Integrations overview
Everything TraceX connects to, code, databases, logs, errors, chat, email, tickets, on-call and docs, what each connection is used for, and which to connect first.
TraceX is only as good as what it can see. Integrations give it two things:
- Sources: the systems it reads evidence from: your code, databases, logs and errors.
- Intake: the places issues come from: Slack, a support inbox, tickets and alerts, so investigations start on their own and answers go back where the issue was raised.
Connect integrations per project under Integrations in the dashboard. Credentials are checked with a real read when you add them, stored encrypted, and never shown again.
Which to connect first
Section titled “Which to connect first”| Priority | Connect | Why |
|---|---|---|
| 1 | A repository (GitHub) | Turns errors into files and lines, and finds the change behind an incident. |
| 2 | Your main database (PostgreSQL, MongoDB) | Lets TraceX look up the actual records an issue is about. |
| 3 | Errors or logs (Sentry, Datadog, …) | Shows the failure itself: what, how often, since when. |
| 4 | Where issues arrive (Slack, Jira, …) | Starts investigations without anyone opening the dashboard. |
| 5 | Your docs (Notion and Confluence) | Gives TraceX your runbooks and past incidents. |
One repository and one database is enough for a first investigation. See the Quickstart.
Sources
Section titled “Sources”| Category | Integrations | Used for |
|---|---|---|
| Code | GitHub, GitLab and Bitbucket | Reading code, matching errors to changes, PR review and fix pull requests (GitHub) |
| Databases | PostgreSQL, MongoDB | Looking up the records an issue is about, read-only |
| Warehouses | BigQuery, Snowflake | Impact and trends over time, read-only |
| Logs | Datadog, CloudWatch, Loki, Elasticsearch, Google Cloud Logging, Azure Monitor | Searching logs around an incident and for identifiers |
| Errors | Sentry | Errors, stack traces and releases |
| Docs | Notion and Confluence | Runbooks and internal docs |
Intake
Section titled “Intake”| Category | Integrations | How an investigation starts |
|---|---|---|
| Chat | Slack | Mention @TraceX in a thread |
| Gmail support inbox | A new email to the support address | |
| Tickets | Jira, Linear, Zendesk, Intercom | Assign, label or tag the ticket |
| On-call | PagerDuty, Opsgenie | An incident or alert fires |
Automatic intake from Jira, Linear, Zendesk, Intercom, PagerDuty and email is part of the Team plan. Coming later: MySQL, Oracle, Outlook and Microsoft Teams.
Plan limits
Section titled “Plan limits”Starter includes 2 connected sources, Pro 3, Team unlimited (Plans).
Least privilege
Section titled “Least privilege”Give TraceX a dedicated read-only user or token for each system, scoped to what it needs. Each integration page lists the minimum permissions. Then:
- every query, log search and file read is recorded in the data access log;
- you can disconnect one integration from Integrations, or everything at once from Settings → Revoke access.