convex-monitor
Watches for the next dev or prod error or request in a Convex app and returns it as a typed event to react to. Instead of polling logs, it races local error logs, deployment subscriptions, and Sentinel prod-error rows, returning the first to fire so developers can triage and fix issues the moment they appear.
npx skills add https://github.com/convex/agent-skills --skill convex-monitorBefore / After Comparison
1 组Developers had to manually refresh logs and poll deployment state, often learning about an error minutes later and stitching together context by hand.
This Skill blocks on the next typed event, racing local error logs, deployment subscriptions, and prod-error rows to return the first trigger so issues are found and fixed immediately.
Watch for the next thing to react to
Block on the next typed event instead of polling. Races local error logs, deployment subscriptions, and Sentinel prod-error rows; returns the first to fire (or a quiet heartbeat).
Workflow
- Call
wait_for_eventwith {project_dir, event_kinds, timeout_ms}. - On kind=convex_error/next_error: decode and fix it. On kind=prod_error: triage (see sentinel) and fix. On kind=feature_request: build it. On kind=quiet: loop.
- Where a harness has no blocking MCP (e.g. Copilot cloud), the pack runs a poll loop with the SAME event contract — same behavior, different mechanism.
Rules
- Prefer the blocking tool; fall back to a poll loop only where blocking MCP is weak.
- The event schema is fixed and versioned — the same trigger yields the same typed event.
- Prod events (kind=prod_error) require a deployed cloud app plus Sentinel.
User Reviews (0)
Write a Review
No reviews yet
Statistics
User Rating
Rate this Skill