improvement(mcp): push-driven settings freshness + lifecycle-audit fixes by waleedlatif1 · Pull Request #5842 · simstudioai/sim

waleedlatif1 · 2026-07-22T06:29:02Z

Summary

Aligns the MCP settings surface with the reference-client refresh model and folds in the fixes from a full end-to-end lifecycle audit. One clean PR covering the settings freshness work plus every audit finding worth acting on.

Push-driven freshness (settings page)

Subscribe the settings page to the live list_changed → SSE push (useMcpToolsEvents) — the pipeline (connection-manager → pubsub → /api/mcp/events → browser EventSource) already existed but was only mounted in the workflow-editor tool picker. Same shared, reference-counted connection. Real-time tool updates on the settings page.
Raise MCP_SERVER_TOOLS_STALE_TIME 30s → 5min (matching the server-side cache TTL) so revisiting leans on push + stored state instead of re-probing. There is no background poll (no refetchInterval); this only affects refetch-on-visit-if-stale — real changes arrive instantly via push.
Resync on reconnect AND on re-subscribe — a dropped/reconnected EventSource, or leaving and returning to the tab (which tears the connection down), could miss a tools_changed event; a per-workspace sseEverSubscribed flag re-syncs on the first open of a re-subscription and on every reconnect, while skipping only the session's first subscribe (queries fetch fresh then).

Lifecycle-audit fixes

Recovered-server status reconciliation. The per-server tools queryFn refreshed the server list only on error; a server recovering via the stale re-probe showed "tools present but row red" until the list independently refetched. A successful probe against a still-failed server now reconciles the status. (+ corrected an overstated keep/drop comment.)
Stale SSRF docstring.validateMcpServerSsrf's docstring described "pin subsequent connections," which no longer holds post validate-at-connect — the returned IP is a policy signal, and redirect/rebind safety comes from per-connect validation + followRedirectsGuarded, not pinning (only the self-hosted private carve-out still pins). Corrected so no one reintroduces a real pin from the text. Doc-only.
OAuth callback timeout consistency. The callback bounded only its post-auth steps; loadOauthRowByState, getSession, the server SELECT, and the provider_errorclearState were unbounded. Now wrapped in timedStep so a wedged DB read surfaces as a labeled timeout.

Validated against Claude Code / LibreChat / OpenCode / VS Code + the MCP spec: discover-once + list_changed push, no timer polling, resync-on-reconnect. Three-pass adversarial lifecycle audit found no security escapes, count leaks, or double-release; the findings above are everything worth acting on.

Type of Change

Improvement + bug fixes (UX freshness, robustness, doc correctness)

Testing

109 tests green across the MCP query, settings, domain-check, and callback-route suites; tsc + biome clean.

Checklist

Code follows project style guidelines
Self-reviewed my changes
Tests added/updated and passing
No new warnings introduced
I confirm that I have read and agree to the terms outlined in the Contributor License Agreement (CLA)

…t over re-probing The list_changed → SSE push pipeline was already wired end-to-end but only mounted in the workflow-editor tool picker. Subscribe the settings MCP page to the same shared, reference-counted EventSource so tool changes reflect in real time there too — the reference-client model (discover once, refresh on push). With push now active on the settings page, raise MCP_SERVER_TOOLS_STALE_TIME from 30s to 5min (matching the server-side cache TTL) so revisiting the page leans on push + stored state instead of re-probing every connected server. There is no background poll (no refetchInterval) — this only affects refetch-on-visit-if-stale; real changes still arrive instantly via push.

vercel · 2026-07-22T06:29:06Z

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment

Project	Deployment	Actions	Updated (UTC)
docs	Skipped		Jul 22, 2026 8:02am

cursor · 2026-07-22T06:29:12Z

PR Summary

Low Risk
Mostly client caching/SSE behavior and OAuth callback timeouts; SSRF doc is comment-only. OAuth timeouts could surface failures earlier on slow DB but improve hang behavior rather than weakening auth.

Overview
MCP tool lists now lean on SSE push instead of frequent re-discovery: useMcpToolsEvents is mounted inside useMcpToolsQuery, so every consumer (settings, tool picker, etc.) shares one reference-counted tools_changed subscription without duplicating useMcpToolsEvents in components like tool-input. MCP_SERVER_TOOLS_STALE_TIME goes from 30s to 5 minutes (aligned with server cache TTL); live updates still come from push, not polling.

useMcpToolsEvents gains reconnect/resubscribe safety: a per-workspace sseEverSubscribed flag plus onopen / pre-open error tracking trigger a workspace invalidate when the client could have missed events (reconnect, tab return, or first connect after an initial SSE error), while skipping a clean first subscription.

The MCP OAuth callback wraps previously unbounded steps (loadOauthRowByState, getSession, server load, provider-error clearState) in timedStep so wedged DB/session work fails with labeled timeouts instead of hanging.

validateMcpServerSsrf docstring is updated to describe the return value as an SSRF policy signal and validate-at-connect guarding, not legacy connection pinning.

Tests add a FakeEventSource, clear shared SSE globals between runs, and count discovery calls explicitly now that the tools query also opens SSE.

^{Reviewed by Cursor Bugbot for commit 1815e21. Configure here.}

greptile-apps · 2026-07-22T06:31:30Z

Greptile Summary

This PR improves MCP settings freshness and lifecycle handling. The main changes are:

Moves the shared SSE subscription into the MCP tools query.
Resyncs tools after reconnects, re-subscriptions, and pre-open connection errors.
Aligns the client tools stale time with the server cache.
Adds timeout bounds to OAuth callback operations.
Updates MCP query tests and clarifies SSRF documentation.

Confidence Score: 5/5

This looks safe to merge.

The initial connection-error path now triggers a workspace resync after recovery.
Reconnect and re-subscription paths invalidate the relevant MCP queries.
The workflow editor retains SSE updates through its existing MCP tools query.
No blocking issues were found in the changed code.

Important Files Changed

Filename	Overview
apps/sim/hooks/queries/mcp.ts	Centralizes SSE freshness handling and covers reconnect, re-subscription, and initial connection-error recovery.
apps/sim/app/workspace/[workspaceId]/w/[workflowId]/components/panel/components/editor/components/sub-block/components/tool-input/tool-input.tsx	Removes the redundant direct SSE subscription now provided by the MCP tools query.
apps/sim/hooks/queries/mcp.test.tsx	Adds EventSource test support and resets shared SSE state between tests.
apps/sim/app/api/mcp/oauth/callback/route.ts	Adds labeled timeout bounds around OAuth callback reads and cleanup.
apps/sim/lib/mcp/domain-check.ts	Clarifies how MCP SSRF validation, guarded connections, and IP pinning interact.

_{Reviews (9): Last reviewed commit: "fix(mcp): drop the success-path status r..." | Re-trigger Greptile}

…'t strand stale tools A dropped/reconnected EventSource may miss a tools_changed event during the gap, which — with a longer stale time — could leave the page showing old tools until a remount or manual refresh. Invalidate the workspace tools on reconnect (never on the first open), the standard resync-on-reconnect pattern for push clients.

waleedlatif1 · 2026-07-22T06:34:19Z

@greptile

waleedlatif1 · 2026-07-22T06:34:20Z

@cursor review

…losed reconcile Leaving the settings tab tears down the shared EventSource; remounting created a new connection whose first open was skipped, so a tools_changed fired while unsubscribed wasn't reconciled (the 5min stale time also won't refetch on remount). Track a per-workspace 'ever subscribed' flag: skip resync only on the session's first subscription (queries fetch fresh then); resync on the first open of any re-subscription and on every reconnect.

waleedlatif1 · 2026-07-22T06:40:32Z

@greptile

waleedlatif1 · 2026-07-22T06:40:33Z

@cursor review

cursor

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit f8fc2ed. Configure here.}

…ry + fix stale comment Lifecycle-audit findings: - The per-server tools queryFn invalidated the server list only on error. A server that recovered via the stale re-probe returned fresh tools while the cached status still showed failed → 'tools present but row red' until the list independently refetched (a window the 5min stale-time widened). Now a successful probe against a server the cache still shows non-connected refreshes the list so the row clears. - Correct the keep/drop comment: tools drop once the stored status leaves 'connected' (disconnected/error), not at a 3-failure threshold.

waleedlatif1 · 2026-07-22T06:56:08Z

@greptile

waleedlatif1 · 2026-07-22T06:56:09Z

@cursor review

cursor

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit 82d32c4. Configure here.}

…uth callback steps Lifecycle-audit follow-ups (merged-code): - validateMcpServerSsrf's return docstring described 'pin subsequent connections', which no longer holds for the public path — the returned IP is a policy signal selecting the validate-at-connect guarded fetch; redirect/rebind safety comes from per-connect validation + followRedirectsGuarded, not pinning (only the self-hosted private carve-out still literally pins). Corrected so a future reader can't reintroduce a real pin from the misleading text. - The callback wrapped only the five post-auth steps in timedStep; the earlier loadOauthRowByState, getSession, and server SELECT (plus the provider_error clearState) were unbounded, contradicting the 'every awaited step bounded' invariant. Wrapped them so a wedged DB read surfaces as a labeled timeout.

waleedlatif1 · 2026-07-22T07:02:59Z

@greptile

waleedlatif1 · 2026-07-22T07:03:01Z

@cursor review

…onnection error If the initial EventSource connection errors before opening (and the initial tools query fails with retry:false), the first successful onopen was skipped, leaving the query stale. Track erroredBeforeOpen so a first open that followed a connection error also resyncs — only a clean first subscription still skips.

waleedlatif1 · 2026-07-22T07:07:21Z

@greptile

waleedlatif1 · 2026-07-22T07:07:23Z

@cursor review

cursor

✅ Bugbot reviewed your changes and found no new issues!

1 issue from previous review remains unresolved.

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit 0b0658c. Configure here.}

The 5min stale time assumed push, but useMcpToolsEvents was only mounted in the settings page and the tool picker — other consumers of the tools query (dynamic args, tool selector, canvas block via useMcpToolsQuery/useMcpTools) had no subscription, so list_changed updates could stay invisible for the full window. Mount useMcpToolsEvents inside useMcpToolsQuery so every tools consumer gets real-time push from the shared, reference-counted connection — no consumer is left re-probing. Removed the now-redundant explicit mounts from the settings page and tool-input.

waleedlatif1 · 2026-07-22T07:21:12Z

@greptile

waleedlatif1 · 2026-07-22T07:21:13Z

@cursor review

…obalThis mcp.ts captures the connections Map and subscribed Set in module consts at import, so setting the globalThis property to undefined didn't reset what the module uses — subscription state leaked across tests. Clear the shared instances instead.

waleedlatif1 · 2026-07-22T07:28:32Z

@greptile

waleedlatif1 · 2026-07-22T07:28:33Z

@cursor review

The client-side serversList invalidation on a successful probe assumed the probe updated the stored status, but a server-side cache hit returns tools without touching status — so in the failed-cache-delete edge it fired a pointless refetch. The benefit (instant vs the 60s serversList stale-time for status recovery) doesn't justify the incorrect assumption; rely on the existing 60s stale-time + SSE push for status recovery instead. Keeps the corrected keep/drop comment.

waleedlatif1 · 2026-07-22T08:07:34Z

@greptile

waleedlatif1 · 2026-07-22T08:07:35Z

@cursor review

cursor

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit 1815e21. Configure here.}

vercelBottemporarily deployed to Preview July 22, 2026 06:29 Inactive

waleedlatif1 mentioned this pull request Jul 22, 2026
v0.7.41: email fixes, MCP fixes, gemini 3.6 flash, GitHub runners fallback, mothership experience improvements, soft delete chats #5835
Merged

greptile-appsBot reviewed Jul 22, 2026
View reviewed changes

Comment threadapps/sim/hooks/queries/mcp.ts

vercelBottemporarily deployed to Preview July 22, 2026 06:34 Inactive