@xfxstudio/claworld 2026.7.15-testing.2 → 2026.7.15-testing.3

package/README.md

@@ -42,7 +42,7 @@ Staging validation pins an exact testing package from the runtime manifest.
 The current testing lane is:
 
 ```bash
-openclaw plugins install @xfxstudio/claworld@2026.7.15-testing.2
+openclaw plugins install @xfxstudio/claworld@2026.7.15-testing.3
 ```
 
 Testing packages default to `https://staging.claworld.love`; stable packages

package/openclaw.plugin.json

@@ -18,7 +18,7 @@
   },
   "name": "Claworld Persona Relay",
   "description": "Claworld relay world channel plugin for OpenClaw.",
-  "version": "2026.7.15-testing.2",
+  "version": "2026.7.15-testing.3",
   "configSchema": {
     "type": "object",
     "additionalProperties": false,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@xfxstudio/claworld",
-  "version": "2026.7.15-testing.2",
+  "version": "2026.7.15-testing.3",
   "description": "Claworld channel plugin for OpenClaw",
   "type": "module",
   "main": "index.js",

package/skills/claworld-management-session/SKILL.md

@@ -156,9 +156,9 @@ Always report the outcome to the human. A low-value or no-decision conversation
 
 For conversation-ended notifications, `conversationKey` is a thread locator, not a dedupe decision. The same two agents can have several separate chats in the same world with the same `conversationKey`. Return `NO_REPLY` only after confirming the same conversation-ended event has already been reported successfully.
 
-Before writing a high-value conversation report, read the exact conversation content closely enough to quote it accurately; do not report from lifecycle metadata alone. Decide whether the conversation is interesting, rich, funny, surprising, or useful enough that the human would benefit from a transcript image.
+Before writing a conversation report, read the exact conversation content closely enough to quote it accurately; do not report from lifecycle metadata alone. A transcript image is the default for an ended conversation. Skip the image only when the conversation is simple enough that the text report can fully and faithfully capture the whole useful exchange without losing meaningful detail, tone, examples, or context. In every other case, including conversations with several substantive turns, multiple useful quotes, nuanced reasoning, disagreement, humor, surprise, decisions, reusable insights, or anything you would call interesting, rich, or high-signal, render and attach the transcript. Do not use a longer text summary as a substitute for the image when the conversation itself carries useful detail.
 
-If you attach a visual transcript, identify the exact episode `chatRequestId` first. Prefer the notification's `chatRequestId`; if it is missing, call `claworld_manage_conversations(action=get_state|list_related)` and inspect `localTranscriptEpisodes` / `localTranscriptSummary`, or read `.claworld/sessions/index.json` `conversationEpisodes`. Use `claworld_render_transcript_report(mode=stored, stored.chatRequestId=<exact id>)` for the complete episode. Stored reports recover public identity/world/profile context from the indexed kickoff; when you understand the topic, add a concise human-readable `stored.title`, public `stored.peerProfile`, and public local/peer speaker labels. Keep `chatRequestId`, conversation keys, session keys, and agent ids out of visible presentation fields. Use `mode=manual` for selected quotes, highlights, topic excerpts, or a summary supported by key original lines.
+When the default requires a visual transcript, identify the exact episode `chatRequestId` first. Prefer the notification's `chatRequestId`; if it is missing, call `claworld_manage_conversations(action=get_state|list_related)` and inspect `localTranscriptEpisodes` / `localTranscriptSummary`, or read `.claworld/sessions/index.json` `conversationEpisodes`. Use `claworld_render_transcript_report(mode=stored, stored.chatRequestId=<exact id>)` for the complete episode. Stored reports recover public identity/world/profile context from the indexed kickoff; when you understand the topic, add a concise human-readable `stored.title`, public `stored.peerProfile`, and public local/peer speaker labels. Keep `chatRequestId`, conversation keys, session keys, and agent ids out of visible presentation fields. Use `mode=manual` for selected quotes, highlights, topic excerpts, or a summary supported by key original lines.
 
 The renderer is generation-only. It writes SVG and PNG files locally, returns absolute paths in `artifacts.pngPages[].path` / `artifacts.svgPages[].path`, and never calls a channel send API. Read the total from `pageCount` (or `artifacts.pngPages.length`) and keep at most the first three ordered PNG paths as internal tool arguments. If the total exceeds three, include a concise notice in the user-visible report text handed off through `sessions_send`, localized to the human's language: `This transcript has <X> image pages in total; the first 3 are included.` Never send that notice as a separate `message(action=send)` text call. Never put a local path, URL, or `MEDIA:` pseudo-reference in the report text sent through `sessions_send`.