Sourcegraph 6.6.0

Template variables such as repository.search_result_paths and previous_step.modified_files allow you to chain operations and optimize batch changes by processing only relevant files. The templating engine supports Go's text/template features including conditionals, loops, and custom helper functions, providing programmatic control over changeset generation and execution flow.

This feature is available on Enterprise plans and enables more sophisticated automation workflows for large-scale code modifications.

Read the docs for more info.

Inventory Stats, launched last month, make it easy to track how your codebase evolves over time, no search required. We’ve now added support for grouping metrics to help you create cleaner, more focused visualizations.

Grouped charts are especially useful for tracking migrations, understanding repo composition, or comparing multiple teams or projects in one view.

This update gives you more power and clarity when using Inventory Stats to explore and monitor your codebase at scale. Learn more in our docs.

We’ve introduced a new Inventory Stats Code Insight that makes it easy to track how your codebase evolves over time, no search required. This insight complements existing Insights like Track Changes and Language Usage, but brings a more scalable and reliable way to monitor inventory-style metrics across repositories.

With Inventory Stats, you can:

• Track lines of code, file count, and repository size across one or more repos
• See changes over time using configurable time granularity
• Filter repos directly (no search syntax needed)
• Export aggregated data for offline analysis
• Share the insight like any other Code Insight

This insight uses a non-search-based approach that scales to large codebases, giving you a fast, reliable way to monitor growth, complexity, or cleanup work. Live previews help you fine-tune your metrics before saving.

Learn more about Code Insights in our docs.