Configuration Limits
| Setting |
Limit |
| Configuration name |
Max 250 characters |
| Description |
Max 5,000 characters |
| JQL query |
Max 5,000 characters |
| Display JQL |
Max 1,000 characters |
| Display condition |
Max 2,000 characters |
| Configuration size |
Max 230 KB (total serialized size) |
Data Limits
| Setting |
Limit |
Default |
| Max Issues (JQL / Linked Issues) |
1-1,000 |
50 |
| Rows Per Page |
1-100 |
10 |
| Table fields per configuration |
Max 50 |
- |
| Display fields per configuration |
Max 50 |
- |
Issue link types per configuration (linkTypes) |
Max 100 |
- |
| Projects per context filter |
Max 1,000 |
- |
| Issue types per context filter |
Max 1,000 |
- |
A JQL search pages through Jira until it has loaded Max Issues issues (up to 1,000), so a large value is not cut off at a single page of results. The total shown by Show Count is Jira’s approximate count of matching issues, not the number of rows loaded.
Template Engine Limits
| Constraint |
Limit |
| Loop iterations |
Max 10,000 total iterations per render |
range(start, end, step) size |
Max 10,000 values returned per call |
| Execution timeout |
800ms per template rendering |
| Prototype access |
Blocked - templates cannot access JavaScript prototypes |
| Eval / exec |
Not available - no arbitrary code execution |
If a template exceeds the iteration limit or timeout, it stops execution and returns an error message.
Caching
Display Linked Issues uses a 3-level caching strategy to optimize performance:
| Level |
TTL |
Purpose |
| KVS (Forge Storage, server-side) |
Persistent |
Configuration data, module settings |
| IndexedDB (browser) |
4 hours |
Cached configuration lookups |
| localStorage (browser) |
Session |
Quick access for active configurations |
Cache Invalidation
- Saving a configuration immediately invalidates the server-side cache
- Browser caches refresh after the TTL expires
- Users can force a refresh by hard-reloading the page (Ctrl/Cmd + Shift + R)
Optimize Max Issues
| Use Case |
Recommended Max Issues |
| Overview panels (badges, counts) |
25-50 |
| Standard issue tables |
50 |
| Comprehensive lists |
100-200 |
| Large datasets (use sparingly) |
500-1,000 |
Higher Max Issues values mean more API calls and longer rendering times. Start low and increase only when users need to see more data.
Optimize JQL Queries
- Use indexed fields in JQL:
project, status, assignee, priority, issuetype, created, updated
- Avoid text searches on large datasets:
text ~ "keyword" is expensive
- Be specific with project and issue type filters - narrower queries are faster
- Use ORDER BY to ensure the most relevant issues appear first (especially with low Max Issues)
Optimize Templates
- Avoid nested loops - they multiply iteration counts rapidly
- Use
| take(N) to limit items processed in loops
- Minimize filter chains - each filter adds processing time
- Pre-compute values with
{% set %} variables instead of repeating expressions
- Keep conditional logic simple - deeply nested if/else blocks are harder to debug and slower to execute
Optimize Field Selection
- Select only the table fields you actually display - each field adds to the data transfer
- Use display fields sparingly - they add additional field rendering per panel
Reduce Panel Count
- Each visible panel makes independent API calls
- Consider combining related information into a single panel with sections
- Use conditional display (Display JQL, Display Conditions) to prevent unnecessary panel rendering
- Disable panels that are no longer needed rather than leaving them active
Next Steps