Auto-healing and observability add-on for Workato
Workato add-on called Integration Nexus and adds multiple enhancement to standard Workato experience: Auto-Healer, Integration Brain, Data Steward, and SDK Builder + Watcher.
Auto-Healer
Detects failed jobs, creates fix plans, and applies safe retries, restarts, and recipe changes. Production changes stay behind a human approval gate.
Integration Brain
Every integration gets a working memory: what it does, why it exists, who owns it, what can break, and how it should be monitored.
Data Steward
Scans recipes for PII and secret leakage, then creates mitigation steps a developer can review or apply.
SDK Builder + Watcher
Tracks changes in Nordic app APIs and Workato connectors, then prepares the updates needed to keep recipes running.
Fix errors automatically
Watches for production errors, creates precise fix plans, and applies fixes automatically or with human approval
ACTION: Add new retry criteria to Recipe Id: "IAM-009"
WHY: The internal IAM update API intermittently returned "Connection timeout". A retry handler already exists on IAM-009 , but it only watches 500 Internal Server Error, so these failures slipped through.
- 1Open recipe IAM-009 · error-retry step
- 2Add “Connection timeout” to watched errors
- 3Keep existing “500 Internal Server Error”
- 4Set max retries 3 · backoff 30s
- 5Re-test in DEV with replayed payloads
NS-012 polls every 5 min — source changes ~hourly
Switch the trigger to a 30-minute schedule. Same freshness, ~104k fewer tasks / year.
Apply the suggested fix
Applies fixes automatically or with gated approval, moving changes step by step from Dev to Test to Prod
on_error:
- watch: ["500 Internal Server Error"]
+ watch: ["500 Internal Server Error",
+ "Connection timeout"]
+ max_retries: 3
+ backoff: 30sAnswer business questions about your integrations
Search across recipes to trace a data record from source to destination and answer questions like "Did this record sync?"
Provisioning to Eletive most likely did not happen because Anna Berg started 2 days ago. This integration only syncs employees after they have worked for more than 14 days.
Are your integrations running?
Gives a high-level overview of all integrations. Automatically maps integrations built across multiple recipes, functions, and SDK connectors, then recreates the end-to-end path
Warns before an API change breaks a recipe
Generates and manages SDK connectors, then watches for API changes that could break your integrations
- 1Register OAuth 2.0 app in Visma
- 2Add redirect URI · nexus.se/oauth/cb
- 3Swap connection auth on 2 connections
- 4Re-test 6 recipes in DEV
- 5Schedule deploy before 2026-09-01
Stop paying to process unchanged data
Scans recipes for task and credit usage, then suggests optimization plans with concrete savings
You’re paying to re-write 526 unchanged people, every night.
The first task does a GET from the Quinyx API, which already mirrors Visma HRPlus. We compared it against the last sync: 94% of records were identical — yet every person still ran all 12 tasks. Add a change-detection gate after the GET and skip the unchanged ones.
Protect sensitive fields and clean old PII
Continuous scans for PII and security problems and creates fix plans