Inputs & Outputs

SWEny ships three GitHub Actions, each with its own input surface. This page documents all inputs for all three.

Action	Use case	Docs
swenyai/triage@v1	SRE triage — monitors alerts, files tickets, opens fix PRs	Triage inputs
swenyai/e2e@v1	Agentic E2E browser tests	E2E inputs
swenyai/sweny@v5	Generic runner — any workflow YAML	Generic inputs

---

Generic runner inputs

swenyai/sweny@v5 — runs any SWEny workflow YAML. Minimal input surface.

Input	Description	Default
workflow	Path to the workflow YAML file to execute (relative to working-directory)	required
claude-oauth-token	Claude Code OAuth token from a Max/Pro subscription	—
anthropic-api-key	Anthropic API key for Claude (pay-per-use billing)	—
cli-version	Version of @sweny-ai/core to install	latest
node-version	Node.js version to install	24
working-directory	Working directory to run from	.

You must provide either claude-oauth-token or anthropic-api-key.

Pass any additional credentials your workflow needs via env: on the step — the CLI auto-loads environment variables.

---

Triage action inputs

swenyai/triage@v1 — the full triage preset with observability, issue tracking, notification, and investigation inputs.

Authentication

Input	Description	Default
anthropic-api-key	Anthropic API key for Claude (pay-per-use billing)	—
claude-oauth-token	Claude Code OAuth token from a Max/Pro subscription (predictable monthly cost)	—
github-token	GitHub token for API access and PR creation	${{ github.token }}
bot-token	Optional bot token with elevated permissions for cross-repo dispatch or pushing to protected branches	—

You must provide either anthropic-api-key or claude-oauth-token. The github-token is auto-provided by GitHub Actions.

Coding Agent

Input	Description	Default
agent	Agent to use for implementation: claude, codex, or gemini	claude
openai-api-key	OpenAI API key (required when agent is codex)	—
gemini-api-key	Google Gemini API key (required when agent is gemini)	—

Observability

Input	Description	Default
observability-provider	Provider to query for logs, errors, and metrics	datadog

Supported values: datadog, sentry, betterstack, cloudwatch, splunk, elastic, newrelic, loki, honeycomb, axiom, heroku, opsgenie, pagerduty, prometheus, vercel, supabase, netlify, fly, render, file.

Datadog

Input	Description	Default
dd-api-key	Datadog API key	—
dd-app-key	Datadog Application key	—
dd-site	Datadog site (e.g., datadoghq.com, datadoghq.eu)	datadoghq.com

Sentry

Input	Description	Default
sentry-auth-token	Sentry auth token	—
sentry-org	Sentry organization slug	—
sentry-project	Sentry project slug	—
sentry-base-url	Sentry base URL (for self-hosted instances)	https://sentry.io

CloudWatch

Input	Description	Default
cloudwatch-region	AWS region for CloudWatch Logs	us-east-1
cloudwatch-log-group-prefix	CloudWatch log group name or prefix	—

AWS credentials

CloudWatch uses the standard AWS credential chain. Set AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY as repository secrets, or use OIDC federation for keyless auth.

BetterStack

Input	Description	Default
betterstack-api-token	Better Stack API token	—
betterstack-source-id	Better Stack Telemetry source ID	—
betterstack-table-name	Better Stack Telemetry ClickHouse table name	—

Splunk

Input	Description	Default
splunk-url	Splunk REST API base URL (e.g., https://splunk.example.com:8089)	—
splunk-token	Splunk authentication token	—
splunk-index	Splunk index to search	main

Elasticsearch

Input	Description	Default
elastic-url	Elasticsearch base URL (e.g., https://elastic.example.com:9200)	—
elastic-api-key	Elasticsearch API key	—
elastic-index	Elasticsearch index pattern	logs-*

New Relic

Input	Description	Default
newrelic-api-key	New Relic API key (NRAK-...)	—
newrelic-account-id	New Relic account ID	—
newrelic-region	New Relic region: us or eu	us

Grafana Loki

Input	Description	Default
loki-url	Loki base URL (e.g., https://loki.example.com)	—
loki-api-key	Loki API key (optional for unauthenticated instances)	—
loki-org-id	Loki tenant/org ID (optional)	—

Honeycomb

Input	Description	Default
honeycomb-api-key	Honeycomb API key	—
honeycomb-dataset	Honeycomb dataset name	—

Axiom

Input	Description	Default
axiom-api-token	Axiom API token	—
axiom-dataset	Axiom dataset name	—
axiom-org-id	Axiom org ID (required for multi-org accounts)	—

PagerDuty

Input	Description	Default
pagerduty-api-key	PagerDuty API key	—

OpsGenie

Input	Description	Default
opsgenie-api-key	OpsGenie API key	—
opsgenie-region	OpsGenie region: us or eu	us

Prometheus

Input	Description	Default
prometheus-url	Prometheus base URL (e.g., http://prometheus.internal:9090)	—
prometheus-token	Prometheus bearer token (optional, for secured instances)	—

Vercel

Input	Description	Default
vercel-token	Vercel personal access token	—
vercel-project-id	Vercel project ID (prj_...)	—
vercel-team-id	Vercel team ID (team_..., optional for team-owned projects)	—

Supabase

Input	Description	Default
supabase-management-key	Supabase management API key	—
supabase-project-ref	Supabase project reference ID	—

Netlify

Input	Description	Default
netlify-token	Netlify personal access token	—
netlify-site-id	Netlify site ID (find in Site Settings > General)	—

Fly.io

Input	Description	Default
fly-token	Fly.io personal access token	—
fly-app-name	Fly.io application name	—

Render

Input	Description	Default
render-api-key	Render API key	—
render-service-id	Render service ID (srv-...)	—

Heroku

Input	Description	Default
heroku-api-key	Heroku API key	—
heroku-app-name	Heroku application name	—

File

Input	Description	Default
log-file-path	Path to a local JSON log file	—

Use the file provider for testing without connecting to an external service.

Issue Tracking

Input	Description	Default
issue-tracker-provider	Issue tracker to use: github-issues, linear, or jira	github-issues

Linear

Input	Description	Default
linear-api-key	Linear API key	—
linear-team-id	Linear team UUID	—
linear-bug-label-id	Linear label UUID for bugs	—
linear-triage-label-id	Linear label UUID for agent-triage issues	—
linear-state-backlog	Linear workflow state UUID for Backlog	—
linear-state-in-progress	Linear workflow state UUID for In Progress	—
linear-state-peer-review	Linear workflow state UUID for Peer Review	—

Only linear-api-key and linear-team-id are required. The label and state UUIDs are optional and let you control where issues land in your workflow.

Jira

Input	Description	Default
jira-base-url	Jira base URL (e.g., https://myco.atlassian.net)	—
jira-email	Jira user email for API authentication	—
jira-api-token	Jira API token	—

GitHub Issues

GitHub Issues requires no extra configuration. It uses the github-token input, which is auto-provided by GitHub Actions.

Source Control

Input	Description	Default
source-control-provider	Source control provider: github or gitlab	github
gitlab-token	GitLab personal access token	—
gitlab-project-id	GitLab project ID or path (e.g., my-group/my-project)	—
gitlab-base-url	GitLab instance base URL	https://gitlab.com

Notification

Input	Description	Default
notification-provider	Where to send results: github-summary, slack, teams, discord, email, webhook, or file	github-summary
notification-webhook-url	Webhook URL (required for slack, teams, discord, and webhook providers)	—
sendgrid-api-key	SendGrid API key (required when notification-provider is email)	—
email-from	Sender email address (required when notification-provider is email)	—
email-to	Recipient email addresses, comma-separated (required when notification-provider is email)	—
webhook-signing-secret	HMAC-SHA256 signing secret for generic webhook notifications	—
output-dir	Directory for file-based provider output (dry-run and file provider artifacts)	.github/sweny-output

Investigation

Input	Description	Default
time-range	Time window to analyze (e.g., 1h, 6h, 24h, 7d)	24h
severity-focus	What to focus on: errors, warnings, or all	errors
service-filter	Service filter pattern (e.g., my-service, api-*, *)	*
investigation-depth	How deep the agent should investigate: quick, standard, or thorough	standard
max-investigate-turns	Maximum Claude turns for the investigation phase (1—500)	50
max-implement-turns	Maximum Claude turns for the implementation phase (1—500)	30

Matching time-range to your schedule

If you run triage on a daily cron, set time-range: 24h. If you run every 6 hours, use time-range: 6h. This avoids investigating the same errors twice.

PR Settings

Input	Description	Default
base-branch	Target branch for PRs (e.g., main, master, develop)	main
pr-labels	Comma-separated list of labels to apply to created PRs	agent,triage,needs-review

Behavior

Input	Description	Default
dry-run	Analyze only — the executor stops at the first conditional edge, guaranteeing zero side effects. Enforced by the executor, not by prompt.	false
review-mode	PR merge behavior: auto (enable GitHub auto-merge when CI passes, automatically suppressed for high-risk changes like migrations, auth, lockfiles, or >20 files) or review (open PR and wait for human approval)	review
novelty-mode	Only report novel issues not already tracked in your issue tracker	true
issue-override	Existing issue identifier to work on (e.g., ENG-123). Skips log scanning and works directly on this issue.	—
additional-instructions	Extra guidance for the agent (e.g., "focus on the webhook handler")	—

Service Map

Input	Description	Default
service-map-path	Path to the service ownership map YAML file	.github/service-map.yml

See Service Map for the file format and usage guide.

Workspace Tools

Input	Description	Default
workspace-tools	Comma-separated list of workspace integrations to enable. Each tool injects its MCP server when the corresponding credential environment variable is present	—

Supported values: slack, notion, pagerduty, monday, asana.

Example: workspace-tools: "slack,notion"

MCP Servers

Input	Description	Default
mcp-servers-json	Additional MCP servers as a JSON object. Each key is a server name; each value is an MCPServerConfig with fields: type, command, args, env, url, headers	—

Provider-based MCP servers (GitHub, Linear, Datadog) are injected automatically based on your configured providers — you do not need to add them here.

mcp-servers-json: |
  {
    "filesystem": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem@latest", "/workspace"]
    }
  }

See MCP Servers for the full server catalog.

Setup

Input	Description	Default
cli-version	Version of @sweny-ai/core to install	latest
node-version	Node.js version to install	24
working-directory	Working directory to run from	.

---

Triage outputs

After the triage workflow completes, SWEny sets these outputs on the action step. Use them in subsequent workflow steps with ${{ steps.<step-id>.outputs.<output> }}.

Output	Description	Example
issues-found	Whether any issues were found during investigation	true
recommendation	Investigation recommendation	implement, +1 existing ENG-456, skip
issue-identifier	Issue identifier from the configured tracker	ENG-123, #42
issue-url	URL of the created or found issue	https://linear.app/team/issue/ENG-123
pr-url	URL of the created pull request	https://github.com/org/repo/pull/99
pr-number	Number of the created pull request	99

Using outputs in subsequent steps

jobs:
  triage:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0

      - uses: swenyai/triage@v1
        id: sweny
        with:
          claude-oauth-token: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
          dd-api-key: ${{ secrets.DD_API_KEY }}
          dd-app-key: ${{ secrets.DD_APP_KEY }}

      - name: Post to Slack if issues found
        if: steps.sweny.outputs.issues-found == 'true'
        run: |
          curl -X POST "${{ secrets.SLACK_WEBHOOK }}" \
            -H 'Content-type: application/json' \
            -d '{"text": "SWEny found issues: ${{ steps.sweny.outputs.issue-url }}"}'

---

E2E action inputs

swenyai/e2e@v1 — runs agentic browser tests against a deployed app.

Input	Description	Default
workflow	Path to the E2E workflow YAML file to execute	required
claude-oauth-token	Claude Code OAuth token	—
anthropic-api-key	Anthropic API key for Claude	—
base-url	Base URL of the deployed app under test (exposed as BASE_URL)	—
cli-version	Version of @sweny-ai/core to install	latest
agent-browser-version	Version of agent-browser to install	latest
node-version	Node.js version to install	24
working-directory	Working directory to run from	.
screenshots-path	Path where the workflow writes screenshots and artifacts	results/
artifact-name	Name of the uploaded screenshots artifact	e2e-screenshots
artifact-retention-days	How long to retain the screenshots artifact (days)	14

You must provide either claude-oauth-token or anthropic-api-key.