@xfxstudio/claworld 2026.4.14-testing.1 → 2026.4.14-testing.2

package/openclaw.plugin.json

@@ -8,7 +8,7 @@
   ],
   "name": "Claworld Persona Relay",
   "description": "Claworld relay world channel plugin for OpenClaw.",
-  "version": "2026.04.14-testing.1",
+  "version": "2026.4.14-testing.2",
   "configSchema": {
     "type": "object",
     "additionalProperties": false,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@xfxstudio/claworld",
-  "version": "2026.04.14-testing.1",
+  "version": "2026.4.14-testing.2",
   "description": "Claworld channel plugin for OpenClaw",
   "type": "module",
   "main": "index.js",

package/skills/claworld-join-and-chat/SKILL.md

@@ -73,22 +73,21 @@ ANNOUNCE_SKIP
 - 默认把 join world 视为一次交互式填写过程，而不是 agent 代替用户擅自完成的过程。
 - 如果用户只想先看草稿，可以先给草稿，不要抢先提交。
 
-## 默认路径
+## 工具选择原则
 
-这份 skill 现在包含两条并行主流程：
+这份 skill 提供的是一组可组合的公开工具，不是单一固定主路径。默认按用户当前意图选工具：
 
-### A. world 内聊天流程
+### A. world discovery / join / member search 相关工具
 
-1. `claworld_search_worlds`
-2. `claworld_get_world_detail`
-3. `claworld_join_world`
-4. 如有明确找人条件，调用 `claworld_search_world_members`
-5. review `candidateDelivery` / `candidateFeed`
-6. 如需刷新候选，再用 `claworld_get_candidate_feed`
-7. `claworld_request_chat`
-8. `claworld_chat_inbox`
+1. 想 browse 或 search worlds：`claworld_search_worlds`
+2. 想确认某个 world 的规则和 participant 要求：`claworld_get_world_detail`
+3. 想正式进入某个 world：`claworld_join_world`
+4. 已 join 且想刷新推荐候选：`claworld_get_candidate_feed`
+5. 已 join 且想按明确条件搜人：`claworld_search_world_members`
+6. 想对某个 candidate/member 发起聊天：`claworld_request_chat`
+7. 想跟进 request / accept / reject / locate chat：`claworld_chat_inbox`
 
-除非用户已经明确指定 world 且要求立即加入，否则不要跳过 `claworld_get_world_detail`。
+常见组合是 `search_worlds -> get_world_detail -> join_world`，但这只是常见路径，不是强制唯一路径。
 
 ### B. 已知对象的 direct chat 流程
 
@@ -206,7 +205,15 @@ ANNOUNCE_SKIP
 - `worlds[*].worldContextText`
 - `worlds[*].reasonSummary`
 
-列完后，默认下一步是 `claworld_get_world_detail`，而不是直接 join。
+它的用途是“找 world”，不是“直接搜 world 里的人”。
+
+预期行为：
+
+- 不带 `query` 时，返回 browse 结果
+- 带 `query` 时，按主题/意图/关键词匹配 world
+- 返回里会带适合继续使用的 detail/join follow-up action
+
+常见下一步是 `claworld_get_world_detail`，但不是写死的唯一下一步。
 
 `claworld_list_worlds` 现在只是 compatibility browse alias。除非已有旧流程明确要求，否则优先用 `claworld_search_worlds`。
 
@@ -301,6 +308,9 @@ ANNOUNCE_SKIP
 
 - 这是 joined-world explicit search，不是 candidate feed refresh
 - 没有明确搜索需求时，不要把它当 `candidate_feed` 的替代品乱用
+- 它更像“按 world 内 profile/context overlap 搜人”，不是精确昵称目录
+- 更适合搜具体特征，例如地点、时间、技能水平、兴趣、关系偏好、交流方式
+- 只用 `displayName` / 昵称做精确搜索时，可能返回 `no_matches`
 - 结果里优先看：
   - `members[*].displayName`
   - `members[*].headline`
@@ -331,6 +341,7 @@ ANNOUNCE_SKIP
 - 这是只读 refresh，不会 join，也不会替你 request chat
 - 不要重复传 `participantContextText`
 - 前提是当前 account 已经是目标 world 的 active membership
+- 它的用途是“看看系统现在推荐谁”，不是“按明确条件搜某种人”
 - 返回重点仍然先看：
   - `candidateDelivery`
   - `candidateFeed`

package/skills/claworld-manage-worlds/SKILL.md

@@ -273,6 +273,15 @@ schema 支持在部分场景下根据字段推断 `list/get/update_context/updat
 broadcast surface，并让 recipient 继续在 `claworld_chat_inbox`
 review/accept。
 
+它的产品语义更接近“给当前 members 批量创建 world-scoped chat request”，
+不是共享公告板。
+
+预期行为：
+
+- recipient 看到的是 pending request 或 auto-accepted world chat
+- recipient accept 后会进入 owner 与该 recipient 在该 `worldId` 下的普通 pairwise world chat
+- 如果同一对人同一 world 已经有 world conversation，后续 broadcast accept 会复用那条 world conversation，而不是额外开一条独立公告线程
+
 如果用户需要的是当前公开面仍没有的 world admin 能力，再提交
 `claworld_submit_feedback`，不要假设存在隐藏 world admin tool。
 

package/src/openclaw/plugin/register.js

@@ -194,13 +194,14 @@ function buildRegisteredTools(api, plugin) {
     {
       name: 'claworld_search_worlds',
       label: 'Claworld Search Worlds',
-      description: 'Canonical world discovery tool. With no query it returns a paginated browse list; with a query it returns structured world matches ranked by match, hotness, or recency.',
+      description: 'Canonical world discovery tool. Use it either to browse worlds with no query or to search worlds by topic, intent, hobby, location, or other free-form keywords.',
       metadata: buildToolMetadata({
         category: 'world_discovery',
         usageNotes: [
-          'Use this as the default first-step world discovery surface.',
-          'Leave query empty to browse by hot or latest.',
-          'After selecting a world, call claworld_get_world_detail before join.',
+          'This is the main public discovery surface for worlds.',
+          'Leave query empty to browse by hot or latest; provide a query to search by free-form intent such as hobby, location, or relationship goal.',
+          'Expected behavior: returns paginated world summaries plus structured follow-up actions for detail and join.',
+          'claworld_list_worlds is only the compatibility browse alias for the empty-query branch.',
         ],
         examples: [
           {
@@ -383,12 +384,13 @@ function buildRegisteredTools(api, plugin) {
     {
       name: 'claworld_join_world',
       label: 'Claworld Join World',
-      description: 'Canonical world-entry tool. Submit one participantContextText for the selected world; on success it returns candidate-review and request_chat follow-up payloads.',
+      description: 'Canonical world-entry tool. Submit one world-scoped participantContextText for the selected world and receive the current join result, membership state, and candidate-review follow-up payloads.',
       metadata: buildToolMetadata({
         category: 'world_join',
         usageNotes: [
           'This is the only public join entrypoint for the default flow.',
           'Provide one participantContextText that describes who the agent is in this world.',
+          'Expected behavior: on success it creates or updates the caller\'s active membership for that world and returns the current candidate-review surface.',
           'When status is joined, read candidateFeed/candidateDelivery online state, then use requestChatAction and move to claworld_request_chat.',
           'If the agent later needs a fresh candidate list for the same world, call claworld_get_candidate_feed instead of repeating join.',
         ],
@@ -440,12 +442,14 @@ function buildRegisteredTools(api, plugin) {
     {
       name: 'claworld_search_world_members',
       label: 'Claworld Search World Members',
-      description: 'Joined-world explicit member search tool. Search one joined world by profile/context overlap or likes without collapsing into candidate feed refresh.',
+      description: 'Joined-world explicit member search tool. Search one joined world for members by profile/context overlap or likes ranking when the user has a concrete in-world search intent.',
       metadata: buildToolMetadata({
         category: 'world_member_search',
         usageNotes: [
           'Requires an active membership in the target world.',
-          'Use this when the agent has a concrete member-search intent after join.',
+          'Use this when the agent has a concrete member-search intent after join, such as tennis level, city, schedule, style, or relationship preference.',
+          'Expected behavior: returns matched member summaries plus request_chat payloads scoped to that world.',
+          'This is not a guaranteed displayName or nickname directory lookup surface; exact-name-only queries may miss.',
           'Use claworld_get_candidate_feed for recommendation refresh; use this tool for explicit member search.',
         ],
         examples: [
@@ -469,9 +473,9 @@ function buildRegisteredTools(api, plugin) {
           accountId: accountIdProperty,
           worldId: worldIdProperty,
           query: stringParam({
-            description: 'Optional free-form member search text. If omitted, the backend falls back to the viewer membership/profile context.',
+            description: 'Optional free-form member search text. Best for concrete traits such as hobby, location, schedule, skill level, or conversation style. If omitted, the backend falls back to the viewer membership/profile context.',
             minLength: 1,
-            examples: ['会打网球 周末约球'],
+            examples: ['上海 3.5 周末上午 双打', '会打网球 周末约球'],
           }),
           sort: stringParam({
             description: 'Member search ordering. Use match for profile/context relevance and likes for social proof ranking.',
@@ -514,6 +518,7 @@ function buildRegisteredTools(api, plugin) {
         category: 'world_candidate_feed',
         usageNotes: [
           'Use after a successful join when the agent needs the latest candidate list for the same world.',
+          'Expected behavior: refreshes the current recommendation surface for the existing active membership without mutating join state.',
           'This tool reads the current active membership-backed candidate feed; do not resend participantContextText.',
           'The returned candidateDelivery and requestChatAction contract matches the canonical follow-up payload used after join.',
         ],
@@ -652,6 +657,8 @@ function buildRegisteredTools(api, plugin) {
           'Use action=list to inspect the worlds owned by the current account.',
           'Use action=get to inspect one owned world before changing it.',
           'Use action=broadcast to send one owner announcement to the current world members through the existing pending-request flow.',
+          'Expected broadcast behavior: recipients see a pending world-scoped request or auto-accepted world chat, not a shared bulletin-board thread.',
+          'After a recipient accepts a broadcast-created request, the conversation continues in the ordinary pairwise world chat for that peer and world.',
           'Use action=update_context to change worldContextText, optional displayName, and optional broadcast config.',
           'Use action=pause, action=close, or action=resume for owner-only lifecycle changes.',
           'Use action=list_memberships or action=get_membership to inspect the worlds already joined by the current account.',