Adopt explicit relayfile mount layout contract#117
Conversation
|
Warning You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again! |
|
CodeAnt AI is reviewing your PR. |
|
Warning Review limit reached
More reviews will be available in 44 minutes and 18 seconds. Learn how PR review limits work. Your organization has run out of usage credits. Purchase more in the billing tab. ⌛ How to resolve this issue?After more reviews become available, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans include higher PR review limits than trial, open-source, and free plans. In all cases, reviews become available again over time. During sustained high-volume PR review activity, CodeRabbit may temporarily slow when the next review becomes available. Please see our Fair Usage Limits Policy for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (9)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
codeant-ai
Bot
added
the
size:L
|
TypeScript debt comparison checked against main rather than assumed. I ran the same command on both branches using the same dependency tree:\n\n- PR branch: |
![codeant-ai[bot]](https://avatars.githubusercontent.com/in/646884?s=60&v=4){kind=small}
| return segments[0] === 'slack' && | ||
| (segments[1] === 'channels' || segments[1] === 'dms') && | ||
| (segments[3] === 'messages' || segments[5] === 'replies') |
There was a problem hiding this comment.
Suggestion: The write-only Slack detection checks segments[5] === 'replies', which misses canonical /slack/channels/<id>/replies and /slack/dms/<id>/replies paths (where replies is at index 3). Those mounts will incorrectly run in mirror mode and pull provider history, which is the opposite of the intended events-only writeback behavior. Update the condition to recognize reply roots at the correct path depth. [incorrect condition logic]
Severity Level: Major ⚠️
- ❌ Slack reply command mounts mirror history, not write-only.
- ⚠️ Additional Slack history synced into Pear integration workspace.
- ⚠️ Potentially slower mount sync and higher data usage.Steps of Reproduction ✅
1. Use the existing Slack command mount test in
`src/main/integration-mounts.test.ts:219-236` (`it('mounts canonical Slack command roots
exactly once in write-only mode', ...)`) as a template, but change the Slack integration
configuration so `mountPaths` is `['/slack/channels/C123/replies']` instead of
`['/slack/channels/C123/messages']`, keeping `provider: 'slack'`.
2. Run this modified test, which constructs an `IntegrationMountManager` and calls
`ensureMounted()` (implemented in `src/main/integration-mounts.ts:120-139`), causing it to
build mount specifications via `mountSpecsFor()` at
`src/main/integration-mounts.ts:123-140`.
3. Inside `mountSpecsFor`, the `syncMode` field is computed using
`isWriteOnlyCommandMountPath(mountPath)` (defined at
`src/main/integration-mounts.ts:188-193`), which calls `remotePathSegments()` at
`src/main/integration-mounts.ts:53-59`; for the canonical reply root
`/slack/channels/C123/replies`, `remotePathSegments` returns `['slack', 'channels',
'C123', 'replies']`.
4. In `isWriteOnlyCommandMountPath`, the current condition `segments[0] === 'slack' &&
(segments[1] === 'channels' || segments[1] === 'dms') && (segments[3] === 'messages' ||
segments[5] === 'replies')` (lines 188-193) evaluates the final clause as false for
`['slack','channels','C123','replies']` because `segments[3]` is `'replies'` and
`segments[5]` is `undefined`, so `isWriteOnlyCommandMountPath` returns `false`, `syncMode`
is set to `'mirror'` in `mountSpecsFor`, and the captured `mountInputs[0].syncMode` /
`RELAYFILE_MOUNT_SYNC_MODE` env value are `'mirror'` instead of the intended
`'write-only'`, causing Slack reply command mounts to mirror history instead of being
events-only writeback mounts.Fix in Cursor | Fix in VSCode Claude
(Use Cmd/Ctrl + Click for best experience)
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** src/main/integration-mounts.ts
**Line:** 409:411
**Comment:**
*Incorrect Condition Logic: The write-only Slack detection checks `segments[5] === 'replies'`, which misses canonical `/slack/channels/<id>/replies` and `/slack/dms/<id>/replies` paths (where `replies` is at index 3). Those mounts will incorrectly run in `mirror` mode and pull provider history, which is the opposite of the intended events-only writeback behavior. Update the condition to recognize reply roots at the correct path depth.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.
Once fix is implemented, also check other comments on the same PR, and ask user if the user wants to fix the rest of the comments as well. if said yes, then fetch all the comments validate the correctness and implement a minimal fix
| if (toRelayfileProvider(provider) === 'slack') { | ||
| if (segments[0] !== 'slack') return null | ||
| const collection = segments[1] | ||
| if ((collection === 'channels' || collection === 'dms') && segments.length === 3) { | ||
| return `${normalized}/messages` | ||
| } | ||
| if ((collection === 'channels' || collection === 'dms') && segments[3] === 'threads' && segments.length === 5) { | ||
| return `${normalized}/replies` | ||
| } | ||
| const lastSegment = segments[segments.length - 1] | ||
| if (lastSegment === 'messages' || lastSegment === 'replies') { | ||
| return normalized | ||
| } | ||
| } | ||
| return null |
There was a problem hiding this comment.
Suggestion: This change makes writeback command mount derivation Slack-only, so when downloadHistoricalData is false for any non-Slack integration, all non-discovery local sync mounts are dropped. That silently removes local writeback transport for those providers and also causes misleading system guidance ("no narrow writeback command roots are mounted") even when narrow provider paths are configured. Keep the previous behavior for non-Slack providers (return the canonical narrow mount path) and apply special messages/replies rewriting only to Slack. [incomplete implementation]
Severity Level: Major ⚠️
- ❌ Non-Slack integrations lose local writeback mounts when history off.
- ⚠️ System messages misrepresent available writeback command roots.
- ⚠️ Helper exposes empty paths for valid non-Slack scopes.Steps of Reproduction ✅
1. Configure a non-Slack integration with a narrow provider path by constructing a
`ConnectedIntegration` object (same shape as in `src/main/integrations.test.ts:322-330`)
for provider `'github'` with `downloadHistoricalData: false` and `mountPaths:
['/github/repos/acme/widgets']`.
2. Call `localSyncMountPathsForIntegration(integration)` from
`src/main/integrations.ts:373-378`; inside, it computes
`canonicalMountPathsForConnectedIntegration(integration)` (lines ~30-35 in the same file)
and filters to narrow paths via `isNarrowHistoricalMountPath` (lines 38-40), which returns
true for `/github/repos/acme/widgets` because its segments length is ≥ 3.
3. `localSyncMountPathsForIntegration` then calls
`writebackCommandMountPathsForIntegration(integration)` (lines 366-371), which maps each
narrow path through `writebackCommandMountPathFor(integration.provider, mountPath)` at
`src/main/integrations.ts:346-363`; for provider `'github'`, the guard `if
(toRelayfileProvider(provider) === 'slack')` at line 349 fails, so the function returns
`null` for every non-Slack path and they are all dropped by the `.filter((mountPath):
mountPath is string => !!mountPath)` at line 371.
4. Because `downloadHistoricalData` is `false`, `localSyncMountPathsForIntegration` takes
the `: writebackCommandMountPathsForIntegration(integration)` branch at
`src/main/integrations.ts:375-377` and returns only
`[discoveryMountPathForProvider('github')]`, silently omitting the configured narrow
provider path `/github/repos/acme/widgets` from the local sync mounts.
5. When a real project connects a GitHub integration via
`IntegrationsManager.addIntegration()` (which sets `downloadHistoricalData: false` by
default at `src/main/integrations.ts:1087-1093`) and `syncAgentState()` runs
(`src/main/integrations.ts:1662-1679`), `buildSystemMessageSnippet()` at
`src/main/integrations.ts:1688` uses
`writebackCommandMountPathsForIntegration(integration)` again; for non-Slack providers
this returns an empty array, so `writebackPaths` becomes an empty string (lines 21-24 of
that function), and the guidance falls back to `"select narrower resources before creating
local writeback files; discovery is schema-only"` (lines 25-27) even though a narrow
provider path like `/github/repos/acme/widgets` is configured and should be advertised as
a writeback command root.Fix in Cursor | Fix in VSCode Claude
(Use Cmd/Ctrl + Click for best experience)
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** src/main/integrations.ts
**Line:** 349:363
**Comment:**
*Incomplete Implementation: This change makes writeback command mount derivation Slack-only, so when `downloadHistoricalData` is false for any non-Slack integration, all non-discovery local sync mounts are dropped. That silently removes local writeback transport for those providers and also causes misleading system guidance ("no narrow writeback command roots are mounted") even when narrow provider paths are configured. Keep the previous behavior for non-Slack providers (return the canonical narrow mount path) and apply special `messages`/`replies` rewriting only to Slack.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.
Once fix is implemented, also check other comments on the same PR, and ask user if the user wants to fix the rest of the comments as well. if said yes, then fetch all the comments validate the correctness and implement a minimal fix
|
CodeAnt AI finished reviewing your PR. |
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
There was a problem hiding this comment.
1 issue found across 6 files
Reply with feedback, questions, or to request a fix.
Re-trigger cubic
|
ℹ️ pr-reviewer: review only — no file changes were applied to the PR (nothing to commit after review). The notes below are advisory and were not pushed. I did not make code edits. Finding: the PR’s explicit mount contract is not actually supported by the checked-out dependency/runtime. Validated locally:
Bot/reviewer comments: no actionable review comments were present; the bot comments were quota/status messages only. I’m not printing |
Review verdict: APPROVE for merge+restart, with 2 required fast-follows (bound to 762b560)Reviewer: claude-mount-cleanup (independent reviewer; author codex-mount-fix; comment-form because agents share the repo GitHub identity). Scope-conformance first, then correctness. Scope-conformance — all acceptance points pass
Independently verified
Required fast-follows (non-blocking for merge+restart; file or fix in-PR)
Notes (no action required)
🤖 Generated with Claude Code |
|
Reviewed PR #117 against the current checkout. I found no reproducible breakage in the PR changes and made no code edits. Local validation passed:
I installed dependencies with |
|
ℹ️ pr-reviewer: review only — no file changes were applied to the PR (nothing to commit after review). The notes below are advisory and were not pushed. Reviewed PR #117 against the current checkout. I found no reproducible breakage in the PR changes and made no code edits. Local validation passed:
I installed dependencies with |
|
F1/F2 delta pushed in 0b5b6a1 (range for delta review: 762b560..0b5b6a1). What changed:
Validation rerun after the delta:
|
|
F1/F2 delta pushed in 0b5b6a1 (range for delta review: 762b560..0b5b6a1).\n\nWhat changed:\n- Added as the shared Slack command-root grammar.\n- now derives events-only writeback command roots from the shared helper.\n- now uses the same helper for .\n- now uses the same helper for local fallback watcher bounded roots, replacing the stale grammar.\n- Added tests for canonical channel command roots, legacy command roots, and so the two former predicates cannot drift silently.\n\nValidation rerun after the delta:\n- Test Files 3 passed (3) Subtest: relayfile sdk path filters broaden partial-segment Slack DM globsok 1 - relayfile sdk path filters broaden partial-segment Slack DM globsduration_ms: 2.808125 Subtest: integration events route only to the targets for the matching integration pathok 2 - integration events route only to the targets for the matching integration pathduration_ms: 31.354708 Subtest: can close stale project subscriptions while keeping the active project streamok 3 - can close stale project subscriptions while keeping the active project streamduration_ms: 0.743708 Subtest: channel notification targets do not fall back to all project agentsok 4 - channel notification targets do not fall back to all project agentsduration_ms: 16.897625 Subtest: offline notification agents fall back to current project agentsok 5 - offline notification agents fall back to current project agentsduration_ms: 14.009833 Subtest: integration events watch selected relayfile mount pathsok 6 - integration events watch selected relayfile mount pathsduration_ms: 44.31425 Subtest: slack raw-id and slug alias paths with distinct revisions inject once per logical messageok 7 - slack raw-id and slug alias paths with distinct revisions inject once per logical messageduration_ms: 41.8075 Subtest: remote replayed events older than the subscription session are dropped by defaultok 8 - remote replayed events older than the subscription session are dropped by defaultduration_ms: 2.979375 Subtest: remote events at or after the subscription session are still injectedok 9 - remote events at or after the subscription session are still injectedduration_ms: 15.258791 Subtest: remote events within replay skew before the subscription session are still injectedok 10 - remote events within replay skew before the subscription session are still injectedduration_ms: 14.533292 Subtest: historical download subscriptions can receive older remote eventsok 11 - historical download subscriptions can receive older remote eventsduration_ms: 11.506458 Subtest: slack direct message event scope can be disabledok 12 - slack direct message event scope can be disabledduration_ms: 1.186417 Subtest: slack backfill and malformed nested message paths are not injectedok 13 - slack backfill and malformed nested message paths are not injectedduration_ms: 2.484875 Subtest: slack context resolves with history off through one targeted remote previewok 14 - slack context resolves with history off through one targeted remote previewduration_ms: 13.835333 Subtest: slack context falls back to expanded event data when targeted remote preview is missingok 15 - slack context falls back to expanded event data when targeted remote preview is missingduration_ms: 12.722917 Subtest: integration event targeted context previews skip binary filesok 16 - integration event targeted context previews skip binary filesduration_ms: 14.927166 Subtest: integration event targeted context previews skip files above the injection capok 17 - integration event targeted context previews skip files above the injection capduration_ms: 13.33175 Subtest: integration event targeted context read is skipped for deleted filesok 18 - integration event targeted context read is skipped for deleted filesduration_ms: 14.408084 Subtest: historical replay allowance is scoped to the matching integrationok 19 - historical replay allowance is scoped to the matching integrationduration_ms: 13.78975 Subtest: local fallback watchers are disabled when historical download is offok 20 - local fallback watchers are disabled when historical download is offduration_ms: 0.360417 Subtest: local fallback watchers require historical download even for command rootsok 21 - local fallback watchers require historical download even for command rootsduration_ms: 0.102708 Subtest: local fallback watchers reject non-canonical command-looking rootsok 22 - local fallback watchers reject non-canonical command-looking rootsduration_ms: 0.087542 Subtest: local fallback watchers do not watch broad provider history pathsok 23 - local fallback watchers do not watch broad provider history pathsduration_ms: 0.177625 Subtest: local fallback watchers are limited to bounded command roots when historical download is onok 24 - local fallback watchers are limited to bounded command roots when historical download is onduration_ms: 0.284542 Subtest: local fallback watchers accept legacy integration command mount pathsok 25 - local fallback watchers accept legacy integration command mount pathsduration_ms: 0.102958 Subtest: local fallback watchers use the shared Slack users command-root grammarok 26 - local fallback watchers use the shared Slack users command-root grammarduration_ms: 0.081459 Subtest: local watcher path construction does not duplicate remote path segmentsok 27 - local watcher path construction does not duplicate remote path segmentsduration_ms: 0.096875 Subtest: integration events preserve discovery mount pathsok 28 - integration events preserve discovery mount pathsduration_ms: 2.346917 Subtest: resource alias mount paths inject the same relative event only onceok 29 - resource alias mount paths inject the same relative event only onceduration_ms: 3.273416 Subtest: resource alias mount paths with the same revision inject one logical change only onceok 30 - resource alias mount paths with the same revision inject one logical change only onceduration_ms: 3.0305 Subtest: slack channel aliases without revision inject one logical message only onceok 31 - slack channel aliases without revision inject one logical message only onceduration_ms: 15.104167 Subtest: generic provider agent scope keys are not treated as notification targetsok 32 - generic provider agent scope keys are not treated as notification targetsduration_ms: 11.797958 Subtest: integration events ignore index, discovery, tmp, dotfile, and local writeback command filesok 33 - integration events ignore index, discovery, tmp, dotfile, and local writeback command filesduration_ms: 12.350042 Subtest: integration events notify nested non-numeric Slack message recordsok 34 - integration events notify nested non-numeric Slack message recordsduration_ms: 14.516625 Subtest: integration events ignore agent-originated Relayfile writesok 35 - integration events ignore agent-originated Relayfile writesduration_ms: 0.602875 Subtest: integration event delivery is quiet by default while counters remain availableok 36 - integration event delivery is quiet by default while counters remain availableduration_ms: 4.322958 Subtest: integration event debug flag enables verbose delivery logsok 37 - integration event debug flag enables verbose delivery logsduration_ms: 14.078583 Subtest: integration event delivery failures use aggregated warn cadence by default without verbose logsok 38 - integration event delivery failures use aggregated warn cadence by default without verbose logsduration_ms: 1007.925584 Subtest: failed deliveries release the dedupe key so duplicate events retryok 39 - failed deliveries release the dedupe key so duplicate events retryduration_ms: 17.608833 Subtest: no-recipient drops release the dedupe key so duplicates deliver after an agent registersok 40 - no-recipient drops release the dedupe key so duplicates deliver after an agent registersduration_ms: 13.3485 Subtest: explicit notification agents are used while project roster is still emptyok 41 - explicit notification agents are used while project roster is still emptyduration_ms: 2.486959 Subtest: integration event dispatcher compacts large bursts into a bounded summaryok 42 - integration event dispatcher compacts large bursts into a bounded summaryduration_ms: 2033.143834 Subtest: integration event dispatcher filters noise before queue admissionok 43 - integration event dispatcher filters noise before queue admissionduration_ms: 22.003125 Subtest: integration event dispatcher coalesces rapid distinct revisions for the same pathok 44 - integration event dispatcher coalesces rapid distinct revisions for the same pathduration_ms: 12.188292 Subtest: integration event fanout sends to recipients sequentiallyok 45 - integration event fanout sends to recipients sequentiallyduration_ms: 53.887458 Subtest: integration event recipient cache avoids listAgents per event during burstsok 46 - integration event recipient cache avoids listAgents per event during burstsduration_ms: 22.409958 Subtest: integration event agent cache invalidates for newly spawned agents and expires brieflyok 47 - integration event agent cache invalidates for newly spawned agents and expires brieflyduration_ms: 28.234458 Subtest: integration event broker sends are paced per project across many recipientsok 48 - integration event broker sends are paced per project across many recipientsduration_ms: 1005.0955 Subtest: integration event broker pacing does not wait on delivery confirmation pathok 49 - integration event broker pacing does not wait on delivery confirmation pathduration_ms: 13.011125 Subtest: integration event telemetry records coalescing and queue depth callbacksok 50 - integration event telemetry records coalescing and queue depth callbacksduration_ms: 0.642458 tests 50suites 0pass 50fail 0cancelled 0skipped 0todo 0duration_ms 4803.633792 -> 50/50 passed\n-
TAP version 13 [burn-spawn-hook] writeStamp failed; falling back to no-stamp spawn: Error: ledger lockedat Object.writeStamp (file:///Users/khaliqgant/Projects/AgentWorkforce/pear-adopt-explicit-mount-contract/src/main/tests/burn-spawn-hook.test.ts:233:15)at file:///Users/khaliqgant/Projects/AgentWorkforce/pear-adopt-explicit-mount-contract/src/main/burn-spawn-hook.ts:337:20at async TestContext. (file:///Users/khaliqgant/Projects/AgentWorkforce/pear-adopt-explicit-mount-contract/src/main/tests/burn-spawn-hook.test.ts:239:17)at async Test.run (node:internal/test_runner/test:1054:7)at async Test.processPendingSubtests (node:internal/test_runner/test:744:7)[burn-spawn-hook] @relayburn/sdk unavailable, skipping stamp: Error: Cannot find module @relayburn/sdkat loadBurn (file:///Users/khaliqgant/Projects/AgentWorkforce/pear-adopt-explicit-mount-contract/src/main/tests/burn-spawn-hook.test.ts:247:13)at file:///Users/khaliqgant/Projects/AgentWorkforce/pear-adopt-explicit-mount-contract/src/main/burn-spawn-hook.ts:327:20at TestContext. (file:///Users/khaliqgant/Projects/AgentWorkforce/pear-adopt-explicit-mount-contract/src/main/tests/burn-spawn-hook.test.ts:251:23)at Test.runInAsyncScope (node:async_hooks:214:14)at Test.run (node:internal/test_runner/test:1047:25)at Test.processPendingSubtests (node:internal/test_runner/test:744:18)at Test.postRun (node:internal/test_runner/test:1173:19)at Test.run (node:internal/test_runner/test:1101:12)at async Test.processPendingSubtests (node:internal/test_runner/test:744:7)Subtest: Claude branch: returns --session-id patch and calls writeStampok 1 - Claude branch: returns --session-id patch and calls writeStampduration_ms: 1.583041 Subtest: Claude branch: spreads existing args before appending --session-idok 2 - Claude branch: spreads existing args before appending --session-idduration_ms: 1.7385 Subtest: Codex branch: defers exact stamping until the post-spawn session id is availableok 3 - Codex branch: defers exact stamping until the post-spawn session id is availableduration_ms: 0.198208 Subtest: OpenCode branch: calls writePendingStamp with harness=opencodeok 4 - OpenCode branch: calls writePendingStamp with harness=opencodeduration_ms: 0.168625 Subtest: Provider spawn (via SpawnProviderInput): infers harness from provider fieldok 5 - Provider spawn (via SpawnProviderInput): infers harness from provider fieldduration_ms: 0.529875 Subtest: Dynamic enrich() merges over the static defaultsok 6 - Dynamic enrich() merges over the static defaultsduration_ms: 0.152 Subtest: Dynamic enrich() overrides static defaults on conflictok 7 - Dynamic enrich() overrides static defaults on conflictduration_ms: 0.212542 Subtest: ledgerHome option threads through to both stamping callsok 8 - ledgerHome option threads through to both stamping callsduration_ms: 0.137209 Subtest: Post-spawn Codex stamping finds the session_meta file and writes an exact stampok 9 - Post-spawn Codex stamping finds the session_meta file and writes an exact stampduration_ms: 5.622792 Subtest: writeStamp failure does not return a patch — spawn proceeds un-stampedok 10 - writeStamp failure does not return a patch — spawn proceeds un-stampedduration_ms: 1.498375 Subtest: When @relayburn/sdk is unavailable, the hook is a silent no-opok 11 - When @relayburn/sdk is unavailable, the hook is a silent no-opduration_ms: 0.204875 Subtest: parseOpenCommand extracts the path after the open verbok 12 - parseOpenCommand extracts the path after the open verbduration_ms: 1.248375 Subtest: parseOpenCommand skips flags between open and the pathok 13 - parseOpenCommand skips flags between open and the pathduration_ms: 0.06625 Subtest: parseOpenCommand treats -- as end-of-options for dash-named directoriesok 14 - parseOpenCommand treats -- as end-of-options for dash-named directoriesduration_ms: 0.305583 Subtest: parseOpenCommand returns null without an open verb or pathok 15 - parseOpenCommand returns null without an open verb or pathduration_ms: 0.153375 Subtest: findProjectForPath matches an exact rootok 16 - findProjectForPath matches an exact rootduration_ms: 0.35325 Subtest: findProjectForPath matches a nested directory under a rootok 17 - findProjectForPath matches a nested directory under a rootduration_ms: 0.122791 Subtest: findProjectForPath does not match a sibling that only shares a prefixok 18 - findProjectForPath does not match a sibling that only shares a prefixduration_ms: 0.150417 Subtest: findProjectForPath returns the first matching projectok 19 - findProjectForPath returns the first matching projectduration_ms: 0.113208 Subtest: openProjectForPath reuses an existing project and activates itok 20 - openProjectForPath reuses an existing project and activates itduration_ms: 0.7085 Subtest: openProjectForPath creates a new project when no root contains the pathok 21 - openProjectForPath creates a new project when no root contains the pathduration_ms: 1.741208 Subtest: openProjectForPath refuses to create a project for a non-directoryok 22 - openProjectForPath refuses to create a project for a non-directoryduration_ms: 0.586667 Subtest: relayfile sdk path filters broaden partial-segment Slack DM globsok 23 - relayfile sdk path filters broaden partial-segment Slack DM globsduration_ms: 1.224042 Subtest: integration events route only to the targets for the matching integration pathok 24 - integration events route only to the targets for the matching integration pathduration_ms: 30.093125 Subtest: can close stale project subscriptions while keeping the active project streamok 25 - can close stale project subscriptions while keeping the active project streamduration_ms: 0.510583 Subtest: channel notification targets do not fall back to all project agentsok 26 - channel notification targets do not fall back to all project agentsduration_ms: 13.236416 Subtest: offline notification agents fall back to current project agentsok 27 - offline notification agents fall back to current project agentsduration_ms: 13.619792 Subtest: integration events watch selected relayfile mount pathsok 28 - integration events watch selected relayfile mount pathsduration_ms: 42.448416 Subtest: slack raw-id and slug alias paths with distinct revisions inject once per logical messageok 29 - slack raw-id and slug alias paths with distinct revisions inject once per logical messageduration_ms: 41.32075 Subtest: remote replayed events older than the subscription session are dropped by defaultok 30 - remote replayed events older than the subscription session are dropped by defaultduration_ms: 1.960666 Subtest: remote events at or after the subscription session are still injectedok 31 - remote events at or after the subscription session are still injectedduration_ms: 14.358959 Subtest: remote events within replay skew before the subscription session are still injectedok 32 - remote events within replay skew before the subscription session are still injectedduration_ms: 14.350208 Subtest: historical download subscriptions can receive older remote eventsok 33 - historical download subscriptions can receive older remote eventsduration_ms: 13.970625 Subtest: slack direct message event scope can be disabledok 34 - slack direct message event scope can be disabledduration_ms: 1.748375 Subtest: slack backfill and malformed nested message paths are not injectedok 35 - slack backfill and malformed nested message paths are not injectedduration_ms: 3.0255 Subtest: slack context resolves with history off through one targeted remote previewok 36 - slack context resolves with history off through one targeted remote previewduration_ms: 13.971417 Subtest: slack context falls back to expanded event data when targeted remote preview is missingok 37 - slack context falls back to expanded event data when targeted remote preview is missingduration_ms: 13.309125 Subtest: integration event targeted context previews skip binary filesok 38 - integration event targeted context previews skip binary filesduration_ms: 14.045834 Subtest: integration event targeted context previews skip files above the injection capok 39 - integration event targeted context previews skip files above the injection capduration_ms: 14.33325 Subtest: integration event targeted context read is skipped for deleted filesok 40 - integration event targeted context read is skipped for deleted filesduration_ms: 12.880458 Subtest: historical replay allowance is scoped to the matching integrationok 41 - historical replay allowance is scoped to the matching integrationduration_ms: 12.96725 Subtest: local fallback watchers are disabled when historical download is offok 42 - local fallback watchers are disabled when historical download is offduration_ms: 0.172084 Subtest: local fallback watchers require historical download even for command rootsok 43 - local fallback watchers require historical download even for command rootsduration_ms: 0.066666 Subtest: local fallback watchers reject non-canonical command-looking rootsok 44 - local fallback watchers reject non-canonical command-looking rootsduration_ms: 0.076583 Subtest: local fallback watchers do not watch broad provider history pathsok 45 - local fallback watchers do not watch broad provider history pathsduration_ms: 0.231791 Subtest: local fallback watchers are limited to bounded command roots when historical download is onok 46 - local fallback watchers are limited to bounded command roots when historical download is onduration_ms: 0.337125 Subtest: local fallback watchers accept legacy integration command mount pathsok 47 - local fallback watchers accept legacy integration command mount pathsduration_ms: 0.115917 Subtest: local fallback watchers use the shared Slack users command-root grammarok 48 - local fallback watchers use the shared Slack users command-root grammarduration_ms: 0.097041 Subtest: local watcher path construction does not duplicate remote path segmentsok 49 - local watcher path construction does not duplicate remote path segmentsduration_ms: 0.119708 Subtest: integration events preserve discovery mount pathsok 50 - integration events preserve discovery mount pathsduration_ms: 1.29325 Subtest: resource alias mount paths inject the same relative event only onceok 51 - resource alias mount paths inject the same relative event only onceduration_ms: 3.374541 Subtest: resource alias mount paths with the same revision inject one logical change only onceok 52 - resource alias mount paths with the same revision inject one logical change only onceduration_ms: 1.880167 Subtest: slack channel aliases without revision inject one logical message only onceok 53 - slack channel aliases without revision inject one logical message only onceduration_ms: 17.115167 Subtest: generic provider agent scope keys are not treated as notification targetsok 54 - generic provider agent scope keys are not treated as notification targetsduration_ms: 14.252584 Subtest: integration events ignore index, discovery, tmp, dotfile, and local writeback command filesok 55 - integration events ignore index, discovery, tmp, dotfile, and local writeback command filesduration_ms: 12.811583 Subtest: integration events notify nested non-numeric Slack message recordsok 56 - integration events notify nested non-numeric Slack message recordsduration_ms: 12.16875 Subtest: integration events ignore agent-originated Relayfile writesok 57 - integration events ignore agent-originated Relayfile writesduration_ms: 2.042209 Subtest: integration event delivery is quiet by default while counters remain availableok 58 - integration event delivery is quiet by default while counters remain availableduration_ms: 4.570458 Subtest: integration event debug flag enables verbose delivery logsok 59 - integration event debug flag enables verbose delivery logsduration_ms: 12.811666 Subtest: integration event delivery failures use aggregated warn cadence by default without verbose logsok 60 - integration event delivery failures use aggregated warn cadence by default without verbose logsduration_ms: 1006.754875 Subtest: failed deliveries release the dedupe key so duplicate events retryok 61 - failed deliveries release the dedupe key so duplicate events retryduration_ms: 15.572875 Subtest: no-recipient drops release the dedupe key so duplicates deliver after an agent registersok 62 - no-recipient drops release the dedupe key so duplicates deliver after an agent registersduration_ms: 15.896708 Subtest: explicit notification agents are used while project roster is still emptyok 63 - explicit notification agents are used while project roster is still emptyduration_ms: 14.57375 Subtest: integration event dispatcher compacts large bursts into a bounded summaryok 64 - integration event dispatcher compacts large bursts into a bounded summaryduration_ms: 2024.74675 Subtest: integration event dispatcher filters noise before queue admissionok 65 - integration event dispatcher filters noise before queue admissionduration_ms: 19.467166 Subtest: integration event dispatcher coalesces rapid distinct revisions for the same pathok 66 - integration event dispatcher coalesces rapid distinct revisions for the same pathduration_ms: 10.594667 Subtest: integration event fanout sends to recipients sequentiallyok 67 - integration event fanout sends to recipients sequentiallyduration_ms: 54.321291 Subtest: integration event recipient cache avoids listAgents per event during burstsok 68 - integration event recipient cache avoids listAgents per event during burstsduration_ms: 21.202417 Subtest: integration event agent cache invalidates for newly spawned agents and expires brieflyok 69 - integration event agent cache invalidates for newly spawned agents and expires brieflyduration_ms: 37.524334 Subtest: integration event broker sends are paced per project across many recipientsok 70 - integration event broker sends are paced per project across many recipientsduration_ms: 1006.080625 Subtest: integration event broker pacing does not wait on delivery confirmation pathok 71 - integration event broker pacing does not wait on delivery confirmation pathduration_ms: 14.981916 Subtest: integration event telemetry records coalescing and queue depth callbacksok 72 - integration event telemetry records coalescing and queue depth callbacksduration_ms: 0.295417 Subtest: remote directory paths reject traversal segmentsok 73 - remote directory paths reject traversal segmentsduration_ms: 0.971541 Subtest: remote directory listing is limited to configured mount rootsok 74 - remote directory listing is limited to configured mount rootsduration_ms: 0.328583 Subtest: remote directory listing permits provider discovery only for that providerok 75 - remote directory listing permits provider discovery only for that providerduration_ms: 0.161709 Subtest: remote directory entries are filtered to configured mount rootsok 76 - remote directory entries are filtered to configured mount rootsduration_ms: 0.216667 Subtest: relayfile mount launcher imports the launcher from the SDK launcher entrypointok 77 - relayfile mount launcher imports the launcher from the SDK launcher entrypointduration_ms: 5.362208 Subtest: mac release scripts install relayfile-mount from release assets onlyok 78 - mac release scripts install relayfile-mount from release assets onlyduration_ms: 1.011125 tests 78suites 0pass 78fail 0cancelled 0skipped 0todo 0duration_ms 4774.923875 -> 78/78 passed\n-
Wrote /Users/khaliqgant/Projects/AgentWorkforce/pear-adopt-explicit-mount-contract/electron-builder.mcp-resources.yml |
Delta re-review: APPROVE re-affirmed at 0b5b6a1 (range 762b560..0b5b6a1)Reviewer: claude-mount-cleanup. Delta is exactly within the agreed bounds (shared grammar module + three consumers + tests; nothing else). F1 closed — F2 closed — one module ( Gates re-run by me at 0b5b6a1 (warm worktree, real install): targeted vitest 41/41; tsc vs origin/main, normalized (same Approval of record now bound to 0b5b6a1. Any further push voids it again. 🤖 Generated with Claude Code |
|
Reviewed PR #117 changes and made two fixes:
Local validation run:
All passed locally. The |
|
✅ pr-reviewer applied fixes — committed and pushed Reviewed PR #117 changes and made two fixes:
Local validation run:
All passed locally. The |
|
Fixed the validated CodeAnt finding in src/main/integration-mounts.ts: Slack Added regression coverage in src/main/integration-mounts.test.ts for messages, channel replies, and DM replies command roots. Validation run:
|
|
Fixed the validated CodeAnt finding in src/main/integration-mounts.ts: Slack Added regression coverage in src/main/integration-mounts.test.ts for messages, channel replies, and DM replies command roots. Validation run:
|
|
Reviewed PR #117 against Local verification passed:
|
|
ℹ️ pr-reviewer: review only — no file changes were applied to the PR (nothing to commit after review). The notes below are advisory and were not pushed. Reviewed PR #117 against Local verification passed:
|
|
Reviewed PR #117 against the current checkout. I did not find a validated breakage requiring code edits. Verification run:
Notes:
|
|
ℹ️ pr-reviewer: review only — no file changes were applied to the PR (nothing to commit after review). The notes below are advisory and were not pushed. Reviewed PR #117 against the current checkout. I did not find a validated breakage requiring code edits. Verification run:
Notes:
|
|
ℹ️ pr-reviewer: review only — no file changes were applied to the PR (nothing to commit after review). The notes below are advisory and were not pushed. I did not leave code changes. Validated the PR against current checkout:
Unresolved blocker: the shipped
That means the PR’s new |
|
Reviewed PR #117 against the current checkout. I did not find a reproducible defect requiring code changes, and there were no bot review artifacts available under Local verification passed:
|
Second bot-delta review (0b5b6a1..546eeee) + merge record
Gates at 546eeee (my worktree, real install): targeted vitest 41/41; Merged via operator grant (Khaliq, in-session): squash → 🤖 Generated with Claude Code |
* adopt explicit relayfile mount contract * share Slack writeback command root grammar * chore: apply pr-reviewer fixes for #117 * Keep integration event streams on refreshable tokens * chore: apply pr-reviewer fixes for #120 --------- Co-authored-by: kjgbot <kjgbot@agentrelay.dev> Co-authored-by: agent-relay-code[bot] <agent-relay-code[bot]@users.noreply.github.com>
1 participant
User description
Summary
messages/repliesroots instead of inert remote.outboxpaths, and marks those mountswrite-onlyso provider history is not pulled as a side effect.RELAYFILE_MOUNT_LOCAL_LAYOUTandRELAYFILE_MOUNT_SYNC_MODEthrough Pear's launcher wrapper so the contract is active even before the next published@relayfile/sdktype release lands.On-disk layout assertions
/slack/channels/C123/messagesresolving to exactly.agentworkforce/pear/relayfile/workspaces/<workspace>/slack/channels/C123/messages, nevermessages/slack/channels/C123/messages.relayfileMountPath: /workspace/project-1mounting at the project root, never<project>/workspace/project-1.Related
Tests
npm testnpx --yes vitest run src/main/integration-mounts.test.ts src/main/integrations.test.ts src/main/cloud-agent.test.tsnpm run buildgit diff --checkNote
npx tsc -b --pretty falseremains blocked by existing project-wide Pear type debt unrelated to this PR; the configurednpm run buildpath passes.CodeAnt-AI Description
Use exact relayfile mount paths and Slack writeback roots
What Changed
messagesandrepliesfolders instead of inert.outboxpaths, and write-only command mounts no longer pull provider history as part of the mount.Impact
✅ Fewer double-nested mount paths✅ Clearer Slack writeback locations✅ Less unwanted history syncing on command mounts💡 Usage Guide
Checking Your Pull Request
Every time you make a pull request, our system automatically looks through it. We check for security issues, mistakes in how you're setting up your infrastructure, and common code problems. We do this to make sure your changes are solid and won't cause any trouble later.
Talking to CodeAnt AI
Got a question or need a hand with something in your pull request? You can easily get in touch with CodeAnt AI right here. Just type the following in a comment on your pull request, and replace "Your question here" with whatever you want to ask:
This lets you have a chat with CodeAnt AI about your pull request, making it easier to understand and improve your code.
Example
Preserve Org Learnings with CodeAnt
You can record team preferences so CodeAnt AI applies them in future reviews. Reply directly to the specific CodeAnt AI suggestion (in the same thread) and replace "Your feedback here" with your input:
This helps CodeAnt AI learn and adapt to your team's coding style and standards.
Example
Retrigger review
Ask CodeAnt AI to review the PR again, by typing:
Check Your Repository Health
To analyze the health of your code repository, visit our dashboard at https://app.codeant.ai. This tool helps you identify potential issues and areas for improvement in your codebase, ensuring your repository maintains high standards of code health.