fix(gateway): keep shared-group sender prefix outermost

[jsky-james]

Summary

Fix shared-group sender attribution when inbound messages receive prepend-style enrichment.

Previously _prepare_inbound_message_text() applied [DisplayName] before image/STT/audio/video/document, Discord, and reply context processing. Later enrichment could prepend text ahead of the sender tag, so media turns no longer began with the sender identity.

This change:

• applies the sanitized sender prefix once after all inbound enrichments (including reply and @ context expansion),
• keeps channel_context outermost, so only the triggering turn is attributed,
• preserves untagged behavior for per-user group sessions and missing sender names.

Regression coverage

Adds a shared-group regression covering an audio attachment, reply-to context, channel backfill, control-character sanitization, and exactly-once sender attribution. Also covers missing sender names.

Verification

• python -m pytest -q tests/gateway/test_shared_group_sender_prefix.py tests/gateway/test_reply_to_injection.py tests/gateway/test_video_context_note.py
  • 11 passed
• python -m compileall -q gateway/run.py
• git diff --check

[alt-glitch]

This was generated by AI during triage.

Related to #44705, #13939, and #16850's sender-attribution work. This PR is a narrower ordering fix: it keeps the existing shared-group prefix outermost after media/reply enrichment.

[x7peeps]

Closing as duplicate

[teknium1]

Thanks for the focused regression and ordering fix. The current main implementation still has the reported ordering issue: gateway/run.py:12873-12898 adds sender/backfill before later media and reply prepend paths (:13014-13031, :13118-13132).

Problems

• The moved block predates main's Slack current-author-ID behavior. Current main adds | Slack user <@...> at gateway/run.py:12888-12891, with coverage in tests/gateway/test_shared_group_sender_prefix.py:74-131; preserve that nested branch when moving the prefix or shared Slack turns lose their verified mention target.

Suggested changes

• Salvage the reorder over current gateway/run.py and add an enriched shared-Slack regression asserting the outer prefix retains <@U...>.

This is an automated hermes-sweeper review.

[teknium1]

When salvaging this moved construction onto current main, retain the Slack-specific | Slack user <@{source.user_id}> augmentation added by commit 503c0c0e51 before constructing this prefix. Otherwise enriched shared Slack turns lose the verified current-author mention target.

[GottZ]

This was generated by AI during triage.

Summary

Five PRs address or reference the two issues: #18711 exposes the verified Slack author mention, #66507 fixes internal-event impersonation and redundant async follow-ups, #67886 repairs sender-prefix ordering, while #13939 broadens ID-qualified attribution and #44705 labels unnamed shared-session participants. The latter three attribution changes overlap in gateway code but address distinct causes and carry documented compatibility or rebase requirements.

Related pull requests

• feat(gateway): authoritative sender attribution in all chat contexts #13939 fixes — (+240/-7) — n/a: Adds default-on ID-qualified attribution across inbound contexts and spoof-prefix stripping, broadly addressing ambiguous sender identity rather than Slack self-mention binding specifically. The contributor COMMENTED keep_open review identifies required salvage work: use user_id_alt or user_id, move behavioral configuration into config.yaml, document it, and add Signal/Feishu plus config-loading coverage.
• fix(slack): expose shared-thread author mention #18711 [closed] best fix — (+61/-1) — n/a: The recorded best-fix for Slack shared-thread self-mention can target the wrong user when current author ID is not exposed to the agent #17916 adds the verified current Slack <@user_id> to shared-turn prefixes and warns against guessing historical mentions; it remains relevant because contributor teknium1 states that it was cherry-picked with authorship via fix(slack): thread-context lifecycle — cold-start, restart rehydration, mention refresh, reply-wake #69320. Despite the earlier automated keep_open verdict—which noted that this diff lacked gateway-side outbound validation—the PR is already closed after that integration and must not be reopened.
• Label unnamed shared-session participants #44705 related — (+166/-18) — n/a: Adds sanitized, stable hashed labels for unnamed shared-session participants, but its fallback can double-prefix Yuanbao messages that are already attributed. The COMMENTED keep_open review supports salvage only after preserving an explicit no-second-prefix path and adding a Yuanbao regression; the unrelated transport-fanout and release-attribution hunks should be split out.
• fix(gateway): skip sender prefix for internal events + NO_REPLY contract in async delegation (#66480) #66507 best fix — (+91/-2) — n/a: The recorded best-fix for [Bug]: Async delegation completion impersonates user and prompts redundant follow-up #66480 prevents internal synthetic events from receiving a human sender prefix and adds the exact NO_REPLY no-news contract to both single and batch async-delegation formatters. This matches the automated keep_open verdict, whose cited current-main paths show both causes remain and the focused hunks are mechanically salvageable.
• fix(gateway): keep shared-group sender prefix outermost #67886 related — (+95/-18) — n/a: Moves shared-session attribution after media, reply, and @ enrichment so the sender prefix wraps the complete triggering turn while channel backfill remains outermost. Despite the earlier contributor comment saying “Closing as duplicate,” the diff is a distinct ordering fix; the later COMMENTED keep_open review requires rebasing the move onto current main, retaining commit 503c0c0e51's Slack <@user_id> augmentation, and adding an enriched shared-Slack regression.

Duplicates

No listed PRs are exact duplicates: #13939 and #44705 overlap on sender attribution, while #67886 changes ordering in the same path but fixes a separate enrichment bug. #18711 is the closed source implementation cherry-picked through #69320, and #66507 addresses the separate internal-event/async-completion issue.

Suggested consolidation

Keep #66507 open with a salvage path for #66480, and keep #67886 open with a salvage path that rebases its ordering change onto current main while preserving the Slack mention augmentation and adding the requested regression. Author action for #13939: rebase or split out the viable attribution work after resolving the contributor's identity-precedence, configuration, documentation, and coverage findings; author action for #44705: split the participant-labeling fix from unrelated hunks and add the Yuanbao no-double-prefix guard. Leave #18711 closed as the source already integrated via #69320; none of the remaining listed PRs has sufficient evidence to close as a duplicate of another listed PR.

Cross-PR triage: Reviewed 5 pull requests and 2 issues in this complex. Each diff was read against this issue; Assessment working set: 42 kB of PR diffs, 18 kB of issue/PR text, 9 kB of discussion (10 comments), 6 verify verdicts. verdicts reflect diff content, not PR titles. Part of an automated triage batch.