Skip to content

SWEny

SWEny is a platform for building AI-powered engineering workflows. Connect any observability source, take action with AI, and report results through any channel. Triage is the first recipe — it monitors your production logs, investigates errors, and opens fix PRs automatically.

SWEny’s workflow engine orchestrates three phases — Learn, Act, Report — to automate engineering tasks end-to-end:

  1. Learn — connects to your observability stack, pulls errors and log data, checks your issue tracker for duplicates, and builds context
  2. Act — an AI agent investigates root causes through your codebase, files tickets, creates branches, and opens fix PRs
  3. Report — sends results to the channels your team already uses: Slack, Teams, Discord, email, or webhooks

Triage is the first recipe built on this engine. It runs on a schedule, finds novel production errors, and fixes them without waking anyone up. See Engine & Recipes to learn how the engine works and how to build your own recipes.

Want to understand the full flow? Read the end-to-end walkthrough — a real error spike from Datadog through to a merged PR.

Observability

Datadog, Sentry, CloudWatch, Splunk, Elasticsearch, New Relic, Grafana Loki — query logs and aggregate errors from your existing monitoring.

Issue Tracking

Linear, GitHub Issues, Jira — create tickets, search for duplicates, link PRs automatically.

Source Control

GitHub, GitLab — create branches, push commits, open PRs, dispatch cross-repo workflows.

Notification

GitHub Summary, Slack, Teams, Discord, Email, Webhook — send results wherever your team works.

Incident

PagerDuty — create incidents, acknowledge, resolve, check on-call schedules.

Storage

Filesystem, S3 — persist sessions, memory, and workspace files for the agent.

Both modes are powered by SWEny’s workflow engine, which manages provider connections, step orchestration, and the Learn/Act/Report pipeline.

GitHub Action — runs on a cron schedule in your CI. No servers, no infrastructure. This is the primary deployment.

Interactive Agent — a bot your team can chat with. Ask it to investigate errors, query logs, or write fixes interactively. Currently supports Slack, with more channels planned. Deploy as a container or run locally as a CLI.

.github/workflows/sweny-triage.yml
- uses: swenyai/sweny@v0.2
with:
claude-oauth-token: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
dd-api-key: ${{ secrets.DD_API_KEY }}
dd-app-key: ${{ secrets.DD_APP_KEY }}
dry-run: true # Remove when ready to create PRs

3 secrets. GitHub Issues by default. Full setup guide →