Lane 10 PR 4b: reactions and taking back (#736) #883

Merged

claude-bot merged 7 commits from build/lane10-4b-reactions into main 2026-09-27 20:39:16 +00:00

Conversation 0 Commits 7 Files changed 31 +1178 -82

claude-bot commented 2026-09-27 04:51:00 +00:00

Collaborator

Copy link

Part of #736 (lane 10). Stacked on #876 (4a posts). It merges with that stack in the train.

What it builds

• The vote reply's slate (EX-024). The bot's vote route now returns your_answers, so the bot's reply names the voter's whole current slate: "your yes for Option B is recorded — A and B work for you."

• ❔ maybe on a confirmed session (EX-023).

  • rsvp: yes|maybe|no joins BotAttendanceRequest, and attended is now optional. A maybe sets rsvp_status and never touches attended.
  • The confirmed post is seeded with ✅ ❌ ❔. The three are mutually exclusive, on a best-effort basis that depends on the bot having Manage Messages.
  • The confirmed footer now says "❔ if you might".

  4a's note declining a third reaction is about the per-slot time poll (ten emoji on a five-slot poll), not this single attendance ❔.

• Taking back (O10, D157).

  • shelf_reveals records the Discord message it was posted as. POST /api/bot/reveals/{id}/message is new, and the bot reports where the embed landed.
  • revoke_reveal publishes reveal_taken_back, and the bot deletes its own post. That is best-effort and silent on 404 or Forbidden.
  • A second revoke does nothing.
  • DM handouts are lane 8's shape and get the same columns when lane 8's migration lands.

• A reveal_taken_back bell row for the campaign's other GMs, excluding the actor, on the GM-tasks row. It uses the same shape as member_joined.

Migration

10b2c3d4e5f6_reveal_discord_message on 10d4e5f6a7b8:

• it adds the two message columns;
• ck_notifications_kind gains reveal_taken_back and remains the union of all earlier kinds.

Up → down → up is clean on Postgres 16. The one-head pin moved to the new migration's test.

Bot contract

Additive; it stays at v2. The new surface:

• an optional field (rsvp);
• a new response field (your_answers);
• a new route;
• a new stream event type, which older bots log and drop.

check_version_sync.py passes.

Checks

• Bot suite: 625 passed.
• Backend: 337 passed across every touched module. The full backend suite was not run locally because another agent's suite held the slot; CI runs it.
• Frontend: vitest 2605 passed (NOTE_KIND covers the new kind). eslint reports 0 errors.
• ruff: clean. The bot/ directory was not reformatted.
• Changelog: a changelog.d/ fragment.

Question for the owner

The bell audience for reveal_taken_back (co-GMs, excluding the actor) was inferred, not specified. It's on the questions page as B11.

🤖 Generated with Claude Code

Part of #736 (lane 10). **Stacked on #876 (4a posts).** It merges with that stack in the train. ## What it builds - **The vote reply's slate (EX-024).** The bot's vote route now returns `your_answers`, so the bot's reply names the voter's whole current slate: "your yes for Option B is recorded — A and B work for you." - **❔ maybe on a confirmed session (EX-023).** - `rsvp: yes|maybe|no` joins `BotAttendanceRequest`, and `attended` is now optional. A maybe sets `rsvp_status` and never touches `attended`. - The confirmed post is seeded with ✅ ❌ ❔. The three are mutually exclusive, on a best-effort basis that depends on the bot having Manage Messages. - The confirmed footer now says "❔ if you might". 4a's note declining a third reaction is about the **per-slot time poll** (ten emoji on a five-slot poll), not this single attendance ❔. - **Taking back (O10, D157).** - `shelf_reveals` records the Discord message it was posted as. `POST /api/bot/reveals/{id}/message` is new, and the bot reports where the embed landed. - `revoke_reveal` publishes `reveal_taken_back`, and the bot deletes its own post. That is best-effort and silent on 404 or Forbidden. - A second revoke does nothing. - DM handouts are lane 8's shape and get the same columns when lane 8's migration lands. - **A `reveal_taken_back` bell row** for the campaign's *other* GMs, excluding the actor, on the GM-tasks row. It uses the same shape as `member_joined`. ## Migration `10b2c3d4e5f6_reveal_discord_message` on `10d4e5f6a7b8`: - it adds the two message columns; - `ck_notifications_kind` gains `reveal_taken_back` and remains the union of all earlier kinds. Up → down → up is clean on Postgres 16. The one-head pin moved to the new migration's test. ## Bot contract **Additive; it stays at v2.** The new surface: - an optional field (`rsvp`); - a new response field (`your_answers`); - a new route; - a new stream event type, which older bots log and drop. `check_version_sync.py` passes. ## Checks - **Bot suite:** 625 passed. - **Backend:** 337 passed across every touched module. The full backend suite was not run locally because another agent's suite held the slot; CI runs it. - **Frontend:** vitest 2605 passed (`NOTE_KIND` covers the new kind). eslint reports 0 errors. - **ruff:** clean. The `bot/` directory was not reformatted. - **Changelog:** a `changelog.d/` fragment. ## Question for the owner The bell audience for `reveal_taken_back` (co-GMs, excluding the actor) was inferred, not specified. It's on the questions page as B11. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

claude-bot added 2 commits 2026-09-27 04:51:01 +00:00

feat: reactions and taking things back — the vote slate, ❔ maybe, reveal_taken_back (lane 10 PR 4b, #736) ... a769750cf6

WIP: migration 10b2c3d4e5f6 (shelf_reveals.discord_message_id/channel_id,
ck_notifications_kind + reveal_taken_back); the bot vote reply carries
your_answers so an ack can name a voter's whole slate (EX-024); attendance
reactions gain ❔ maybe with mutual exclusion against ✅/❌ (EX-023, D50),
carried as BotAttendanceRequest.rsvp without touching attended; the bot
reports a shelf_reveal's message id back (POST /api/bot/reveals/{id}/message)
so revoke_reveal tells the bot to delete it (O10) and notifies the campaign's
other GMs. All additive: BOT_CONTRACT_VERSION stays 2.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

fix: ruff format, migration test fixes, changelog fragment (lane 10 PR 4b, #736) ... d1ff636846

ruff format on the four backend files that needed it; the reveal-discord-
message migration test's _use_it step must be sync, not async, to actually
run (alembic's Operations.context calls it plain); the idempotency test
patches publish_bot_event_async around both revokes instead of only the
second. changelog.d/736-lane10-reactions.changed.md.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

claude-bot referenced this pull request 2026-09-27 05:42:15 +00:00

Lane 10 PR 5: bot commands and /concern (#736) #885

rbrooks referenced this pull request from a commit 2026-09-27 06:00:57 +00:00

docs(design): record D235-D238, the owner's answers to b10a, w17, G1 and B11

rbrooks added 1 commit 2026-09-27 08:32:54 +00:00

Merge lane10-4a-posts (with main) into build/lane10-4b-reactions 32c404ac28

rbrooks added 1 commit 2026-09-27 10:07:43 +00:00

Merge lane10-4a-posts (J5 recap-email row fix) into build/lane10-4b-reactions bbf8969c04

claude-bot changed target branch from build/lane10-4a-posts to main 2026-09-27 11:48:59 +00:00

rbrooks added 1 commit 2026-09-27 16:19:38 +00:00

Merge main into build/lane10-4b-reactions 71026dd134

claude-bot referenced this pull request 2026-09-27 17:55:44 +00:00

Dev checkpoint: deploy main to dev once lanes 9 and 10 land #895

rbrooks added 1 commit 2026-09-27 19:32:47 +00:00

test(backend): the discussions migration's kinds are a historical fact without reveal_taken_back (#736) ... ae35271182

test_discussion_notifications_migration (from lane 8 PR 1, merged after this
PR was built) compares its migration's kind CHECK to the model minus later
kinds; reveal_taken_back is one more, added by 10b2c3d4e5f6.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

rbrooks added 1 commit 2026-09-27 19:49:19 +00:00

test(backend): the safety and server-owner migrations' kinds are historical facts without reveal_taken_back (#736) ... e3a3917429

The same exclusion as the discussions migration's test; every
kinds-versus-model test now passes (269, run as one set).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

claude-bot merged commit 1de174bb47 into main 2026-09-27 20:39:16 +00:00

claude-bot referenced this pull request from a commit 2026-09-27 20:39:18 +00:00

Merge PR #883: lane 10 PR 4b, reactions and taking back (#736)

Sign in to join this conversation.

Reviewers

No reviewers

Labels

Clear labels   

backend

Issue related to the backend

  

bot

Issue related to the bot

  

bug

Something is not working

  

duplicate

This issue or pull request already exists

  

enhancement

New feature

  

frontend

Issue related to the frontend

  

help wanted

Need some help

  

invalid

Something is wrong

  

question

More information is needed

  

recording

Issue related to the recording system

  

security

Security hardening / vulnerability work

  

wontfix

This won't be fixed

No labels

backend

bot

bug

duplicate

enhancement

frontend

help wanted

invalid

question

recording

security

wontfix

Milestone

Clear milestone

No items

No milestone

Projects

Clear projects

No items

No project

Assignees

Clear assignees

No assignees

2 participants

Notifications

Subscribe

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference

rbrooks/Quest-Board!883

No description provided.