You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: CHANGELOG.md
+1Lines changed: 1 addition & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -12,6 +12,7 @@ Docs: https://docs.openclaw.ai
12
12
13
13
### Fixes
14
14
15
+
- Sessions/WebChat: filter OpenClaw delivery-mirror transcript artifacts from LLM context, `chat.history`, and `sessions.get` so internal delivery audit rows do not appear as duplicate assistant messages. Carries forward #40716. Thanks @kiyoakii.
15
16
- NVIDIA/NIM: persist the `NVIDIA_API_KEY` provider marker and mark bundled NVIDIA Chat Completions models as string-content compatible, so NIM models load from `models.json` and OpenAI-compatible subagent calls send plain text content. Fixes #73013 and #50107; refs #73014. Thanks @bautrey, @iot2edge, @ifearghal, and @futhgar.
16
17
- Channels/Discord: let text-only configs drop the `GuildVoiceStates` gateway intent and expose a bounded `/gateway/bot` metadata timeout with rate-limited fallback logs, reducing idle CPU and warning floods. Fixes #73709 and #73585. Thanks @sanchezm86 and @trac3r00.
17
18
- CLI/plugins: use plugin metadata snapshots for install slot selection and add opt-in plugin lifecycle timing traces, so plugin install avoids runtime-loading the plugin registry for metadata-only decisions. Thanks @shakkernerd.
- `sessions.reset`, `sessions.delete`, and `sessions.compact` perform session maintenance.
401
-
- `sessions.get` returns the full stored session row.
402
-
- Chat execution still uses `chat.history`, `chat.send`, `chat.abort`, and `chat.inject`. `chat.history` is display-normalized for UI clients: inline directive tags are stripped from visible text, plain-text tool-call XML payloads (including `<tool_call>...</tool_call>`, `<function_call>...</function_call>`, `<tool_calls>...</tool_calls>`, `<function_calls>...</function_calls>`, and truncated tool-call blocks) and leaked ASCII/full-width model control tokens are stripped, pure silent-token assistant rows such as exact `NO_REPLY` / `no_reply` are omitted, and oversized rows can be replaced with placeholders.
401
+
- `sessions.get` returns the stored session messages after omitting internal OpenClaw delivery-mirror transcript artifacts.
402
+
- Chat execution still uses `chat.history`, `chat.send`, `chat.abort`, and `chat.inject`. `chat.history` is display-normalized for UI clients: inline directive tags are stripped from visible text, plain-text tool-call XML payloads (including `<tool_call>...</tool_call>`, `<function_call>...</function_call>`, `<tool_calls>...</tool_calls>`, `<function_calls>...</function_calls>`, and truncated tool-call blocks) and leaked ASCII/full-width model control tokens are stripped, internal OpenClaw delivery-mirror transcript artifacts and pure silent-token assistant rows such as exact `NO_REPLY` / `no_reply` are omitted, and oversized rows can be replaced with placeholders.
`<function_calls>...</function_calls>`, and truncated tool-call blocks), and
35
-
leaked ASCII/full-width model control tokens are stripped from visible text,
36
-
and assistant entries whose whole visible text is only the exact silent
37
-
token `NO_REPLY` / `no_reply` are omitted.
35
+
leaked ASCII/full-width model control tokens are stripped from visible text.
36
+
Internal OpenClaw delivery-mirror transcript artifacts and assistant entries
37
+
whose whole visible text is only the exact silent token `NO_REPLY` /
38
+
`no_reply` are omitted.
38
39
- Reasoning-flagged reply payloads (`isReasoning: true`) are excluded from WebChat assistant content, transcript replay text, and audio content blocks, so thinking-only payloads do not surface as visible assistant messages or playable audio.
39
40
-`chat.inject` appends an assistant note directly to the transcript and broadcasts it to the UI (no agent run).
40
41
- Aborted runs can keep partial assistant output visible in the UI.
0 commit comments