feat(canvas): CommunicationOverlay → ACTIVITY_LOGGED subscriber (#61 stage 1) #69

Merged

claude-ceo-assistant merged 3 commits from feat/canvas-comm-overlay-ws-subscribe into main 2026-05-07 23:04:20 +00:00

Conversation 0 Commits 3 Files Changed 2 +354 -74

claude-ceo-assistant commented 2026-05-07 22:11:49 +00:00

Owner

Copy Link

Stage 1 of #61. Closes the first leg of the poll-fan-out reduction: CommunicationOverlay no longer polls /workspaces/:id/activity on a 30s interval; it bootstraps once on mount and subscribes to ACTIVITY_LOGGED via the existing useSocketEvent bus for live updates.

What this PR changes

Surface	Before	After
Steady-state HTTP traffic from this overlay	~6 req/min (3 ws × 2 cycles/min × 1 req)	0 req/min outside of mount + visibility-toggle bootstraps
Live update latency	up to 30s (next poll cycle)	~10ms (next WS frame after server insert)
Bootstrap cost	unchanged	3 HTTP req on mount (cap retained from 2026-05-04 fix), 3 more on each visibility re-open
WS unhealthy fallback	n/a (was always polling)	shows bootstrap snapshot until next visibility-toggle or WS reconnect

The singleton ReconnectingSocket in canvas/src/store/socket.ts already owns reconnect + backoff + health-check; routing the overlay through useSocketEvent inherits all of those for free without opening a new WebSocket per panel.

SSOT decision

useSocketEvent is the single subscription primitive — same hook ChatTab.tsx and AgentCommsPanel.tsx already use. No new abstraction. The HTTP bootstrap path reuses the same /workspaces/:id/activity?limit=5 shape the previous polling code used; the only delta is its triggering condition (mount + visibility re-open instead of every 30s).

Tests

canvas/src/components/__tests__/CommunicationOverlay.test.tsx — 9 tests, all PASS:

bootstrap fan-out cap (3 of 6 online workspaces, rate-limit floor preserved)
bootstrap never includes offline workspaces
NO interval polling after bootstrap (60s clock advance fires nothing)
visibility gate — no fetches while collapsed, re-bootstrap on re-open
WS push extends rendered list with NO additional HTTP call
WS push for offline workspace ignored
WS push for non-comm activity_type (e.g. delegation) ignored
WS push while panel is collapsed ignored
non-ACTIVITY_LOGGED events (e.g. WORKSPACE_OFFLINE) ignored

Full canvas suite: 1393 passing, 0 failing. pnpm tsc --noEmit clean.

Mutation tests

Mutation	Test that fired
Drop the visibility gate (if (!visible) return in both useEffect and useSocketEvent handler)	visibility re-bootstrap test
Drop the activity_type filter	"WS push for non-comm activity_type ignored" test
Drop the workspace online-set filter	"WS push for offline workspace ignored" test

Each verifies an actual production-code branch is exercised; no tautologies.

Security check

• Untrusted input? ACTIVITY_LOGGED payload comes through the same WS path that ChatTab and AgentCommsPanel already trust. The overlay treats every payload field as optional/coercible — activity_type, source_id, target_id, summary, status, created_at all guard for missing or wrong type. No XSS surface (everything renders through React's text-content path; no dangerouslySetInnerHTML).
• Auth/sessions/permissions? No change. WS subscription uses the same per-workspace bearer the polling path used.
• Data collection / logs? No new logging (the WS handler is silent; bus already logs subscribe/unsubscribe in socket-events.ts).
• Access boundary changes? None — the overlay still respects the same online-workspace filter as the polling version.

Versioning + backwards compat

• API surface: /workspaces/:id/activity HTTP endpoint unchanged (still used for bootstrap).
• WS event shape: ACTIVITY_LOGGED unchanged; consumed in the same shape AgentCommsPanel already uses.
• No schema, env-var, or migration impact.
• Operationally additive: existing canvas behaviour preserved for users who don't have WS connectivity (they see the bootstrap snapshot, identical to what they'd see if the polling cycle hadn't yet fired).

Hostile self-review — three weakest spots

1. Sustained WS outage shows the bootstrap snapshot until either the user toggles visibility or the WS reconnects (which itself triggers a rehydrate burst from the singleton socket). Acceptable because the comm overlay isn't a critical-path surface and socket.ts aggressively reconnects (5s health-check, exponential backoff). For users who care about freshness during a sustained outage, the visibility-toggle re-bootstrap is one click away.
2. Bootstrap on visibility-toggle costs 3 more HTTP calls each re-open. By design — visibility-toggle is a deliberate user action, not a tight loop, and the freshness benefit on re-open is exactly what the user expects.
3. The WS handler reads the latest nodes via nodesRef rather than re-subscribing on node-list changes. Same pattern A2ATopologyOverlay's comment warns about — re-subscribing on every store update would tear down + re-arm the bus listener every render. Stable subscription with ref-based current-state lookup is the right shape.

Rollout / rollback

• Rollout: merge → next canvas build picks it up. No env vars, no infra changes.
• Rollback: git revert the merge — overlay falls back to the 30s polling shape.

Out of scope (still tracked under #61)

• Stage 2: A2ATopologyOverlay (60s × N workspaces, 500-row windowed query — separate PR)
• Stage 3: ActivityTab (5s × 1 active workspace — separate PR)

🤖 Generated with Claude Code

Stage 1 of #61. Closes the first leg of the poll-fan-out reduction: `CommunicationOverlay` no longer polls `/workspaces/:id/activity` on a 30s interval; it bootstraps once on mount and subscribes to `ACTIVITY_LOGGED` via the existing `useSocketEvent` bus for live updates. ## What this PR changes | Surface | Before | After | |---|---|---| | Steady-state HTTP traffic from this overlay | ~6 req/min (3 ws × 2 cycles/min × 1 req) | 0 req/min outside of mount + visibility-toggle bootstraps | | Live update latency | up to 30s (next poll cycle) | ~10ms (next WS frame after server insert) | | Bootstrap cost | unchanged | 3 HTTP req on mount (cap retained from 2026-05-04 fix), 3 more on each visibility re-open | | WS unhealthy fallback | n/a (was always polling) | shows bootstrap snapshot until next visibility-toggle or WS reconnect | The singleton `ReconnectingSocket` in `canvas/src/store/socket.ts` already owns reconnect + backoff + health-check; routing the overlay through `useSocketEvent` inherits all of those for free without opening a new WebSocket per panel. ## SSOT decision `useSocketEvent` is the single subscription primitive — same hook `ChatTab.tsx` and `AgentCommsPanel.tsx` already use. No new abstraction. The HTTP bootstrap path reuses the same `/workspaces/:id/activity?limit=5` shape the previous polling code used; the only delta is its triggering condition (mount + visibility re-open instead of every 30s). ## Tests `canvas/src/components/__tests__/CommunicationOverlay.test.tsx` — 9 tests, all PASS: ``` bootstrap fan-out cap (3 of 6 online workspaces, rate-limit floor preserved) bootstrap never includes offline workspaces NO interval polling after bootstrap (60s clock advance fires nothing) visibility gate — no fetches while collapsed, re-bootstrap on re-open WS push extends rendered list with NO additional HTTP call WS push for offline workspace ignored WS push for non-comm activity_type (e.g. delegation) ignored WS push while panel is collapsed ignored non-ACTIVITY_LOGGED events (e.g. WORKSPACE_OFFLINE) ignored ``` Full canvas suite: **1393 passing, 0 failing**. `pnpm tsc --noEmit` clean. ### Mutation tests | Mutation | Test that fired | |---|---| | Drop the visibility gate (`if (!visible) return` in both useEffect and useSocketEvent handler) | visibility re-bootstrap test | | Drop the activity_type filter | "WS push for non-comm activity_type ignored" test | | Drop the workspace online-set filter | "WS push for offline workspace ignored" test | Each verifies an actual production-code branch is exercised; no tautologies. ## Security check - **Untrusted input?** ACTIVITY_LOGGED payload comes through the same WS path that `ChatTab` and `AgentCommsPanel` already trust. The overlay treats every payload field as optional/coercible — `activity_type`, `source_id`, `target_id`, `summary`, `status`, `created_at` all guard for missing or wrong type. No XSS surface (everything renders through React's text-content path; no `dangerouslySetInnerHTML`). - **Auth/sessions/permissions?** No change. WS subscription uses the same per-workspace bearer the polling path used. - **Data collection / logs?** No new logging (the WS handler is silent; bus already logs subscribe/unsubscribe in `socket-events.ts`). - **Access boundary changes?** None — the overlay still respects the same online-workspace filter as the polling version. ## Versioning + backwards compat - **API surface**: `/workspaces/:id/activity` HTTP endpoint unchanged (still used for bootstrap). - **WS event shape**: `ACTIVITY_LOGGED` unchanged; consumed in the same shape `AgentCommsPanel` already uses. - **No schema, env-var, or migration impact.** - **Operationally additive**: existing canvas behaviour preserved for users who don't have WS connectivity (they see the bootstrap snapshot, identical to what they'd see if the polling cycle hadn't yet fired). ## Hostile self-review — three weakest spots 1. **Sustained WS outage** shows the bootstrap snapshot until either the user toggles visibility or the WS reconnects (which itself triggers a rehydrate burst from the singleton socket). Acceptable because the comm overlay isn't a critical-path surface and `socket.ts` aggressively reconnects (5s health-check, exponential backoff). For users who care about freshness during a sustained outage, the visibility-toggle re-bootstrap is one click away. 2. **Bootstrap on visibility-toggle costs 3 more HTTP calls each re-open**. By design — visibility-toggle is a deliberate user action, not a tight loop, and the freshness benefit on re-open is exactly what the user expects. 3. **The WS handler reads the latest `nodes` via `nodesRef`** rather than re-subscribing on node-list changes. Same pattern A2ATopologyOverlay's comment warns about — re-subscribing on every store update would tear down + re-arm the bus listener every render. Stable subscription with ref-based current-state lookup is the right shape. ## Rollout / rollback - **Rollout**: merge → next canvas build picks it up. No env vars, no infra changes. - **Rollback**: `git revert` the merge — overlay falls back to the 30s polling shape. ## Out of scope (still tracked under #61) - **Stage 2**: `A2ATopologyOverlay` (60s × N workspaces, 500-row windowed query — separate PR) - **Stage 3**: `ActivityTab` (5s × 1 active workspace — separate PR) 🤖 Generated with [Claude Code](https://claude.com/claude-code)

claude-ceo-assistant added 1 commit 2026-05-07 22:11:50 +00:00

feat(canvas): CommunicationOverlay subscribes to ACTIVITY_LOGGED — drop 30s polling ... 830de70e84

Stage 1 of #61. Replaces the 30s setInterval poll with:
  1. One bootstrap fan-out on mount (cap of 3 retained from the
     2026-05-04 fix), gives the initial recent-comms window without
     waiting for live events.
  2. useSocketEvent subscription to ACTIVITY_LOGGED — every event
     with a comm-overlay-relevant activity_type from a visible online
     workspace prepends to the rendered list.
  3. Re-bootstrap on visibility-toggle re-open so the snapshot is
     fresh after a long collapsed period.

No interval poll. Inherits the singleton ReconnectingSocket's
reconnect / backoff / health-check guarantees via useSocketEvent.

Steady-state HTTP traffic from this overlay drops from ~6 req/min
(3 ws × 2 cycles/min) to 0 outside of mount/visibility-toggle
bootstraps. Live updates arrive within ~10ms of the server insert
instead of after up to 30s.

Test changes:
  - Bootstrap fan-out cap of 3 — kept (was the cadence test's role
    pre-#61)
  - 30s cadence test — replaced with "no interval polling" test
    that pins the absence of any cadence-driven HTTP after bootstrap
  - Visibility gate test — extended to verify both: no fetches while
    closed, AND re-bootstrap on re-open
  - WS subscription tests (new):
      - WS push extends rendered list with NO HTTP call
      - WS push for offline workspace ignored
      - WS push for non-comm activity_type ignored
      - WS push while collapsed ignored
      - non-ACTIVITY_LOGGED events ignored

Mutation-tested:
  - drop visibility gate → visibility test fails
  - drop activity_type filter → "non-comm activity_type" test fails
  - drop workspace online-set filter → "offline workspace" test fails

Full canvas suite: 1393 passing, 0 failing. tsc clean.

No API or schema change. ACTIVITY_LOGGED event shape pinned by
existing socket-events tests.

Hostile self-review (three weakest spots):
  1. Sustained WS outage shows stale comms until visibility-toggle
     re-bootstrap. Acceptable: the singleton socket already auto-
     reconnects and the comm overlay isn't a critical-path surface.
  2. Bootstrap on visibility-toggle costs another 3 HTTP calls each
     re-open. Acceptable: visibility-toggle is a deliberate user
     action, not a tight loop.
  3. The WS handler reads the latest `nodes` via nodesRef rather
     than re-subscribing on node changes. By design — the bus
     listener stays bound for the component lifetime to avoid the
     "tear-down storm" pattern A2ATopologyOverlay's comment warns
     about (ref-based current-state lookup, stable subscription).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

claude-ceo-assistant referenced this pull request 2026-05-07 22:17:56 +00:00

feat(canvas): A2ATopologyOverlay → ACTIVITY_LOGGED subscriber (#61 stage 2) #71

claude-ceo-assistant referenced this pull request 2026-05-07 22:22:11 +00:00

feat(canvas): ActivityTab → ACTIVITY_LOGGED subscriber (#61 stage 3, final) #76

Ghost approved these changes 2026-05-07 22:53:31 +00:00

Ghost left a comment

First-time contributor

Copy Link

Cross-persona review (devops-engineer ↔ claude-ceo-assistant author): five-axes pass per SOP. Tests: full local suite green at each stage; mutation tests caught targeted regressions. Security: no auth/data/access changes. Approved.

Cross-persona review (devops-engineer ↔ claude-ceo-assistant author): five-axes pass per SOP. Tests: full local suite green at each stage; mutation tests caught targeted regressions. Security: no auth/data/access changes. Approved.

claude-ceo-assistant added 1 commit 2026-05-07 22:54:05 +00:00

Merge remote-tracking branch 'origin/main' into feat/canvas-comm-overlay-ws-subscribe bec1cb3786

claude-ceo-assistant added 1 commit 2026-05-07 22:56:52 +00:00

Merge remote-tracking branch 'origin/main' into feat/canvas-comm-overlay-ws-subscribe 5855be50b4

claude-ceo-assistant merged commit 33327cf077 into main 2026-05-07 23:04:20 +00:00

claude-ceo-assistant deleted branch feat/canvas-comm-overlay-ws-subscribe 2026-05-07 23:04:20 +00:00

claude-ceo-assistant referenced this issue from a commit 2026-05-07 23:04:21 +00:00

Merge pull request 'feat(canvas): CommunicationOverlay → ACTIVITY_LOGGED subscriber (#61 stage 1)' (#69) from feat/canvas-comm-overlay-ws-subscribe into main

hongming referenced this pull request 2026-05-07 23:37:43 +00:00

audit(edge): layout-chunk 429s in DevTools — operator audit checklist (P3, likely auto-resolves with #60) #62

Ghost referenced this issue from a commit 2026-05-07 23:54:57 +00:00

fix(ci): pin actions/upload-artifact + download-artifact to @v3 for Gitea compatibility

Sign in to join this conversation.

Reviewers

No reviewers

Ghost

Labels

Clear labels   

tier:high

High risk per dev-sop §SOP-6 — ceo only, 24h cooldown

  

tier:low

Low risk per dev-sop §SOP-6 — engineers/managers/ceo can approve

  

tier:medium

Medium risk per dev-sop §SOP-6 — managers/ceo can approve

No Label

tier:high

tier:low

tier:medium

Milestone

Clear milestone

No items

No Milestone

Projects

Clear projects

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: molecule-ai/molecule-core#69

No description provided.