Managing integrations
Manage connected tools so agents and workflows have the right capabilities without unnecessary access.
Use this state to check connector availability, connection status, and whether the tool maps to a real workflow or agent action.
Follow the real configuration that turns an operation into a system.
These captures favor full, readable product states: agents, workflows, channels, logs, analytics, and publishing controls without floating labels or artificial step boxes.
Use this state to check connector availability, connection status, and whether the tool maps to a real workflow or agent action.
Use this state to check connector availability, connection status, and whether the tool maps to a real workflow or agent action.
Use this state to check connector availability, connection status, and whether the tool maps to a real workflow or agent action.
Summary
Manage connected tools so agents and workflows have the right capabilities without unnecessary access.
Concepts covered
Step breakdown
- Open IntegrationsStart in the Integrations area and confirm what the real screen is showing.
- Inspect product behaviorUse screenshots and visible product states to understand what the screen does, why it exists, and how teams use it.
- Connect the platformSee how the screen connects to agents, workflows, records, activity, channels, integrations, or ownership.
What the Integrations panel shows
The Studio Integrations panel lists every external system connected to the workspace: Gmail, Slack, WhatsApp, custom APIs, and any other authorized tools. Each entry shows connection status, the account or endpoint connected, and which agents or workflows depend on it.
This is the governance view. Before changing, rotating, or disconnecting a tool, check which workflows and agents reference it. Disconnecting an integration that a production workflow depends on breaks the workflow silently at the node that uses it.
Minimum access principle
Each integration should have access to exactly what the workflow or agent needs — no more. A workflow that sends Slack notifications does not need read access to Slack messages. An agent that sends Gmail follow-up does not need access to the full inbox.
Review scopes when connecting any integration. If the OAuth or API scope is broader than what the workflow requires, document why — or narrow it.
Rotating credentials and reconnecting
API keys and OAuth tokens expire or are revoked. When an integration fails, the first thing to check is whether the credential is still valid.
Store API credentials as workspace secrets rather than hardcoded strings in nodes. When a credential needs to be rotated, update the secret in one place and all nodes that use it pick up the new value without editing each workflow.
Auditing as workflows grow
As the workspace grows, integrations accumulate. Review the Integrations panel quarterly: remove integrations that are no longer referenced by any active workflow or agent, update credentials that are approaching expiry, and confirm that active integrations are still scoped appropriately.
An integration audit is also a workflow audit — unused integrations often point to deprecated workflows that should be archived.
Operational playbook
Use Managing integrations as part of the Frontline Studio Integrations operating loop: inspect the current product state, confirm the source context, and decide what should happen next.
The goal is not to memorize screens. The goal is to understand how the product surface supports repeatable work, AI assistance, and accountable handoff.
Best practices
Start with the operational job before changing configuration. Name the owner, define the trigger or source context, and decide how the result should be reviewed.
Prefer narrow, inspectable setups over broad automation. Teammates should be able to explain why the system took an action from the visible product state.
Troubleshooting
If the result does not match expectation, check the source context first, then permissions, connected integrations, required fields, workflow logs, and any AI-generated output used by downstream steps.
When in doubt, compare the latest product state with the related record, activity, or workflow execution so debugging starts from evidence rather than guesswork.
Access and permissions
Before connecting a tool, decide which agent or workflow needs it and what action it will perform. Use the smallest useful permission scope whenever the integration supports it.
Document which workflows depend on the integration so future debugging can trace failures back to a connected account, connector state, or external API response.
FAQs
What does Managing integrations teach?
Manage connected tools so agents and workflows have the right capabilities without unnecessary access.
How should teams use this lesson?
Use it as a product walkthrough: understand the real screen, the product behavior, the operational outcome, and how the area connects with the rest of Frontline.
When should I connect an integration?
Connect an integration when an agent or workflow has a clear need to read context, send a message, update a system, or trigger an external action. Avoid connecting tools without a defined operating path.
How do integrations affect workflow reliability?
Integrations introduce permissions, API responses, rate limits, and failure modes. Use logs, payload mapping, and fallback routing so a failed external action does not become invisible.
Who should manage integration access?
Workspace owners or operators responsible for the workflow should manage access. The person connecting the tool should understand what the workflow can read, write, and trigger.
How do I debug an integration failure?
Check whether the account is connected, the workflow node has the right action, required fields are mapped, credentials are valid, and the external system returned an error.