Integrations
Flintwake belongs in the tools your team already has open.
Alerts go where your rota lives. Wakes go where your analysts work. Deploy markers come from the pipeline you already run. Everything else is a signed webhook away.
- 34ConnectorsAcross six categories
- 6Data destinationsWarehouse, object store and analytics
- 8RuntimesEdge, gateway, serverless and Kubernetes
- 1FallbackSigned webhooks to any endpoint you control
34 integrations
- Slack
Alert routing, incident threads and a daily digest per channel.
AlertingPush - Microsoft Teams
Adaptive cards for SLO burn, quota warnings and failed deliveries.
AlertingPush - PagerDuty
Burn-rate alerts open and resolve incidents with the wake list attached.
IncidentsTwo-way - Opsgenie
Route alerts by endpoint owner using your existing escalation policies.
IncidentsPush - Incident.io
Attach a filtered wake view to an incident channel automatically.
IncidentsTwo-way - Statuspage
Mirror Flintwake component health onto your own public status page.
IncidentsPush - Datadog
Forward endpoint and consumer metrics as custom metrics with tags intact.
DataPush - Grafana
A data source plugin for wake counts, latency quantiles and error rates.
DataPull - Snowflake
Hourly export of captured wakes into a table you control.
DataPush - BigQuery
Streaming inserts with a stable schema and automatic partitioning by day.
DataPush - Amazon S3
Compressed NDJSON delivered to your bucket with a manifest per hour.
DataPush - ClickHouse
Direct insert for teams already running their own analytics cluster.
DataPush - Honeycomb
Send wakes as events with the timing breakdown as fields.
DataPush - New Relic
Metric and event forwarding through the standard ingest endpoint.
DataPush - Elastic
Index wakes into an Elasticsearch cluster with an included mapping.
DataPush - GitHub
Deploy markers from Actions, and redaction rules applied from a repo.
SourceTwo-way - GitLab
Pipeline deploy markers and rule sync from a project repository.
SourceTwo-way - Linear
Create an issue from an error group with the wake IDs attached.
SourcePush - Jira
Open a ticket from an investigation, preserving the filtered view.
SourcePush - Vercel
Deploy markers and an edge collector for Next.js route handlers.
RuntimeTwo-way - Cloudflare Workers
A Worker-native collector with sub-millisecond added latency.
RuntimePush - Fastly
Compute@Edge collector distributed as a Fastly package.
RuntimePush - AWS Lambda
Extension-based collector for functions behind API Gateway.
RuntimePush - Kubernetes
Sidecar collector with a Helm chart and a validating webhook.
RuntimePush - Envoy
An ext_authz filter that captures without terminating the request.
RuntimePush - NGINX
A dynamic module for teams running NGINX as their API front door.
RuntimePush - Kong
Plugin-based capture for Kong Gateway, configured per service.
RuntimePush - Okta
SSO and SCIM provisioning for console access.
IdentityTwo-way - Entra ID
SAML sign-in and group-based role assignment.
IdentityTwo-way - Google Workspace
SSO with domain-restricted sign-in.
IdentityTwo-way - Auth0
Map Auth0 client IDs to Flintwake consumers automatically.
IdentityTwo-way - 1Password
Push newly issued API keys straight into a shared vault.
IdentityPush - Terraform
A provider for projects, environments, keys, limits and rules.
SourceTwo-way - Webhooks
Signed JSON to any endpoint you control, with retry and replay.
AlertingPush
Build your own
If it is not in the directory, it is a webhook.
Every event type can be delivered to any HTTPS endpoint you control, signed, retried and replayable. Most of the connectors above are built on exactly this.
- Timestamped HMAC signature on every delivery
- Exponential retry for eight attempts, then dead-letter
- Manual replay from the console or the API
- Verification helper in all seven SDKs
curl -X POST https://api.flintwake.com/v2/webhooks/endpoints \
-H "Authorization: Bearer $FLINTWAKE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"url": "https://hooks.acme.dev/flintwake",
"events": ["wake.error", "slo.burn", "consumer.quota.warning"]
}'Connect the first one in under a minute.
Alerting and incident connectors need a URL and a channel. Data destinations need credentials and a table name.