fix: 修复上游健康巡检并同步已审查的市场引用 (#207)

Upstream health checks could select a different branch ending in `main`,
index internal skills or lowercase reference documents, treat a valid
pinned path as broken after an upstream move, and close the tracking
issue while unpinned or unknown entries remained. Resolve exact refs and
tracked filenames, compare the installed snapshot, and drive issue
decisions from structured results.

Refresh 26 reviewed pointers, pin DingTalk and Matt Pocock collections,
remove the deleted merge-conflict skill, and add the six new public
children with reviewed translations. Synchronize sidecars, observed
release facts, license evidence and external dependency disclosures.
Identical-tree Agent pins and their versions remain unchanged. The
per-entry review is recorded in
`docs/upstream-health-189-remediation.md`.

Validation: 99 unit tests pass; all nine pinned collections reproduce;
catalog/i18n and translation-freshness checks pass; online health checks
all 41 external entries with zero broken, actionable or unknown results.
Existing catalog warnings decrease from 123 to 105 with no new warning
groups. Python/workflow shell syntax and whitespace checks pass.

Closes #189.

skills/khazix-skills/entry.json

@@ -12,11 +12,11 @@
   "i18n": {
     "zh-CN": {
       "name": "卡兹克 Skills",
-      "shortDesc": "卡兹克日常在用的开源 AI Skill 合集，遵循 AgentSkills 标准，适配 Claude Code、Codex、OpenCode 等"
+      "shortDesc": "卡兹克日常在用的开源 AI Skill 合集，遵循 AgentSkills 标准，适配 Claude Code、Codex、OpenCode 等。AIHOT 资讯依赖第三方服务；面向外部的商业使用须另获 AIHOT 授权。"
     },
     "en-US": {
       "name": "Khazix Skills",
-      "shortDesc": "Khazix's open-source collection of daily-use AI skills following the AgentSkills standard, working with Claude Code, Codex and OpenCode"
+      "shortDesc": "Khazix's open-source collection of daily-use AI skills following the AgentSkills standard, working with Claude Code, Codex and OpenCode. AIHOT news depends on a third-party service; external commercial use requires separate AIHOT authorization."
     }
   },
   "maintainer": {
@@ -32,7 +32,7 @@
     "kind": "git",
     "repoUrl": "https://github.com/KKKKhazix/khazix-skills.git",
     "repoBranch": "main",
-    "ref": "d4e43c91f16dcd859748c1d71ec7d8aa1ebb4694"
+    "ref": "322346ded8129436b3f64707789a73e732ae24d9"
   },
   "children": [
     {
@@ -40,10 +40,10 @@
       "path": "aihot",
       "i18n": {
         "zh-CN": {
-          "shortDesc": "AI HOT (aihot.virxact.com) 中文 AI 资讯查询 Skill。当用户想知道\"今天 AI 圈有什么\"、\"AI 日报\"、\"AI HOT\"、\"AI 资讯\"、\"AI 热点\"、\"最近 AI\"、\"OpenAI/Anthropic/Google 最近发布了什么\"、\"AI hot today\"、\"AI new…"
+          "shortDesc": "查询 AIHOT（aihot.news）的中文 AI 资讯。用户问今天或最近 AI 圈发生了什么、AI 新闻和 AI 日报、大模型与 AI 产品发布、OpenAI／Anthropic／Google 等公司或某个 AI 话题的最新消息、当前最热的 AI 事件及来龙去脉、AI 模型排行、Codex 额度重置，或其它 AI…"
         },
         "en-US": {
-          "shortDesc": "AI HOT (aihot.virxact.com) 中文 AI 资讯查询 Skill。当用户想知道\"今天 AI 圈有什么\"、\"AI 日报\"、\"AI HOT\"、\"AI 资讯\"、\"AI 热点\"、\"最近 AI\"、\"OpenAI/Anthropic/Google 最近发布了什么\"、\"AI hot today\"、\"AI new…"
+          "shortDesc": "查询 AIHOT（aihot.news）的中文 AI 资讯。用户问今天或最近 AI 圈发生了什么、AI 新闻和 AI 日报、大模型与 AI 产品发布、OpenAI／Anthropic／Google 等公司或某个 AI 话题的最新消息、当前最热的 AI 事件及来龙去脉、AI 模型排行、Codex 额度重置，或其它 AI…"
         }
       }
     },
@@ -71,15 +71,27 @@
         }
       }
     },
+    {
+      "id": "leader",
+      "path": "leader",
+      "i18n": {
+        "zh-CN": {
+          "shortDesc": "将想法整理成可执行的 Agent 任务书，包含调研证据、范围、验收要求和恢复安排。"
+        },
+        "en-US": {
+          "shortDesc": "Turn an idea into an executable agent brief with research evidence, scope, acceptance criteria and recovery steps."
+        }
+      }
+    },
     {
       "id": "neat-freak",
       "path": "neat-freak",
       "i18n": {
         "zh-CN": {
-          "shortDesc": "End-of-session knowledge cleanup with OCD-level rigor — reconciles project docs (CLAUDE.md, README.md, docs/) and agent memory against the code, and audits whet…"
+          "shortDesc": "Knowledge and governance closeout: reconcile project docs, rule files (CLAUDE.md/AGENTS.md), authorized agent memory, and workspace residue with what the code a…"
         },
         "en-US": {
-          "shortDesc": "End-of-session knowledge cleanup with OCD-level rigor — reconciles project docs (CLAUDE.md, README.md, docs/) and agent memory against the code, and audits whet…"
+          "shortDesc": "Knowledge and governance closeout: reconcile project docs, rule files (CLAUDE.md/AGENTS.md), authorized agent memory, and workspace residue with what the code a…"
         }
       }
     },