Add W3C Baggage propagation

README.md

@@ -13,6 +13,7 @@
 
 ## Table of Contents
 
+- [What's new (2026-06-21) — W3C Baggage Propagation](#whats-new-2026-06-21--w3c-baggage-propagation)
 - [What's new (2026-06-21) — Dataset Diff (Row-Set Change Report)](#whats-new-2026-06-21--dataset-diff-row-set-change-report)
 - [What's new (2026-06-21) — Distribution Drift Detection](#whats-new-2026-06-21--distribution-drift-detection)
 - [What's new (2026-06-21) — Layered Configuration Resolver](#whats-new-2026-06-21--layered-configuration-resolver)
@@ -136,6 +137,12 @@
 
 ---
 
+## What's new (2026-06-21) — W3C Baggage Propagation
+
+Carry cross-cutting key-value context across HTTP. Full reference: [`docs/source/Eng/doc/new_features/v84_features_doc.rst`](docs/source/Eng/doc/new_features/v84_features_doc.rst).
+
+- **`Baggage` / `parse_baggage` / `format_baggage` / `inject_baggage` / `extract_baggage`** (`AC_baggage_parse`, `AC_baggage_format`): `trace_context` carried trace/span identity but nothing propagated cross-cutting context (`run_id`/`tenant`/`experiment`). This implements the W3C Baggage header — a percent-encoded `key=value` list — with an immutable `Baggage` (set/remove return new instances) and case-insensitive inject/extract over a headers dict. Pairs with `trace_context`. Pure-stdlib, deterministic.
+
 ## What's new (2026-06-21) — Dataset Diff (Row-Set Change Report)
 
 Diff two tabular extracts by key. Full reference: [`docs/source/Eng/doc/new_features/v83_features_doc.rst`](docs/source/Eng/doc/new_features/v83_features_doc.rst).

README/README_zh-CN.md

@@ -12,6 +12,7 @@
 
 ## 目录
 
+- [本次更新 (2026-06-21) — W3C Baggage 传播](#本次更新-2026-06-21--w3c-baggage-传播)
 - [本次更新 (2026-06-21) — 数据集差异(数据行变更报告)](#本次更新-2026-06-21--数据集差异数据行变更报告)
 - [本次更新 (2026-06-21) — 分布漂移检测](#本次更新-2026-06-21--分布漂移检测)
 - [本次更新 (2026-06-21) — 分层配置解析器](#本次更新-2026-06-21--分层配置解析器)
@@ -135,6 +136,12 @@
 
 ---
 
+## 本次更新 (2026-06-21) — W3C Baggage 传播
+
+跨 HTTP 携带横切键值上下文。完整参考:[`docs/source/Zh/doc/new_features/v84_features_doc.rst`](../docs/source/Zh/doc/new_features/v84_features_doc.rst)。
+
+- **`Baggage` / `parse_baggage` / `format_baggage` / `inject_baggage` / `extract_baggage`**(`AC_baggage_parse`、`AC_baggage_format`):`trace_context` 携带 trace/span 身份,但没有东西传播横切上下文(`run_id`/`tenant`/`experiment`)。本功能实现 W3C Baggage 标头 —— percent-encoded 的 `key=value` 列表 —— 以不可变的 `Baggage`(set/remove 返回新实例)与不分大小写的 inject/extract。与 `trace_context` 搭配。纯标准库、确定。
+
 ## 本次更新 (2026-06-21) — 数据集差异(数据行变更报告)
 
 按键比对两份表格式提取。完整参考:[`docs/source/Zh/doc/new_features/v83_features_doc.rst`](../docs/source/Zh/doc/new_features/v83_features_doc.rst)。

README/README_zh-TW.md

@@ -12,6 +12,7 @@
 
 ## 目錄
 
+- [本次更新 (2026-06-21) — W3C Baggage 傳播](#本次更新-2026-06-21--w3c-baggage-傳播)
 - [本次更新 (2026-06-21) — 資料集差異(資料列變更報告)](#本次更新-2026-06-21--資料集差異資料列變更報告)
 - [本次更新 (2026-06-21) — 分布漂移偵測](#本次更新-2026-06-21--分布漂移偵測)
 - [本次更新 (2026-06-21) — 分層設定解析器](#本次更新-2026-06-21--分層設定解析器)
@@ -135,6 +136,12 @@
 
 ---
 
+## 本次更新 (2026-06-21) — W3C Baggage 傳播
+
+跨 HTTP 攜帶橫切鍵值脈絡。完整參考:[`docs/source/Zh/doc/new_features/v84_features_doc.rst`](../docs/source/Zh/doc/new_features/v84_features_doc.rst)。
+
+- **`Baggage` / `parse_baggage` / `format_baggage` / `inject_baggage` / `extract_baggage`**(`AC_baggage_parse`、`AC_baggage_format`):`trace_context` 攜帶 trace/span 身分,但沒有東西傳播橫切脈絡(`run_id`/`tenant`/`experiment`)。本功能實作 W3C Baggage 標頭 —— percent-encoded 的 `key=value` 清單 —— 以不可變的 `Baggage`(set/remove 回傳新實例)與不分大小寫的 inject/extract。與 `trace_context` 搭配。純標準函式庫、具決定性。
+
 ## 本次更新 (2026-06-21) — 資料集差異(資料列變更報告)
 
 依鍵比對兩份表格式萃取。完整參考:[`docs/source/Zh/doc/new_features/v83_features_doc.rst`](../docs/source/Zh/doc/new_features/v83_features_doc.rst)。

docs/source/Eng/doc/new_features/v84_features_doc.rst

@@ -0,0 +1,41 @@
+W3C Baggage Propagation
+=======================
+
+``trace_context`` carries trace and span identity across an HTTP boundary, but
+there was no way to propagate cross-cutting key-value context (``run_id`` /
+``tenant`` / ``experiment``) alongside it. This implements the W3C Baggage
+header — a percent-encoded ``key=value`` list — so a run can attach such context
+to outgoing requests and read it back on the other side.
+
+Pure standard library (``urllib.parse``); imports no ``PySide6``. ``Baggage`` is
+immutable (mutators return a new instance), so propagation is deterministic.
+
+Headless API
+------------
+
+.. code-block:: python
+
+    from je_auto_control import Baggage, inject_baggage, extract_baggage
+
+    bag = Baggage({"tenant": "acme"}).set("run_id", "42")
+    headers = inject_baggage(outgoing_headers, bag)
+    # headers["baggage"] == "tenant=acme,run_id=42"
+
+    received = extract_baggage(request_headers)
+    tenant = received.get("tenant")
+
+``Baggage`` wraps an immutable key-value map: ``get`` reads, ``set`` / ``remove``
+return new instances, and ``to_dict`` exports the entries. ``parse_baggage``
+reads the header (dropping optional ``;metadata`` and rejecting empty keys),
+``format_baggage`` percent-encodes keys and values back into a header value,
+and ``inject_baggage`` / ``extract_baggage`` write and read the ``baggage``
+header on a request dict (extraction is case-insensitive). Pairs naturally with
+``trace_context`` to carry context alongside the trace.
+
+Executor commands
+-----------------
+
+``AC_baggage_parse`` parses a ``header`` into ``{items}``; ``AC_baggage_format``
+serialises an ``items`` object into ``{header}``. Both are exposed as MCP tools
+(``ac_baggage_parse`` / ``ac_baggage_format``) and as Script Builder commands
+under **Data**.

docs/source/Eng/eng_index.rst

@@ -106,6 +106,7 @@ Comprehensive guides for all AutoControl features.
    doc/new_features/v81_features_doc
    doc/new_features/v82_features_doc
    doc/new_features/v83_features_doc
+   doc/new_features/v84_features_doc
    doc/ocr_backends/ocr_backends_doc
    doc/observability/observability_doc
    doc/operations_layer/operations_layer_doc

docs/source/Zh/doc/new_features/v84_features_doc.rst

@@ -0,0 +1,35 @@
+W3C Baggage 傳播
+===============
+
+``trace_context`` 能跨 HTTP 邊界攜帶 trace 與 span 身分,但沒有辦法在旁邊一併傳播橫切的鍵值脈絡
+(``run_id`` / ``tenant`` / ``experiment``)。本功能實作 W3C Baggage 標頭 —— 一個 percent-encoded 的
+``key=value`` 清單 —— 讓一次執行能把這類脈絡附加到外送請求,並在另一端讀回。
+
+純標準函式庫(``urllib.parse``);不匯入 ``PySide6``。``Baggage`` 不可變(變更操作回傳新實例),因此
+傳播具決定性。
+
+無頭 API
+--------
+
+.. code-block:: python
+
+    from je_auto_control import Baggage, inject_baggage, extract_baggage
+
+    bag = Baggage({"tenant": "acme"}).set("run_id", "42")
+    headers = inject_baggage(outgoing_headers, bag)
+    # headers["baggage"] == "tenant=acme,run_id=42"
+
+    received = extract_baggage(request_headers)
+    tenant = received.get("tenant")
+
+``Baggage`` 包裝一個不可變的鍵值對應:``get`` 讀取,``set`` / ``remove`` 回傳新實例,``to_dict`` 匯出
+條目。``parse_baggage`` 解析標頭(去除選用的 ``;metadata`` 並拒絕空鍵),``format_baggage`` 將鍵與值
+percent-encode 回標頭值,``inject_baggage`` / ``extract_baggage`` 在請求 dict 上寫入與讀取 ``baggage``
+標頭(讀取不分大小寫)。與 ``trace_context`` 自然搭配,在 trace 之外攜帶脈絡。
+
+執行器命令
+----------
+
+``AC_baggage_parse`` 把 ``header`` 解析成 ``{items}``;``AC_baggage_format`` 把 ``items`` 物件序列化成
+``{header}``。兩者皆以 MCP 工具(``ac_baggage_parse`` / ``ac_baggage_format``)以及 Script Builder 中
+**Data** 分類下的命令提供。

docs/source/Zh/zh_index.rst

@@ -106,6 +106,7 @@ AutoControl 所有功能的完整使用指南。
    doc/new_features/v81_features_doc
    doc/new_features/v82_features_doc
    doc/new_features/v83_features_doc
+   doc/new_features/v84_features_doc
    doc/ocr_backends/ocr_backends_doc
    doc/observability/observability_doc
    doc/operations_layer/operations_layer_doc

je_auto_control/__init__.py

@@ -410,6 +410,10 @@
     format_traceparent, format_tracestate, inject_context, new_root_context,
     new_span_id, new_trace_id, parse_traceparent, parse_tracestate,
 )
+# W3C Baggage propagation (cross-cutting key-value context)
+from je_auto_control.utils.baggage import (
+    Baggage, extract_baggage, format_baggage, inject_baggage, parse_baggage,
+)
 # Background popup/interrupt watchdog (unattended automation)
 from je_auto_control.utils.watchdog import (
     PopupWatchdog, WatchdogRule, default_popup_watchdog,
@@ -922,6 +926,8 @@ def start_autocontrol_gui(*args, **kwargs):
     "format_traceparent", "format_tracestate", "inject_context",
     "new_root_context", "new_span_id", "new_trace_id", "parse_traceparent",
     "parse_tracestate",
+    "Baggage", "extract_baggage", "format_baggage", "inject_baggage",
+    "parse_baggage",
     # MCP server
     "AuditLogger", "HttpMCPServer", "MCPContent", "MCPPrompt",
     "MCPPromptArgument", "MCPResource", "MCPServer", "MCPTool",

je_auto_control/gui/script_builder/command_schema.py

@@ -1624,6 +1624,22 @@ def _add_resilience_specs(specs: List[CommandSpec]) -> None:
         ),
         description="Extract the W3C trace context from request headers.",
     ))
+    specs.append(CommandSpec(
+        "AC_baggage_parse", "Data", "Baggage: Parse Header",
+        fields=(
+            FieldSpec("header", FieldType.STRING,
+                      placeholder="tenant=acme,run=42"),
+        ),
+        description="Parse a W3C baggage header into key-value items.",
+    ))
+    specs.append(CommandSpec(
+        "AC_baggage_format", "Data", "Baggage: Format Header",
+        fields=(
+            FieldSpec("items", FieldType.STRING,
+                      placeholder='{"tenant": "acme", "run": "42"}'),
+        ),
+        description="Serialise items into a percent-encoded baggage header.",
+    ))
     specs.append(CommandSpec(
         "AC_profile_rows", "Data", "Data Profile: Profile Rows",
         fields=(

je_auto_control/utils/baggage/__init__.py

@@ -0,0 +1,9 @@
+"""W3C Baggage propagation for AutoControl runs."""
+from je_auto_control.utils.baggage.baggage import (
+    Baggage, extract_baggage, format_baggage, inject_baggage, parse_baggage,
+)
+
+__all__ = [
+    "Baggage", "extract_baggage", "format_baggage", "inject_baggage",
+    "parse_baggage",
+]

je_auto_control/utils/baggage/baggage.py

@@ -0,0 +1,102 @@
+"""W3C Baggage propagation (cross-cutting key-value context).
+
+The just-shipped ``trace_context`` carries trace/span identity across an HTTP
+boundary, but there was no way to propagate cross-cutting key-value context
+(``run_id`` / ``tenant`` / ``experiment``) alongside it. This implements the
+W3C Baggage header — a percent-encoded ``key=value`` list — so a run can attach
+such context to outgoing requests and read it back on the other side.
+
+Pure standard library (``urllib.parse``); imports no ``PySide6``. ``Baggage`` is
+immutable (mutators return a new instance), so propagation is deterministic in
+CI.
+"""
+from typing import Dict, List, Mapping, Optional, Tuple
+from urllib.parse import quote, unquote
+
+_MAX_ENTRIES = 180
+
+
+class Baggage:
+    """An immutable set of W3C baggage key-value entries."""
+
+    def __init__(self, items: Optional[Mapping[str, str]] = None) -> None:
+        self._items: Dict[str, str] = {str(key): str(value)
+                                       for key, value in (items or {}).items()}
+
+    def to_dict(self) -> Dict[str, str]:
+        """Return the entries as a plain dict."""
+        return dict(self._items)
+
+    def get(self, key: str, default: Optional[str] = None) -> Optional[str]:
+        """Return the value for ``key`` or ``default``."""
+        return self._items.get(key, default)
+
+    def set(self, key: str, value: str) -> "Baggage":
+        """Return a new baggage with ``key`` set to ``value``."""
+        updated = dict(self._items)
+        updated[str(key)] = str(value)
+        return Baggage(updated)
+
+    def remove(self, key: str) -> "Baggage":
+        """Return a new baggage with ``key`` removed."""
+        updated = dict(self._items)
+        updated.pop(key, None)
+        return Baggage(updated)
+
+    def __len__(self) -> int:
+        return len(self._items)
+
+    def __contains__(self, key: object) -> bool:
+        return key in self._items
+
+    def __eq__(self, other: object) -> bool:
+        return isinstance(other, Baggage) and other._items == self._items
+
+    def __repr__(self) -> str:
+        return f"Baggage({self._items!r})"
+
+
+def _parse_member(member: str) -> Optional[Tuple[str, str]]:
+    pair = member.split(";", 1)[0].strip()    # drop optional ;metadata
+    if not pair:
+        return None
+    key, sep, value = pair.partition("=")
+    if not sep or not key.strip():
+        return None
+    return unquote(key.strip()), unquote(value.strip())
+
+
+def parse_baggage(header: Optional[str]) -> Baggage:
+    """Parse a W3C ``baggage`` header into a :class:`Baggage`."""
+    items: Dict[str, str] = {}
+    for member in (header or "").split(","):
+        parsed = _parse_member(member)
+        if parsed is not None:
+            items[parsed[0]] = parsed[1]
+    return Baggage(items)
+
+
+def format_baggage(baggage: Baggage) -> str:
+    """Serialise a :class:`Baggage` into a percent-encoded header value."""
+    parts: List[str] = []
+    for key, value in list(baggage.to_dict().items())[:_MAX_ENTRIES]:
+        parts.append(f"{quote(key, safe='')}={quote(value, safe='')}")
+    return ",".join(parts)
+
+
+def inject_baggage(headers: Optional[Dict[str, str]],
+                   baggage: Baggage) -> Dict[str, str]:
+    """Return ``headers`` with the ``baggage`` header set (when non-empty)."""
+    out = dict(headers or {})
+    encoded = format_baggage(baggage)
+    if encoded:
+        out["baggage"] = encoded
+    return out
+
+
+def extract_baggage(headers: Optional[Dict[str, str]]) -> Baggage:
+    """Extract a :class:`Baggage` from request headers (case-insensitive)."""
+    for key, value in (headers or {}).items():
+        if str(key).lower() == "baggage":
+            return parse_baggage(value)
+    return Baggage()

je_auto_control/utils/executor/action_executor.py

@@ -3000,6 +3000,21 @@ def _trace_extract(headers: Any) -> Dict[str, Any]:
     return {"context": ctx.to_dict() if ctx is not None else None}
 
 
+def _baggage_parse(header: str) -> Dict[str, Any]:
+    """Adapter: parse a W3C baggage header into {items}."""
+    from je_auto_control.utils.baggage import parse_baggage
+    return {"items": parse_baggage(header).to_dict()}
+
+
+def _baggage_format(items: Any) -> Dict[str, Any]:
+    """Adapter: serialise an items dict into a W3C baggage {header}."""
+    import json
+    from je_auto_control.utils.baggage import Baggage, format_baggage
+    if isinstance(items, str):
+        items = json.loads(items)
+    return {"header": format_baggage(Baggage(items))}
+
+
 def _profile_rows(rows: Any, columns: Any = None) -> Dict[str, Any]:
     """Adapter: profile a row-set into per-column statistics."""
     import json
@@ -4203,6 +4218,8 @@ def __init__(self):
             "AC_http_replay": _http_replay,
             "AC_trace_inject": _trace_inject,
             "AC_trace_extract": _trace_extract,
+            "AC_baggage_parse": _baggage_parse,
+            "AC_baggage_format": _baggage_format,
             "AC_profile_rows": _profile_rows,
             "AC_infer_schema": _infer_schema,
             "AC_parse_problem": _parse_problem,

je_auto_control/utils/mcp_server/tools/_factories.py

@@ -3509,6 +3509,27 @@ def data_profile_tools() -> List[MCPTool]:
     ]
 
 
+def baggage_tools() -> List[MCPTool]:
+    return [
+        MCPTool(
+            name="ac_baggage_parse",
+            description=("Parse a W3C 'baggage' header (percent-encoded "
+                         "key=value list, optional ;metadata) into {items}."),
+            input_schema=schema({"header": {"type": "string"}}, ["header"]),
+            handler=h.baggage_parse,
+            annotations=READ_ONLY,
+        ),
+        MCPTool(
+            name="ac_baggage_format",
+            description=("Serialise an 'items' object into a percent-encoded "
+                         "W3C baggage {header}."),
+            input_schema=schema({"items": {"type": "object"}}, ["items"]),
+            handler=h.baggage_format,
+            annotations=READ_ONLY,
+        ),
+    ]
+
+
 def trace_context_tools() -> List[MCPTool]:
     return [
         MCPTool(
@@ -5070,7 +5091,8 @@ def media_assert_tools() -> List[MCPTool]:
     search_index_tools, stats_tools, recurrence_tools, text_diff_tools,
     feature_flag_tools, provenance_tools, json_contract_tools, chaos_tools,
     slo_tools, percentiles_tools, bulkhead_tools, http_cassette_tools,
-    trace_context_tools, data_profile_tools, http_problem_tools, dotenv_tools,
+    trace_context_tools, baggage_tools,
+    data_profile_tools, http_problem_tools, dotenv_tools,
     sse_client_tools, layered_config_tools, data_drift_tools,
     dataset_diff_tools,
     saga_tools, decision_table_tools, locator_repair_tools,

je_auto_control/utils/mcp_server/tools/_handlers.py

@@ -1726,6 +1726,16 @@ def trace_extract(headers):
     return _trace_extract(headers)
 
 
+def baggage_parse(header):
+    from je_auto_control.utils.executor.action_executor import _baggage_parse
+    return _baggage_parse(header)
+
+
+def baggage_format(items):
+    from je_auto_control.utils.executor.action_executor import _baggage_format
+    return _baggage_format(items)
+
+
 def profile_rows(rows, columns=None):
     from je_auto_control.utils.executor.action_executor import _profile_rows
     return _profile_rows(rows, columns)