Track content freshness
Topicary flags stale content so you can prioritize updates before documentation drifts out of date.
Staleness indicators
A topic counts as stale when its last edit is 90 or more days old. On the Topics listing page, stale topics carry a health indicator on their row; hover it to see the issue list, where staleness appears as Stale (90d+) alongside any other health issues the topic has.
The 90-day cutoff behind the Topics page's staleness indicator and filters is fixed and cannot be changed. The nightly health scan is separate: it has its own Staleness threshold (days) setting, per project under Settings ▸ Content & AI ▸ Content health scans, which also starts at 90. A topic flagged as stale is not necessarily wrong. It is a signal to verify, not an error.
Filter for stale content
Go to Topics.
Open the Health filter chip and select Stale (90d+) to show only topics whose last edit is 90 or more days old.
The Modified filter chip offers the same cutoff from the other direction: Today, This week, and This month show recently edited topics, and Over 90 days shows the stale ones.
Review and refresh
Open each stale topic and assess whether the content is still accurate:
If it is still correct, make a small edit (even a whitespace change) and save to reset the staleness timer.
If it needs updates, revise the content and save.
Combine freshness filters with tag filters to create targeted review queues. For example, filter for stale topics tagged "Authentication" to review all authentication content before a security-related release. This is more efficient than reviewing every stale topic across the entire project.
Mark a topic as verified
Staleness measures how recently a topic was edited. Verification is a separate, stronger signal: a deliberate human confirmation that the content is accurate, even if it has not changed.
To verify a topic:
Open the topic in the editor.
In the Properties panel, find the Verification section.
Click Mark as verified.
The section then shows "Content verified" with the date it was confirmed accurate. To clear it, click Remove verification.
Only a signed-in person can verify a topic; AI features cannot. Verification records who confirmed the content and when, which is why it is treated as a governance signal rather than an automated check. Verifying a topic does not change its content or its staleness timer.
Verification age is one of the inputs to the Content debt score, and the Automated lifecycle recommendations can flag published topics whose verification has expired or that have never been verified.
Additional content health checks
Beyond staleness, the Topics page can filter for:
Empty: topics with no content
Orphaned: topics not included in any map
Broken references: topics referencing deleted components
Undefined variables: topics using variable tokens that do not exist in any set
Use these filters during regular content audits to maintain documentation quality.
See also
Content health and governance: Broader overview of automatic health tracking across staleness, orphans, references, and structure
Content debt score: How staleness and verification age roll into a single 0-10 debt score
Automated lifecycle recommendations: Nightly scans that flag stale and unverified topics for action
Content health indicators: Reference for all health indicator types, thresholds, and severity levels
Content statuses: Editorial statuses that complement freshness tracking for workflow management
Classify content with tags: Combine tag filters with freshness filters for targeted review queues