fix(cli): 兼容轻量 schema 并升级到 0.11.3

Author: sincerely90

.agents/plugins/marketplace.json

@@ -1,6 +1,6 @@
 {
   "name": "aisee-plugin",
-  "version": "0.11.2",
+  "version": "0.11.3",
   "interface": {
     "displayName": "Aisee Plugin",
     "description": "Aisee plugin marketplace for OpenSpec, project memory, team knowledge, and Compound Engineering workflows."

CHANGELOG.md

@@ -2,6 +2,14 @@
 
 本项目遵循 [Keep a Changelog](https://keepachangelog.com/en/1.1.0/) 的结构，并使用 SemVer 管理版本。
 
+## [0.11.3] - 2026-06-13
+
+### Changed
+
+- `aisee context pack` 与 `aisee:implementation-bridge` 兼容 OpenSpec 官方 `spec-driven` 和其它轻量 schema，不再因为缺少 Aisee 专属 `capabilities`、artifact `requiredness` 或 artifact `capabilities` 字段而误报 blocker。
+- `aisee context pack --for ce-work --json` 在无 `source-map.md` 的轻量 schema 下继续从当前 schema artifacts 提取实现路径；仅在真实 schema / artifact / 路径信息缺口上保留阻塞。
+- `aisee:implementation-bridge` 与 README 文案同步明确轻量 schema 兼容边界，避免把 Aisee 增强字段误当成 OpenSpec schema 的准入条件。
+
 ## [0.11.2] - 2026-06-13
 
 ### Changed

README.md

@@ -155,6 +155,7 @@ Compound Engineering = 可选的执行 / 审查 / 测试消费方
 - **Schema-aware change planning**：`aisee:change-plan` 将已确认输入映射为可独立交付的 OpenSpec changes。
 - **OpenSpec schema pack**：提供 app、device、docsite、infra、security、quick-fix、quick-research、collaboration 等 schema。
 - **Context packs**：`aisee context pack` 为实现、验证和 review 生成 JSON 上下文。
+- **轻量 schema 兼容**：`aisee context pack` 与 `aisee:implementation-bridge` 可直接消费 OpenSpec 官方 `spec-driven` 及其它轻量 schema；不会因为缺少 Aisee 专属增强字段而拒绝生成 bridge 上下文。
 - **项目记忆**：`aisee memory` 受控检索和写入当前仓库长期 guidance，不替代 OpenSpec 事实源。
 - **团队知识 Guardrails**：`aisee knowledge` 基于 pack/card 协议按需检索少量已审查工程经验，不把知识库变成第二份规范事实源。
 - **轻量上下文路由**：`aisee context pack` 在 `source-map.md` 存在时解析来源、编号、候选路径和 evidence 入口。

plugins/aisee-plugin/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "aisee-plugin",
-  "version": "0.11.2",
+  "version": "0.11.3",
   "description": "AI-Enhanced Software Engineering plugin for OpenSpec and Compound Engineering workflows.",
   "author": {
     "name": "Fengliang"

plugins/aisee-plugin/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "aisee-plugin",
-  "version": "0.11.2",
+  "version": "0.11.3",
   "description": "AI-Enhanced Software Engineering plugin for OpenSpec and Compound Engineering workflows.",
   "author": {
     "name": "Fengliang"

plugins/aisee-plugin/.cursor-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "aisee-plugin",
-  "version": "0.11.2",
+  "version": "0.11.3",
   "description": "AI-Enhanced Software Engineering plugin for OpenSpec and Compound Engineering workflows.",
   "author": {
     "name": "Fengliang"

plugins/aisee-plugin/skills/aisee-implementation-bridge/SKILL.md

@@ -69,6 +69,7 @@ Implementation Brief 是实现阶段的交接输出，不是新的规范事实
 
 - 当前只处理一个 OpenSpec change。
 - schema metadata 已声明且项目内已安装该 schema；`SCHEMA_METADATA_MISSING`、`SCHEMA_MISMATCH`、`SCHEMA_NOT_INSTALLED`、`SCHEMA_NOT_FOUND` 都是实现 blocker。
+- 不要把 Aisee 增强字段本身当成准入门槛：官方或轻量 schema 缺少顶层 `capabilities`、artifact `requiredness`、artifact `capabilities` 时，不因此阻塞实现交接；只要 OpenSpec schema 本身可解析，仍按该 schema 的实际 artifacts 和 apply tracks 继续 bridge。
 - change 已经通过 `aisee:change-author` 补齐当前 schema 的必需 artifacts；只有 schema 生成 `source-map.md` 时才要求 source-map 说明按需 artifacts 的 Required=yes/no 与 N/A 原因。
 - 当前 schema 的 apply tracks 存在；通常是 `tasks.md`，但 `quick-research` 等无 apply schema 可以没有 tasks。
 - `source-map.md` 只在当前 schema 生成它时必需。
@@ -105,6 +106,7 @@ aisee context pack --change <change> --for ce-work --json
 处理规则：
 
 - `context pack` 返回 `SCHEMA_METADATA_MISSING` / `SCHEMA_MISMATCH` / `SCHEMA_NOT_INSTALLED` / `SCHEMA_NOT_FOUND`：停止，不生成可执行 Brief。
+- 如果 `context pack` 读取的是官方或轻量 schema，且缺少 Aisee 专属增强字段，不把这类缺口升级成 blocker；继续根据 schema 已声明的 artifacts、`apply.tracks`、`source-map.md` 是否存在来判断执行入口。
 - `context pack.facts.derived.execution.requires_ce_plan=true`：Brief 中推荐先用 `ce-plan`，但结论必须回写当前 schema 的 apply tracks；只有 schema 生成 `source-map.md` 时才回写 source-map。
 - `ID_RESERVATION_REQUIRED`、`ID_UNREGISTERED_REFERENCE`、`ID_INACTIVE_REFERENCE` 只能作为阻塞或风险进入 Brief；不得在实现阶段由 `ce-work` 临时发明 ID。
 
@@ -174,6 +176,7 @@ Brief 只做执行索引，不复制 OpenSpec artifact 正文：
 - `specs/**/spec.md` 是行为契约。
 - 当前 schema 的 apply tracks 是长期执行清单；app/device/quick-fix 通常是 `tasks.md`，无 apply schema 不强制 tasks。
 - `source-map.md` 只在当前 schema 生成它时是代码定位入口；不生成 source-map 的 schema 使用 context pack 从 schema artifacts 中提取的 implementation references。
+- `spec-driven`、`quick-fix`、`quick-research` 或其它轻量 schema 只要 OpenSpec 侧有效，就允许进入 bridge；不要要求它们补 Aisee app/device schema 才有的增强字段。
 - app schema 中，按 `source-map.md` 的 Artifact Applicability 判断哪些 supporting artifacts 是 Required=yes；不要要求每个 change 都展开全部 contracts。
 - device schema 中，`design.md` 和 hardware/firmware/runtime/verification contracts 是实现 contracts；不要把 app contract 假设套到设备项目。
 - `ce-work` 的允许路径应来自 context pack 的 `allowed_paths`、`code_paths` 和 `test_paths`。缺失时标记 `[SOURCE-MAP-GAP]` 或 `[IMPLEMENTATION-PATH-GAP]`，不要自行扩大实现范围。

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "aisee-plugin"
-version = "0.11.2"
+version = "0.11.3"
 description = "AI-Enhanced Software Engineering plugin for OpenSpec and Compound Engineering workflows."
 readme = "README.md"
 requires-python = ">=3.10"

src/aisee_cli/__init__.py

@@ -2,4 +2,4 @@
 
 __all__ = ["__version__"]
 
-__version__ = "0.11.2"
+__version__ = "0.11.3"

src/aisee_cli/context_pack.py

@@ -345,10 +345,17 @@ def parse_schema(schema_path: Path) -> dict[str, Any]:
     issues: list[dict[str, str]] = []
     raw_capabilities = data.get("capabilities")
     capabilities = parse_capabilities(raw_capabilities)
-    if raw_capabilities is None:
-        issues.append(schema_issue("SCHEMA_CAPABILITIES_MISSING", "blocker", "schema.yaml must define top-level capabilities"))
-    elif not isinstance(raw_capabilities, list) or any(not isinstance(item, str) or not item.strip() for item in raw_capabilities):
-        issues.append(schema_issue("SCHEMA_CAPABILITIES_INVALID", "blocker", "schema capabilities must be a non-empty string list"))
+    if raw_capabilities is not None and (
+        not isinstance(raw_capabilities, list)
+        or any(not isinstance(item, str) or not item.strip() for item in raw_capabilities)
+    ):
+        issues.append(
+            schema_issue(
+                "SCHEMA_CAPABILITIES_INVALID",
+                "risk",
+                "schema capabilities should be a non-empty string list when declared",
+            )
+        )
 
     artifacts: list[ArtifactSpec] = []
     raw_artifacts = data.get("artifacts")
@@ -367,7 +374,7 @@ def parse_schema(schema_path: Path) -> dict[str, Any]:
     apply_tracks = normalize_track_value(apply.get("tracks"))
     archive_tracks = parse_tracks(archive.get("tracks"))
     if apply_tracks is None and "apply_execution" in capabilities:
-        issues.append(schema_issue("SCHEMA_APPLY_TRACKS_MISSING", "blocker", "schema capability apply_execution requires apply.tracks"))
+        issues.append(schema_issue("SCHEMA_APPLY_TRACKS_MISSING", "risk", "schema capability apply_execution requires apply.tracks"))
     if "archive_authority" in capabilities and not archive_tracks:
         issues.append(schema_issue("SCHEMA_ARCHIVE_TRACKS_MISSING", "risk", "schema capability archive_authority should declare archive.tracks"))
 
@@ -434,40 +441,35 @@ def normalize_track_value(value: Any) -> str | None:
 def validate_artifact_schema(data: dict[str, Any]) -> list[dict[str, str]]:
     issues: list[dict[str, str]] = []
     artifact_id = str(data.get("id") or "")
-    requiredness = str(data.get("requiredness") or "")
+    raw_requiredness = data.get("requiredness")
+    requiredness = str(raw_requiredness or "always")
     capabilities = data.get("capabilities")
     if not artifact_id:
         issues.append(schema_issue("SCHEMA_ARTIFACT_ID_MISSING", "blocker", "schema artifact is missing id"))
-    if requiredness not in REQUIREDNESS_VALUES:
+    if raw_requiredness is not None and requiredness not in REQUIREDNESS_VALUES:
         issues.append(
             schema_issue(
                 "SCHEMA_ARTIFACT_REQUIREDNESS_INVALID",
-                "blocker",
+                "risk",
                 f"artifact {artifact_id or '<unknown>'} must declare requiredness as one of: always, conditional, never",
             )
         )
-    if capabilities is None:
-        issues.append(
-            schema_issue(
-                "SCHEMA_ARTIFACT_CAPABILITIES_MISSING",
-                "blocker",
-                f"artifact {artifact_id or '<unknown>'} must declare capabilities",
-            )
-        )
-    elif not isinstance(capabilities, list) or any(not isinstance(item, str) or not item.strip() for item in capabilities):
+    if capabilities is not None and (
+        not isinstance(capabilities, list) or any(not isinstance(item, str) or not item.strip() for item in capabilities)
+    ):
         issues.append(
             schema_issue(
                 "SCHEMA_ARTIFACT_CAPABILITIES_INVALID",
-                "blocker",
+                "risk",
                 f"artifact {artifact_id or '<unknown>'} capabilities must be a string list",
             )
         )
     na_requires_reason = data.get("na_requires_reason")
-    if requiredness == "conditional" and not isinstance(na_requires_reason, bool):
+    if raw_requiredness is not None and requiredness == "conditional" and not isinstance(na_requires_reason, bool):
         issues.append(
             schema_issue(
                 "SCHEMA_ARTIFACT_NA_POLICY_MISSING",
-                "blocker",
+                "risk",
                 f"artifact {artifact_id or '<unknown>'} must declare na_requires_reason for conditional requiredness",
             )
         )
@@ -496,7 +498,16 @@ def artifact_has_capability(artifact: ArtifactSpec | dict[str, Any], capability:
 
 
 def schema_generates_source_map(schema_info: dict[str, Any]) -> bool:
-    return schema_has_capability(schema_info, "source_map_routing")
+    if schema_has_capability(schema_info, "source_map_routing"):
+        return True
+    return any(
+        isinstance(spec, ArtifactSpec)
+        and (
+            spec.artifact_id == "source-map"
+            or spec.generates == "source-map.md"
+        )
+        for spec in schema_info.get("artifacts", [])
+    )
 
 
 def schema_requires_tasks(schema_info: dict[str, Any]) -> bool: