* fix(feishu): log message content JSON parse failures instead of silently swallowing
Replace formatErrorMessage(err) with safe metadata-only logging in
parseFeishuMessageContent to prevent potential message content leaks
through V8 JSON.parse error messages.
Changes:
- Remove formatErrorMessage import (security: V8 JSON.parse errors can
include input content in the message)
- Log only msgType and optional messageId (safe metadata) when parse
fails, never the exception message or raw content
- Add assertion that raw content is NOT present in the log output
- Pass messageId through to enable richer diagnostics
The raw content is still preserved as the function return value (existing
fallback behavior).
* fix(feishu): move parse-failure test into getMessageFeishu suite
The test 'logs a safe diagnostic (not raw content) when message content
is not valid JSON' was declared after the closing brace of
describe('getMessageFeishu'), so it did not inherit that suite's fixture
setup and reset hooks (beforeEach/afterAll). Move it inside the suite
so it benefits from the shared mock reset and cleanup.
Fixes ClawSweeper P2: 'Keep the parse-failure test inside the fetch suite'
---------
Co-authored-by: Peter Steinberger <steipete@gmail.com>
* fix(synology-chat): stop fabricating send message ids
The incoming-webhook send contract acks with a bare boolean and carries
no platform message id, but the adapter returned a synthetic
`sc-${Date.now()}` id and stamped it into the delivery receipt. Return
the established empty-id sentinel and an empty receipt instead, matching
the honest no-platform-id shape used by sibling channels (qqbot,
googlechat) and the plugin-sdk empty-id contract.
Co-Authored-By: Claude <noreply@anthropic.com>
* test(synology-chat): verify truthful webhook receipts
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Peter Steinberger <steipete@gmail.com>
The react message-action handler dispatched straight to the sender without
checking the resolved account, so a disabled Nextcloud Talk account
(`enabled:false`) that still had a baseUrl/botSecret in config could keep
emitting reactions. `describeMessageTool` already hides the tool for
unconfigured accounts, but an explicit accountId can reach `handleAction`
directly. Enforce the same enabled+configured gate at dispatch, mirroring
the Signal reaction fix (#112607).
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(codex): restore connected apps for token-authenticated runs
* fix(codex): keep app inventory protocol types private
* fix(codex): align native runtime with Codex 0.146.0
Co-authored-by: Steven Lee <stevenlee@openai.com>
* fix(codex): clean up latest app-server integration
Co-authored-by: Steven Lee <stevenlee@openai.com>
* fix(codex): keep internal protocol types private
* fix(ci): repair current main Codex landing gates
* fix(ci): format inherited code mode matrix
* fix(codex): reconcile native app-server contracts with main
Prepare a verified GitHub-hosted mainline merge while preserving the reviewed Codex 0.146.0 fixes and canonical OpenAI authentication.
Co-authored-by: Steven Lee <stevenlee@openai.com>
* fix(codex): keep QA evidence in its owning plugin
Resolve the current-main Code Mode test rename without resurrecting the retired core test path.
Co-authored-by: Steven Lee <stevenlee@openai.com>
* fix(codex): enforce canonical OpenAI app-server auth
Reject retired provider aliases without runtime compatibility, direct operators to the doctor migration, and remove the redundant OpenAI API-key predicate.
Co-authored-by: Steven Lee <stevenlee@openai.com>
* chore(codex): reconcile latest main dependency graph
Preserve current main dependency changes while preparing the original Codex PR for an ancestry-preserving signed mainline merge.
Co-authored-by: Steven Lee <stevenlee@openai.com>
* fix(codex): unify bundled Codex 0.146 runtimes
Keep the ACP adapter on the same 0.146.0 Codex release as the managed runtime, remove obsolete 0.145.0 platform artifacts and unused semver compatibility, and preserve the latest main dependency upgrades.
Co-authored-by: Steven Lee <stevenlee@openai.com>
---------
Co-authored-by: Peter Steinberger <peter@steipete.me>
Co-authored-by: Peter Steinberger <steipete@gmail.com>
* fix(msteams): keep file replies in channel threads
* fix(msteams): keep SharePoint files in channel threads
Fixes#88836 by forwarding channel thread roots through the existing Microsoft Teams proactive activity boundary while preserving top-level, group chat, and personal delivery.
Co-authored-by: Jasmine Zhang <jasminezhang@JasminedeMac-mini.local>
* test(msteams): use lint-safe real SDK HTTP capture
---------
Co-authored-by: Jasmine Zhang <jasminezhang@JasminedeMac-mini.local>
Co-authored-by: Peter Steinberger <steipete@gmail.com>
* fix(telegram): linkPreview:false is ignored on streamed replies
Replies delivered through draft streaming still unfurled the first URL
even with channels.telegram.linkPreview: false. Non-streamed sends
already honored the flag.
createTelegramDraftStream never received linkPreview and never set
link_preview_options, on either its initial sendMessage or its
editMessageText calls. Finalization could not clean it up: it skips the
final edit when the streamed draft text already equals the final text,
which is the common case, so the draft message stays as-is with the
preview attached.
Passes linkPreview from telegramCfg into the draft stream and applies
link_preview_options: { is_disabled: true } on the send and on every
edit, since an edit that omits the field re-enables the preview
server-side. Rich messages keep expressing this as skip_entity_detection
at render time, which already worked.
Call arity is preserved when no preview options apply, so existing edit
assertions are unaffected.
Closes#111525
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* test(telegram): cover streamed reply option parity
Co-authored-by: Zakaria Rahali <zakariarahali288@gmail.com>
---------
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
Co-authored-by: Peter Steinberger <steipete@gmail.com>
Apply the shared, surrogate-safe outbound chunker to the existing Nostr text limit so the real delivery planner produces ordered NIP-04 events. Preserve assistant-visible sanitization and prove short, word, newline, hard-split, and odd-prefix Unicode boundaries.
Co-authored-by: liyuanbin <li.yuanbin1@xydigit.com>
* docs(agents): standardize canonical record guards
* refactor(model-catalog): use canonical record guard
* refactor(scripts): use canonical record guard
* refactor(ui): use canonical record guard
* refactor(codex): use canonical record guard
* refactor(agents): use canonical record guard
* refactor(sessions): use canonical record guard
* refactor(doctor): use canonical record guard
* refactor(infra): use canonical record guard
Allow before_agent_reply plugins to declare host-enforced trigger eligibility so scheduled-only hooks do not block interrupted user-turn recovery. Keep omitted and malformed scopes fail-closed, scope both memory-core maintenance hooks, and cover three runner reload cycles through the public registration contract.
Refs: #111442
Source: #114836
Co-authored-by: Vincent Koc <vincentkoc@ieee.org>
Preserve the authoritative inbound group origin for private Mattermost channels without changing public channels, direct messages, thread keys, or the native channel delivery target. Prove both Bot API channel types over real HTTP, gateway discovery, and SQLite-backed outbound persistence.
Credit the original group-origin observation in #95669.
Co-authored-by: hansraj <hansraj136@gmail.com>
Emit the established OpenClaw durable-delivery response marker only after an authenticated, non-empty LINE webhook event has been persisted. Verify the production gateway route with signed real HTTP requests covering verification, delayed admission, invalid signatures, and storage failures.
Co-authored-by: 許元豪 <146086744+edenfunf@users.noreply.github.com>
* fix(mattermost): record pending history for non-allowlisted group senders
When `groupPolicy` is `"allowlist"`, messages from senders not in
`groupAllowFrom` were silently dropped before reaching
`recordPendingHistoryEntryIfEnabled()`. This meant the bot had no
context of what non-allowlisted users said, breaking use cases like
conversation summarization.
This change introduces a `senderBlockedByGroupAllowlist` flag so that
non-allowlisted messages continue through the history-recording code
path before returning without generating a response.
Changes:
1. Add `senderBlockedByGroupAllowlist` flag before access check
2. `GROUP_POLICY_NOT_ALLOWLISTED` sets flag instead of returning
3. Skip `commandGate` check for blocked senders
4. Call `recordPendingHistory()` then return before response processing
Fixesopenclaw/openclaw#57607
* test(mattermost): verify denied history over real transport
Co-authored-by: Jason Wang <jasonwang@barkingdog.ai>
---------
Co-authored-by: Jason Wang <jasonwang@barkingdog.ai>
* fix(msteams): honor group conversation allowlists
Fixes#95737. Preserve opaque Teams conversation identities, the documented group fallback, personal-DM isolation, and display-name spoof protection.
Based on the original contributor fix in #95764.
Co-authored-by: 周鹤0668001310 <zhou.he3@xydigit.com>
* test(msteams): isolate conversation allowlist regression proof
Co-authored-by: 周鹤0668001310 <zhou.he3@xydigit.com>
---------
Co-authored-by: 周鹤0668001310 <zhou.he3@xydigit.com>
Sanitize both Nostr delivery paths with the shared assistant-visible text contract. Preserve ordinary replies and suppress internal-only messages before Markdown conversion.
Related: #90684
Co-authored-by: liyuanbin <li.yuanbin1@xydigit.com>
Qualify non-shared sandbox identities by resolved workspace while preserving shared runtime names. Existing non-shared runtimes reset once under the new identity.
Related: #51363
Co-authored-by: Tayoun <39609208+tayoun@users.noreply.github.com>
zw-xysk