Query Convex deployment logs and health insights through the official MCP read tools.
npx skills add https://github.com/openclaw/clawhub --skill convex-insights
Please help me install this Agent Skill. Skill: convex-insights Source: https://github.com/openclaw/clawhub/tree/main/.agents/skills/convex-insights Command: npx skills add https://github.com/openclaw/clawhub --skill convex-insights
Investigating failures, slow functions, and deploy regressions in a running Convex app requires knowing the right log filters and identifiers, and raw log dumps are hard to act on. This Skill turns natural-language operational questions into scoped, evidence-backed queries against the official Convex MCP read tools.
Ask the agent to check what functions are failing in my Convex deployment since the last deploy and show the evidence with a dashboard link.
High-intent search queries and answers about installing and using this skill.
Use the failures view: fetch a bounded recent window with logs --history <n> --jsonl, then filter client-side to group failures by function and error message. The result includes counts, a representative stack trace, and a dashboard deep link.
Use the trace view: fetch a recent log window with logs --history <n> --jsonl, then filter client-side to one requestId or function name to read the full execution. There is no server-side requestId filter, so filtering happens locally.
No. The logs tool accepts only --history (a count, not a time window), --success, --jsonl, --prod, and --deployment, and insights has no function filter. You fetch a bounded recent window and filter client-side by function, status, or requestId.
No. The insights tool is cloud-only and requires user authentication, covering dev and prod cloud deployments. It surfaces typed 72-hour read-limit and OCC conflict events, which this Skill ranks and reports.
Compare the failure onset timestamps from the logs against the deployment version reported by the status tool. The Skill correlates these signals to answer deploy-causality questions without asserting causation beyond the evidence.
Use convex-advisor when the question is about root-causing and fixing performance or cost problems. This Skill stays read-only for observability and hands off perf and cost findings to convex-advisor as pointer findings.
matixlolQuery Convex databases via the Convex MCP server for schema and data inspection.
allcllAnalyze backend code and system behavior to identify root causes and risks.
brainbrewlabsTriage monitoring alerts by assessing severity and diagnosing root causes.
mindcockpit-aiScan repositories for plaintext secrets and generate env.tpl templates.
