A top-level Errors dashboard automatically isolates and groups similar failures, giving developers a timeline of occurrences and a direct path to bulk-replay affected runs.
Developers using Supabase branching can automatically inject the correct database URLs and API keys for preview environments during the build step. Vercel staging synchronization also received stricter mapping requirements.
Batch processing throughput is restored by tracking actual items processed instead of claim attempts, while a new backpressure system prevents visibility timeouts under heavy load.
A new input streams API enables bidirectional communication, allowing developers to stream real-time data into paused or running task processes to build interactive workflows and human-in-the-loop systems.
Developers can process more background jobs simultaneously, with concurrency limits bumped from one to five for Free accounts and ten to fifty for Pro accounts.
The automated release pipeline now explicitly targets the main repository when updating Docker links, resolving potential context failures during builds.
Trigger.dev documentation has been updated to cover Input Streams, allowing real-time data to be sent into running background tasks for human-in-the-loop approvals and mid-flight cancellations.
Docker builds are now insulated from unexpected upstream breakages by pinning the Goose migration tool to a specific version compatible with the current Go environment.
The automated release workflow now dispatches an event to the separate marketing site repository to draft a changelog pull request.
By breaking a massive root context file into directory-specific instructions and adding an automated CI audit, the repository provides hyper-local guidance to AI coding agents.
External pipeline workflows are locked to exact cryptographic hashes to protect against upstream supply-chain attacks, while redundant repository checkout steps are removed to speed up execution.