diff --git a/.claude-plugin/marketplace.json b/.claude-plugin/marketplace.json
index 0efef0f..827fcf0 100644
--- a/.claude-plugin/marketplace.json
+++ b/.claude-plugin/marketplace.json
@@ -58,7 +58,7 @@
     {
       "name": "ragkit",
       "source": "./plugins/ragkit",
-      "version": "0.1.2",
+      "version": "0.1.3",
       "description": "Standalone knowledge-base RAG plugin (vector/lexical/metadata multi-channel recall).",
       "category": "database",
       "homepage": "https://github.com/qxbyte/pluginhub",
diff --git a/plugins/ragkit/.claude-plugin/plugin.json b/plugins/ragkit/.claude-plugin/plugin.json
index 37f8290..e1ab33f 100644
--- a/plugins/ragkit/.claude-plugin/plugin.json
+++ b/plugins/ragkit/.claude-plugin/plugin.json
@@ -1,6 +1,6 @@
 {
   "name": "ragkit",
-  "version": "0.1.2",
+  "version": "0.1.3",
   "description": "Standalone knowledge-base RAG: chunk + embed + multi-channel recall (vector/lexical/metadata, RRF-fused) over distill-style knowledge-base/ directories.",
   "author": {
     "name": "xueqiang",
diff --git a/plugins/ragkit/CHANGELOG.md b/plugins/ragkit/CHANGELOG.md
index 4e3e7b2..91332d4 100644
--- a/plugins/ragkit/CHANGELOG.md
+++ b/plugins/ragkit/CHANGELOG.md
@@ -2,6 +2,11 @@
 
 ## Unreleased
 
+## 0.1.3 (2026-07-08) — 修 skill YAML + 临时探针 spike
+
+- **修 frontmatter YAML 解析失败(P0)**：0.1.2 的四个 skill `description` 含 `Trigger: `（冒号+空格），未加引号导致 YAML 解析失败——`claude plugin validate` 报错、且**运行时 frontmatter 被整体静默丢弃**（name/description 全丢，影响 skill 发现）。修法：四个 description 全部加双引号。（同款 bug 也在 specode / task-swarm 的 skill 里，单独处理。）
+- **临时探针 spike**：`skills/query/SKILL.md` 加 `__env_probe__` 哨兵——检索词为 `__env_probe__` 时不查询、改跑自包含 bash,打印各 harness 在 **skill 上下文**里提供的 `CLAUDE_SKILL_DIR` / `CLAUDE_PLUGIN_ROOT` / `CODEBUDDY_*` 变量、路径形态、及候选 resolver 是否命中。用于定死"harness 无关的 skill 脚本定位"最终写法(去掉 `${VAR:-default}` + `find`，消除 Windows msys 路径 bug)。正常检索不受影响,**下一版删探针、落地正式 resolver**。
+
 ## 0.1.2 (2026-07-06) — skill description 统一模板
 
 - 四个 skill（embed / query / status / eval）的 frontmatter `description`（渐进式加载唯一常驻的元数据）统一为「Use when 场景 → 做什么 → Trigger 触发词/命令 → 边界」轻量模板：补上 `/ragkit:*` 触发命令与彼此的交叉引用边界（query↔embed↔status）。零行为变化。
diff --git a/plugins/ragkit/skills/embed/SKILL.md b/plugins/ragkit/skills/embed/SKILL.md
index c5eff42..960b9bb 100644
--- a/plugins/ragkit/skills/embed/SKILL.md
+++ b/plugins/ragkit/skills/embed/SKILL.md
@@ -1,7 +1,7 @@
 ---
 name: embed
 user-invocable: false
-description: Use when building or refreshing the RagKit index for a project `knowledge-base/` (e.g. after distill added/updated knowledge points). Trigger: /ragkit:embed. Querying instead → ragkit:query; index health → ragkit:status.
+description: "Use when building or refreshing the RagKit index for a project `knowledge-base/` (e.g. after distill added/updated knowledge points). Trigger: /ragkit:embed. Querying instead → ragkit:query; index health → ragkit:status."
 ---
 
 # RagKit Embed
diff --git a/plugins/ragkit/skills/eval/SKILL.md b/plugins/ragkit/skills/eval/SKILL.md
index 91ae71e..ac80632 100644
--- a/plugins/ragkit/skills/eval/SKILL.md
+++ b/plugins/ragkit/skills/eval/SKILL.md
@@ -1,7 +1,7 @@
 ---
 name: eval
 user-invocable: false
-description: Use when measuring or tuning RagKit retrieval accuracy — runs the golden evalset, reports recall@k / MRR per bucket. Trigger: /ragkit:eval. Run before/after any retrieval-param change to keep a baseline.
+description: "Use when measuring or tuning RagKit retrieval accuracy — runs the golden evalset, reports recall@k / MRR per bucket. Trigger: /ragkit:eval. Run before/after any retrieval-param change to keep a baseline."
 ---
 
 # RagKit Eval
diff --git a/plugins/ragkit/skills/query/SKILL.md b/plugins/ragkit/skills/query/SKILL.md
index 7ca8406..41b6cd6 100644
--- a/plugins/ragkit/skills/query/SKILL.md
+++ b/plugins/ragkit/skills/query/SKILL.md
@@ -1,13 +1,37 @@
 ---
 name: query
 user-invocable: false
-description: Use when retrieving from a project `knowledge-base/` — multi-channel RAG recall (vector + lexical + metadata, RRF-fused) returning locate-pointer cards; multi-round/multi-angle querying allowed. Trigger: /ragkit:query, or called by name by workflows like specode. Building/refreshing the index instead → ragkit:embed.
+description: "Use when retrieving from a project `knowledge-base/` — multi-channel RAG recall (vector + lexical + metadata, RRF-fused) returning locate-pointer cards; multi-round/multi-angle querying allowed. Trigger: /ragkit:query, or called by name by workflows like specode. Building/refreshing the index instead → ragkit:embed."
 ---
 
 # RagKit Query
 
 对项目 `knowledge-base/` 做多路召回（向量+词汇+元数据，RRF 融合），返回**定位卡片**。
 
+## 环境探针（临时 spike 诊断，仅本版 0.1.3）
+
+**如果检索词恰好是 `__env_probe__`**：不要查询、不要调 `ragkit.py`，改为运行下面这段自包含 bash，把它的输出**原样、完整**回贴给用户，然后结束。（用途：确定当前 harness 在 skill 上下文里到底提供哪个变量、路径是什么形态，据此定死 resolver 的最终写法。请在两台机器各跑一次——Claude Code 与 CodeBuddy——对照输出。）
+
+```sh
+echo "=== RAGKIT HARNESS PROBE ==="
+echo "CLAUDE_SKILL_DIR      =[${CLAUDE_SKILL_DIR}]"
+echo "CODEBUDDY_SKILL_DIR   =[${CODEBUDDY_SKILL_DIR}]"
+echo "CLAUDE_PLUGIN_ROOT    =[${CLAUDE_PLUGIN_ROOT}]"
+echo "CODEBUDDY_PLUGIN_ROOT =[${CODEBUDDY_PLUGIN_ROOT}]"
+echo "combined_dash         =[${CLAUDE_PLUGIN_ROOT:-$CODEBUDDY_PLUGIN_ROOT}]"
+echo "cwd                   =$(pwd)"
+echo "uname                 =$(uname -s 2>/dev/null || echo n/a)"
+A="${CLAUDE_SKILL_DIR}/../../scripts/run.sh"
+B="${CLAUDE_PLUGIN_ROOT}/scripts/run.sh"
+C="${CODEBUDDY_PLUGIN_ROOT}/scripts/run.sh"
+echo "A skilldir/../../scripts/run.sh exists=$([ -f "$A" ] && echo YES || echo NO) [$A]"
+echo "B pluginroot/scripts/run.sh     exists=$([ -f "$B" ] && echo YES || echo NO) [$B]"
+echo "C cbroot/scripts/run.sh         exists=$([ -f "$C" ] && echo YES || echo NO) [$C]"
+echo "=== END PROBE ==="
+```
+
+否则（正常检索词）按下面执行。
+
 ## 执行
 
 ```sh
diff --git a/plugins/ragkit/skills/status/SKILL.md b/plugins/ragkit/skills/status/SKILL.md
index 2ee0b7a..93aa9f2 100644
--- a/plugins/ragkit/skills/status/SKILL.md
+++ b/plugins/ragkit/skills/status/SKILL.md
@@ -1,7 +1,7 @@
 ---
 name: status
 user-invocable: false
-description: Use when checking RagKit index health — doc/chunk counts, backend resolution, staleness and drift between `knowledge-base/` and index. Trigger: /ragkit:status. Rebuilding after drift → ragkit:embed.
+description: "Use when checking RagKit index health — doc/chunk counts, backend resolution, staleness and drift between `knowledge-base/` and index. Trigger: /ragkit:status. Rebuilding after drift → ragkit:embed."
 ---
 
 # RagKit Status