skills/golem-powers/_archive/convex/SKILL.md
Manage Convex backend operations including dev server lifecycle, cloud deployments, function execution, schema management, and data import/export. Wraps the npx convex CLI with project-specific configuration. Includes workflows for starting local dev, deploying to production, running one-off functions, exporting/importing data snapshots, and managing environment variables. Use when starting a Convex dev server, deploying backend changes, running mutations or queries, managing Convex schema, or debugging Convex function errors. Triggers on 'convex', 'backend deploy', 'run function', 'schema change', 'convex dev'. NOT for: Supabase or Firebase operations (use respective tools), frontend-only React work, or general database queries.
npx skillsauth add etanhey/golems convexInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Backend skill wrapping
npx convexCLI. Routes to workflows for specific operations.
Run first to verify you're in a Convex project:
[ -d "convex" ] && echo "Convex project detected" || echo "ERROR: No convex/ directory found"
If no convex/ directory: Run npx convex init to initialize.
| What you want to do | Workflow | |---------------------|----------| | Start dev server | workflows/dev.md | | Deploy to production | workflows/deploy.md | | Run a Convex function | workflows/run-function.md | | Export data | workflows/data-export.md | | Import data | workflows/data-import.md | | Inspect/validate schema | workflows/schema.md | | Delete user and all data | workflows/user-deletion.md | | Fix errors | workflows/troubleshooting.md |
Execute directly - they handle errors and edge cases:
| Script | Purpose | Usage |
|--------|---------|-------|
| scripts/dev.sh | Start dev server with auto-cleanup | bash ~/.claude/commands/convex/scripts/dev.sh [--deployment <name>] |
| scripts/deploy.sh | Deploy to production | bash ~/.claude/commands/convex/scripts/deploy.sh [--1p] [--dry-run] |
| scripts/run-function.sh | Execute queries/mutations/actions | bash ~/.claude/commands/convex/scripts/run-function.sh api:funcName [--args '{}'] |
| scripts/export-data.sh | Export database backup | bash ~/.claude/commands/convex/scripts/export-data.sh [--path <file>] |
| scripts/import-data.sh | Import data (with safety prompts) | bash ~/.claude/commands/convex/scripts/import-data.sh --path <file.zip> |
# Development
npx convex dev # Start dev server with hot reload
# Deployment
npx convex deploy # Deploy to production
# Functions
npx convex run api:functionName # Run a function
npx convex run api:func --arg '{"key":"value"}' # With arguments
# Data
npx convex export --path ./backup.zip # Export all data
npx convex import --path ./backup.zip # Import data
# Environment
npx convex env list # List env vars
npx convex env set KEY=value # Set env var
npx convex env get KEY # Get env var
# Other
npx convex dashboard # Open dashboard in browser
npx convex codegen # Regenerate types
npx convex logs --name prod # View production logs
convex/ directory exists before running commandsCONVEX_DEPLOY_KEY (see deploy workflow)ALWAYS prefix Convex commands with this cleanup:
rm -f convex/*.js 2>/dev/null; npx convex dev
If you see:
✘ [ERROR] Two output files share the same path but have different contents: out/auth.js
Cause: Orphan .js files in convex/ folder (from git worktrees, crashes, or manual operations).
Fix: rm -f convex/*.js then retry.
Prevention: Always use the cleanup prefix. Only .ts files belong in convex/.
See workflows/troubleshooting.md for details.
tools
The human-eval UX contract for Phoenix views: turn-by-turn scrollable replay (not a scorecard), hide-but-copyable IDs, collapsed thinking, identity chips, tool filters, tiny frozen starter datasets, mark-wrong-in-thread, mobile-first. Use when: building or reviewing ANY Phoenix/eval view, annotation UI, session replay, or human-grading surface. Triggers: phoenix view, eval UI, annotation view, session replay, human eval UX, grading interface. NOT for: Phoenix data pipelines/ingest (capture scripts have their own specs).
tools
macOS systems specialist — AppKit NSPanel architecture, launchd services, socket activation, MCP bridge resilience, syspolicyd, and high-frequency SwiftUI dashboards. Use when building menu-bar apps, LaunchAgents, debugging syspolicyd/Gatekeeper/TCC, resilient UDS/MCP bridges, or SwiftUI dashboards at 10Hz+.
development
Bulk LLM-judging protocol for fleet-dispatched verdict runs (KG cluster, eval harness). Use when: dispatching or running judge workers (J1/J2/RT), planning bulk-apply from verdict JSONL, or triaging evidence_degraded outputs. Triggers: judge fleet, bulk judge, R3 verdicts, kg-judge, RT gate, evidence_degraded. NOT for: single-item code review, Phoenix view UX (use phoenix-human-view), or non-judge eval pipelines.
development
Quiet-down protocol for sprint close: when the fleet wraps, delete ALL polling crons and monitors, send ONE final dashboard + ONE message, then go SILENT. Use when: fleet wraps, all workers done, overnight queue exhausted, sprint close, Etan asleep/away with nothing approved left. Triggers: fleet wrap, wrap the fleet, stand down, going quiet, sprint close. NOT for: mid-sprint monitoring (keep your loops), spawning a successor (use /session-handoff first).