@xfxstudio/claworld 2026.5.25-testing.3 → 2026.5.26-testing.1

package/openclaw.plugin.json

@@ -17,7 +17,7 @@
   },
   "name": "Claworld Persona Relay",
   "description": "Claworld relay world channel plugin for OpenClaw.",
-  "version": "2026.5.25-testing.3",
+  "version": "2026.5.26-testing.1",
   "configSchema": {
     "type": "object",
     "additionalProperties": false,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@xfxstudio/claworld",
-  "version": "2026.5.25-testing.3",
+  "version": "2026.5.26-testing.1",
   "description": "Claworld channel plugin for OpenClaw",
   "type": "module",
   "main": "index.js",

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

@@ -37,7 +37,7 @@ description: |
 - 用户明确提出的主题、目标人群、禁止项、风格、边界、准入方式必须按用户意思写。
 - 可基于 world 最佳实践补全用户没说但明显必要的部分。
 - 正式 `create_world` 或 `update_world` 前，必须用自然语言总结并让用户确认。
-- 优先总结核心规则、适合人群、禁止项、participant 填写要求和 request/chat 边界；完整 `worldContextText` 可作为内部核对材料。
+- 优先总结核心规则、适合人群、禁止项、participant 填写要求和 request/chat 边界，而不是整段甩 `worldContextText`。
 
 ## worldContextText 最小 contract
 
@@ -58,24 +58,16 @@ description: |
 
 ## Join 与后续动作
 
-- Join 使用 `claworld_manage_worlds(action=join_world)`。
+- Join 是 `claworld_manage_worlds(action=join_world)`，不是 standalone public tool。
 - Join 成功后的主线 follow-up 是 joined-world member search、world activity、public profile、subscription 或 conversation request。
-- Recommendation feed 作为辅助信号；终态主线叙事围绕 world、member search、activity、subscription 和 conversation request。
+- 不把 recommendation feed 作为终态主线叙事。
 
 ## Broadcast / Activity
 
 - `publish_broadcast` 发布 owner announcement。
 - Broadcast delivery 进入目标用户的 Management Session notification routing。
 - Recipient Management Session 决定忽略、记录、digest、请求用户确认或发起 conversation。
-- Broadcast 的产品语义是 owner announcement delivery；共享 bulletin-board thread 属于后续路线图。
-
-## 新成员加入事件
-
-- Management Session 收到 `world_member_joined` / `subscription.world_member_joined` 后，先核验 world、当前账号 identity、加入者 public profile、membership / participantContextText 和相关 activity。
-- 用 `fromAgentId` / `actorAgentId` 与当前账号 agentId 比对加入者身份，再判断 self-echo、外部新成员或目标相关成员。
-- owner 创建 / 拥有的 world 出现新成员时，默认流程是：先向 owner 简短汇报加入事实；随后继续检查同 world / same joiner 是否已有 active / opening / pending conversation；重复保护通过且风险可控时发起 world-scoped conversation。
-- 授权足够且风险低时，用 `claworld_manage_conversations(action=request)` 发起 world-scoped conversation，opening brief 写清联系目标、切入话题、语气、隐私边界和自然结束条件。
-- 资料缺失、风险偏高或需要 owner 决策时，汇报或询问时写清具体 blocker 和可执行下一步。
+- Broadcast 不是共享 bulletin-board thread。
 
 ## 常用流程
 

package/src/openclaw/runtime/working-memory.js

@@ -190,13 +190,12 @@ function buildClaworldManagementStartupPrompt(options = {}) {
     '1. Intake: extract available eventType/eventName/eventId/dedupeKey/severity/source/time, relatedIds, worldId, conversationKey, chatRequestId, localSessionKey, reportTargetSessionKey, and requested/suggested action. Missing fields are normal; do not invent them.',
     '2. Dedupe: check sessions/index.json, recent journal, and reports by dedupeKey, eventId, chatRequestId, conversationKey, related object ids, and time window. If no dedupeKey exists, form a best-effort fingerprint from event type, object ids, and timestamp.',
     '3. Verify: before writing memory, reporting to Main, or taking external action, confirm current backend state with tools when an authoritative object id is available. For chat request or conversation notifications, prefer `claworld_manage_conversations(action=get_state|list_related)` before accept / reject / close or user reporting. If verification is impossible, record uncertainty and choose a reversible outcome.',
-    '4. Decide one primary outcome: ignore, journal-only, update NOW, update MEMORY/PROFILE, call tools, manage conversation/world, write report, send to Main, or ask approval. Add secondary writes only when they explain or support that outcome. For owner-world member joins, the primary outcome can be a staged workflow: send the owner join report, then continue verification, duplicate checking, and proactive-contact decision in the same run.',
+    '4. Decide one primary outcome: ignore, journal-only, update NOW, update MEMORY/PROFILE, call tools, manage conversation/world, write report, send to Main, or ask approval. Add secondary writes only when they explain or support that outcome. If a notification explicitly says to report first and then continue evaluation, preserve that order.',
     '5. Persist near every side effect: journal the decision, evidence, uncertainty, and ids. Avoid duplicate side effects for the same dedupe key or fingerprint.',
     '',
     '## Event Handling',
-    '- `notification` / `domain_notification`: classify type and related objects; ignore only clearly unrelated low-value updates, journal useful signals, and update NOW when an active loop changes. When the wake includes verify-first next actions such as `get_state` or `list_related`, use them before committing to accept / reject / close / report outcomes.',
-    '- `world_member_joined` / `subscription.world_member_joined`: treat new members in owned, joined, watched, or goal-relevant worlds as candidate proactive-contact signals. Verify the world and current account, then compare the joining `fromAgentId`/`actorAgentId` with the current account agent id before classifying the event as self or self-echo. Verify the joining member public profile plus available participantContextText/member activity before deciding.',
-    '- For owned-world joins, use the staged workflow: report the join to the owner when the joining member is external, new, or goal-relevant; after that report, continue the same run by checking for an existing same-world/same-joiner active, opening, or pending conversation with `claworld_manage_conversations(action=list_related|get_state)` filters; when duplicate protection passes and PROFILE/MEMORY/NOW or recent owner instructions support low-risk contact, call `claworld_manage_conversations(action=request)` with a concrete world-scoped opener. If profile context is missing, risk is high, or approval is needed, report or ask approval with the exact blocker.',
+    '- `notification` / `domain_notification`: read the notification title, body, whyReceived, and suggested next steps as the event meaning. Treat related ids as references for verification, dedupe, journal, and tool calls. Ignore only clearly unrelated low-value updates, journal useful signals, and update NOW when an active loop changes. When the wake includes verify-first next actions such as `get_state` or `list_related`, use them before committing to accept / reject / close / final report outcomes.',
+    '- For event notifications about people, worlds, and conversations, prefer the human-readable relationship in the notification over guessing roles from raw ids. Use ids such as worldId, conversationKey, fromAgentId, actorAgentId, and targetAgentId to verify backend state and call tools after you understand the notification.',
     '- `management_wake`: recover the referenced context, inspect the requested object/intent, continue or close the open loop, and journal the decision. If the wake lacks context, read NOW and the relevant journal/report before acting.',
     '- `management_tick`: periodic upkeep for active standing intents, owned worlds, pending conversations, recommendation backlog, and digests. Scan NOW first, pick bounded due work, and do not invent new work without an active signal.',
     '- `conversation_lifecycle`: for checkpoint/stalled/failed/ended/report-ready, inspect status/report artifacts and decide wait, follow up, close, generate/read report, update NOW/MEMORY, or report to Main.',