feat(messages): per-message send_at on POST /v1/messages (parity port of cueapi/cueapi#623) by mikemolinet · Pull Request #46 · cueapi/cueapi-core

mikemolinet · 2026-05-05T16:23:49Z

Summary

Parity port of cueapi/cueapi#623 (Phase 12.1.7 / roadmap §13). Optional send_at timestamp on MessageCreate delays delivery until the time elapses. Same shape as cue-fire send_at (PR #45 here / cueapi/cueapi#618).

Implementation surfaces

Migration 024 (renumbered from private's 047 to fit OSS sequence after existing 023_messaging_primitive_multi_shell.py) adds messages.send_at TIMESTAMPTZ NULL plus a partial index ix_messages_send_at (WHERE send_at IS NOT NULL) built CONCURRENTLY. Existing rows default to NULL = "send now" (full back-compat).
MessageCreate + MessageResponse schemas grow send_at: Optional[datetime].
create_message plumbs send_at into both Message.send_at and DispatchOutbox.scheduled_at so push-delivery dispatch is also gated. Past timestamps are forgiving fallback (treated as "send now").
list_inbox gates with Message.send_at IS NULL OR send_at <= now() on both the read query AND the queued→delivered transition UPDATE.
list_sent unchanged — sender SHOULD see their scheduled messages.

Files changed

alembic/versions/024_message_send_at.py — new migration (renumbered from private's 047)
app/models/message.py — send_at column
app/schemas/message.py — send_at on MessageCreate + MessageResponse
app/services/message_service.py — send_at parameter + plumbing
app/services/inbox_service.py — send_at gate on inbox query + transition UPDATE
app/routers/messages.py — pass send_at from body to service
tests/test_message_send_at.py — 7 new tests
parity-manifest.json — bump 5 entries to 2026-05-05

Tests

7 new (matching private). 660 passed locally (excluding 7 pre-existing SDK test failures unrelated to this PR).

Phase 12.1.7 OSS port progress

✅ cue-fire send_at (cueapi-core PR feat(cues): per-fire send_at scheduling on POST /v1/cues/{id}/fire (parity port of cueapi/cueapi#618) #45 / cueapi/cueapi#618)
✅ message send_at (this PR / cueapi/cueapi#623)
⏸ §17 BCC light (cueapi/cueapi#619) — separate port, not yet here

Parity Impact

cueapi-core ✓ (this PR)
cueapi-python — TODO: client.messages.create(send_at=...) (Backlog row)
cueapi-mcp — TODO: cue-mac-app's lane (Backlog row)
cueapi-cli — N/A (no message create CLI yet)
cueapi-worker — N/A
cueapi-action — N/A
homebrew-tap — N/A
cuechain — N/A

🤖 Generated with Claude Code

github-actions · 2026-05-05T16:24:00Z

Parity check

This PR modifies files tracked in parity-manifest.json:

app/models/message.py
app/routers/messages.py
app/schemas/message.py
app/services/inbox_service.py
app/services/message_service.py

Please confirm one of the following in a reply or PR description update:

The equivalent change has been applied to the private cueapi monorepo. Link the PR.
This change is OSS-only and does not need porting. Briefly explain why (e.g. "fixes a bug that only exists in the OSS build").
A follow-up issue has been filed to port the reverse direction. Link the issue.

This is a soft check — it does not block merge. The goal is visibility, not friction. See HOSTED_ONLY.md for the open-core policy.

…parity port of cueapi/cueapi#623) Mirrors private PR #623 (Phase 12.1.7 / roadmap §13). Optional ``send_at`` timestamp on MessageCreate delays delivery until the time elapses. Same shape as cue-fire send_at (port #45 / cueapi/cueapi#618). Implementation surfaces: * **Migration 024** (renumbered from private's 047 to fit OSS sequence after existing 023_messaging_primitive_multi_shell.py) adds ``messages.send_at TIMESTAMPTZ NULL`` plus a partial index ``ix_messages_send_at`` (WHERE send_at IS NOT NULL) built CONCURRENTLY to avoid ACCESS EXCLUSIVE lock on a potentially large messages table. Existing rows default to NULL = "send now" (full back-compat). * **MessageCreate** + **MessageResponse** schemas grow ``send_at: Optional[datetime]``. ``extra="forbid"`` already in place. * **create_message** plumbs ``send_at`` into both ``Message.send_at`` and ``DispatchOutbox.scheduled_at`` so push-delivery dispatch is also gated. Past timestamps are forgiving fallback ("send now") — caller doesn't have to worry about clock skew. * **list_inbox** gates with ``Message.send_at IS NULL OR send_at <= now()`` on both the read query AND the queued→delivered transition UPDATE. Recipients can't see scheduled messages until their time; the atomic poll-fetch transition skips them too. * **list_sent** unchanged — sender SHOULD see their scheduled messages (they queued them deliberately). Files: - alembic/versions/024_message_send_at.py — new migration - app/models/message.py — send_at column - app/schemas/message.py — send_at on MessageCreate + MessageResponse - app/services/message_service.py — send_at parameter + plumbing - app/services/inbox_service.py — send_at gate on inbox query + transition UPDATE - app/routers/messages.py — pass send_at from body to service - tests/test_message_send_at.py — 7 new tests - parity-manifest.json — bump 5 entries to 2026-05-05 Tests: 7 new. 660 passed total (excluding 7 pre-existing SDK test failures). Phase 12.1.7 OSS port complete: cue-fire send_at (port/618 / PR #45) + message send_at (this PR). §17 BCC light (private PR #619) is a separate parity port not yet ported here. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

…ueapi/cueapi#630) Mirrors private PR #630 (Phase A of Agent Directory PRD §Surface 1 + §Surface 5). Adds an implicit "online" derivation so callers don't have to PATCH agent status explicitly — the server bumps ``agents.last_seen_at`` on hot-path activity (message send by sender, inbox poll by recipient) and derives ``online`` from recency. What this PR ships: * **Migration 024** — adds ``agents.last_seen_at TIMESTAMPTZ NULL``. Nullable, no backfill (existing rows stay caller-asserted until first hot-path write lands). NOTE: collides with open OSS PR #46 (also targets 024_message_send_at) — whichever lands first, the other rebases + renumbers to 025. * **GET /v1/agents/roster** — display-optimized snapshot for prompt-injection at session-boot. Distinct from the existing management surface (``GET /v1/agents``): - Always-full list (no pagination) - Drops opaque IDs / secrets / timestamps / tenancy metadata - Adds derived ``online`` / ``last_seen_relative`` / ``preferred_contact`` - Always excludes soft-deleted agents - Weak ETag + ``If-None-Match`` → 304 Not Modified for poll efficiency * **Hot-path hooks** write ``last_seen_at = now()``: - ``create_message`` — sender's agent (in same tx as message insert) - ``list_inbox`` — recipient's agent on every poll * **Online derivation** (server-computed): - ``last_seen_at`` ≤ 5 min → ``online`` - ≤ 30 min → ``away`` - older or NULL → ``offline`` - Caller override wins: PATCH status to ``away`` or ``offline`` sticks regardless of activity. Files: - ``alembic/versions/024_agents_last_seen_at.py`` — new migration - ``app/models/agent.py`` — ``last_seen_at`` column - ``app/schemas/agent.py`` — ``AgentRosterEntry`` + ``AgentRosterResponse`` - ``app/services/agent_service.py`` — ``list_roster`` + 7 pure helpers - ``app/services/inbox_service.py`` — ``_bump_last_seen_stmt`` + integration - ``app/services/message_service.py`` — sender's ``last_seen_at`` hook - ``app/routers/agents.py`` — ``GET /v1/agents/roster`` + ``_etag_matches`` - ``tests/test_agent_roster.py`` — 21 new tests (15 pure unit + 6 integration) - ``parity-manifest.json`` — bump 6 entries to 2026-05-05 Tests: 21 new. 657 passed total (excluding 7 pre-existing SDK test failures unrelated to this PR). Out of scope (separate ports / future PRs): * MCP tool ``cueapi_agents_list`` / ``cueapi_agents_describe`` — cue-mac-app's lane * CLI ``agents list --online-only`` / ``agents describe`` — cueapi-secondary's lane * Desktop UI Directory tab — cue-mac-app's lane * Per-agent ``kind`` / ``capabilities`` fields — §1 capability registry territory Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

mikemolinet · 2026-05-11T15:32:54Z

Closing as stale — branch diverged from main by ~8901 deletions across 59 files. CI is FAILURE on this branch (test + sdk-integration). Rebase infeasible given the drift.

Feature is still needed: send_at is absent from cueapi-core's app/schemas/message.py + app/routers/messages.py. Parity port of cueapi/cueapi#623 has not landed in the SERVER side (cli/mcp/python sides shipped 2026-05-10).

Next step: re-port as fresh small PR against current main HEAD (f9ec4ea).

Branch retained for reference.

…parity port of cueapi/cueapi#623) (#77) Re-port of closed [PR #46](#46) which was on a stale base ~8900 deletions behind main. Fresh against current main HEAD. Closes §13 / Phase 12.1.7 — messages-side complement to the cue-fire send_at shipped in PR #618 (which still needs its own re-port). ## What lands - **alembic/versions/030_message_send_at.py** (renumbered from private's 047) Adds `messages.send_at TIMESTAMPTZ NULL` + partial index `ix_messages_send_at` (WHERE send_at IS NOT NULL) built CONCURRENTLY so the index creation doesn't take an ACCESS EXCLUSIVE lock on the messages table during deploy. Existing rows default to NULL = "send now". - **app/models/message.py** — adds `send_at` Column. - **app/schemas/message.py** — `send_at: Optional[datetime]` on MessageCreate + MessageResponse. - **app/routers/messages.py** — passes `send_at=body.send_at` into `create_message`. - **app/services/inbox_service.py** — gates inbox query AND queued→delivered transition UPDATE with `send_at IS NULL OR send_at <= now()`. Scheduled messages are invisible until their time; the atomic poll-fetch transition skips them too. - **app/services/message_service.py** — `create_message` plumbs `send_at` into both `Message.send_at` and `DispatchOutbox.scheduled_at`. Past timestamps are forgiving fallback (treated as send-now). `to_response_dict` surfaces the persisted value. - **tests/test_message_send_at.py** — 7 new tests verbatim from private cueapi covering all 7 semantic paths (omitted, future-invisible, outbox-scheduled, past-fallback, becomes-visible- after-pass, sender-view-shows-it, invalid-timestamp). - **parity-manifest.json** — new entry for migration 030 under `message-send-at-port (private #623)`. ## Wire format `send_at` flows in the BODY of `POST /v1/messages` (server contract: `MessageCreate.send_at`). Same shape as cue-fire `send_at` (PR #618). NULL = send now (default). Future timestamp = inbox-gate + DispatchOutbox.scheduled_at. Past timestamp = forgiving send-now. ## Tests 7/7 new tests pass locally. Full local suite: 836 passed + 18 xfailed (pre-existing) + 3 skipped. Zero regressions. ## Sibling ports - cli, mcp, python, action sides shipped via session 2 (2026-05-10): cueapi-cli #48, cueapi-mcp #33, cueapi-python (private), cueapi-action #12. This is the cueapi-core (OSS server) side that was still missing. ## Re-port note Re-port of closed PR #46. That branch was ~8900 deletions behind main; fresh port against current main HEAD (after PR #74 + #75 merged earlier in this session).

mikemolinet added auto-merge parity labels May 5, 2026

govindkavaturi-art enabled auto-merge (squash) May 5, 2026 16:27

mikemolinet mentioned this pull request May 5, 2026

feat(agents): roster endpoint + last_seen_at + ETag (parity port of cueapi/cueapi#630) #47

Closed

6 tasks

mikemolinet force-pushed the port/623-message-send-at branch from 885e3d0 to fd4396e Compare May 7, 2026 00:34

mikemolinet closed this May 11, 2026

auto-merge was automatically disabled May 11, 2026 15:32
Pull request was closed

mikemolinet mentioned this pull request May 11, 2026

feat(messages): per-message send_at scheduling on POST /v1/messages (parity port of cueapi/cueapi#623) #77

Merged

6 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat(messages): per-message send_at on POST /v1/messages (parity port of cueapi/cueapi#623)#46

feat(messages): per-message send_at on POST /v1/messages (parity port of cueapi/cueapi#623)#46
mikemolinet wants to merge 1 commit into
mainfrom
port/623-message-send-at

mikemolinet commented May 5, 2026

Uh oh!

github-actions Bot commented May 5, 2026 •

edited

Loading

Uh oh!

mikemolinet commented May 11, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

mikemolinet commented May 5, 2026

Summary

Implementation surfaces

Files changed

Tests

Phase 12.1.7 OSS port progress

Parity Impact

Uh oh!

github-actions Bot commented May 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Parity check

Uh oh!

mikemolinet commented May 11, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

github-actions Bot commented May 5, 2026 •

edited

Loading