Menu
Run SWEny triage to investigate production alerts, create issues, and optionally implement fixes. Use when the user wants to analyze recent errors or alerts.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Verify SWEny provider credentials and connectivity. Run to diagnose configuration issues before running workflows.
Run SWEny E2E browser test workflows from .sweny/e2e/. Use when the user wants to execute end-to-end tests.
Pick up an issue and implement a fix with SWEny. Use when the user has an issue ID or URL they want resolved.
Create standard SWEny label set in your issue tracker (Linear or GitHub). Run after init to prepare your tracker for triage.
Render a SWEny workflow as a Mermaid diagram for visualization.
Edit an existing SWEny workflow using natural language instructions.
| name | triage |
| description | Run SWEny triage to investigate production alerts, create issues, and optionally implement fixes. Use when the user wants to analyze recent errors or alerts. |
| disable-model-invocation | true |
| allowed-tools | Bash |
| context | fork |
| agent | sweny-workflow |
| argument-hint | ["--dry-run"] |
Run the SWEny triage workflow. This investigates production alerts from configured observability providers (Datadog, Sentry, BetterStack, etc.), performs root cause analysis, creates issues, and optionally implements fixes.
Prerequisites: A .sweny.yml config file and provider credentials must be set up. Run /sweny:check first if unsure.
Run triage with default settings:
sweny triage --stream
Run in dry-run mode (investigate only, skip creating issues/PRs):
sweny triage --stream --dry-run
If the user provided arguments, pass them through:
sweny triage --stream $ARGUMENTS
sweny check output for credential issues--dry-run — investigate without creating issues or PRs--time-range <range> — time window to analyze (default: 24h)--severity-focus <focus> — severity level (default: errors)--service-filter <filter> — filter by service pattern