,
+    替掉「[exit] 前最后一行永远是通用 tip → 全塌成 Font issues」的误标。"""
+
+    def _checker(self, body: str):
+        with TemporaryDirectory() as tmp:
+            page = _write_page(Path(tmp), "01_cover.svg", body)
+            checker = SVGQualityChecker()
+            checker.check_file(str(page))
+            return checker
+
+    def test_none_when_no_errors(self):
+        # 干净页面(无 error)不出 gate 行 —— exit 0 本就不进失败面板
+        c = self._checker(_text(60, 100, "干净标题"))
+        self.assertEqual(c.summary["errors"], 0)
+        self.assertIsNone(c.gate_verdict_line())
+
+    def test_gate_line_on_error(self):
+        # foreignObject 是 error;gate 行须以 [GATE FAIL] svg_quality_checker 打头,
+        # 带文件计数与 top issues —— toolfail 归一后是诚实签名而非 "Font issues"
+        c = self._checker('<foreignObject x="0" y="0" width="9" height="9"><b/></foreignObject>')
+        self.assertGreater(c.summary["errors"], 0)
+        line = c.gate_verdict_line()
+        self.assertIsNotNone(line)
+        self.assertTrue(line.startswith("[GATE FAIL] svg_quality_checker:"))
+        self.assertIn("top issues:", line)
+        self.assertIn("foreignObject", line)
+
+    def test_gate_line_is_last_stdout_line(self):
+        # 契约:gate 行是 print_summary 之后的最后一行 stdout(toolfail 取尾行)
+        import contextlib
+        import io
+        c = self._checker('<foreignObject x="0" y="0" width="9" height="9"><b/></foreignObject>')
+        buf = io.StringIO()
+        with contextlib.redirect_stdout(buf):
+            c.print_summary()
+            line = c.gate_verdict_line()
+            if line:
+                print(line)
+        last = [ln for ln in buf.getvalue().splitlines() if ln.strip()][-1]
+        self.assertTrue(last.startswith("[GATE FAIL] svg_quality_checker:"))
+
+
 if __name__ == "__main__":
     unittest.main()