Use the Safe Push command-line workflow to submit IntelliJ repository changes for required tests and compilation checks before they reach a protected branch. The guide also explains dry runs, progress links, and common failure troubleshooting.
Push IntelliJ repository changes through the Safe Push workflow.
Run score Not tested
See how to installReview the instructions the agent will receive before you install.
This guide covers the Safe Push process for pushing changes to the IntelliJ repository.
Safe Push is a system that ensures code changes pass all required tests before being merged to protected branches (like master). It's powered by Patronus.
The safePush.cmd script in the repository root provides command-line access to Safe Push without needing the IDE.
# Push current HEAD to master (most common)
./safePush.cmd HEAD:master
# Push specific commit to master
./safePush.cmd <commit-hash>:master
# Push from feature branch to master
./safePush.cmd HEAD:master # while on feature branch
# Dry run (test without pushing)
./safePush.cmd -dry-run HEAD:master
-autosquash Automatically squash fixup! commits (default true)
-dry-run Run tests without push
-emergency Skip tests and push directly (USE ONLY IN EMERGENCIES)
-verbose Verbose output
-help Print help
After starting a Safe Push you'll receive a Patronus URL like:
https://patronus.labs.jb.gg/robot/<uuid>
Always print this URL back to the user — it is their primary handle for the run (link to share, cancel button, restore-branch button). The <uuid> segment is the robot id.
You can also track the Safe Push directly: check current status, enumerate failing checks, drill into a specific attempt, or babysit a long-running robot until it finishes. All of that is in the [[patronus]] skill — pick it up when the user asks for status or when offering to follow up. It's reasonable to proactively suggest tracking or babysitting once the Safe Push is in flight; the user can say no. The [[patronus]] skill also documents the hand-off to [[teamcity-cli]] when an individual TC build needs deeper inspection.
Cancel and restore-branch actions stay in the browser.
Use ./safePush.cmd -emergency ONLY for:
DO NOT use Emergency Push for:
When writing tests that spawn external processes (like bazel.cmd), ensure required environment variables are passed:
// In BUILD.bazel for java_test targets:
env_inherit = [
"LOCALAPPDATA", // Windows
"PROCESSOR_ARCHITECTURE", // Windows
"USERPROFILE", // Windows
"HOME", // Unix/macOS - required for bazel cache directory
],
The Bazel sandbox does NOT inherit environment variables by default. Use env_inherit to explicitly pass variables from the host environment to tests.
#ij-builds Slack channel for infrastructure issues#ij-qa-watch#patronus-support, #ij-builds, #ij-qa-watchNo reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Implement or update IntelliJ `AnAction` classes and their `plugin.xml` registrations. Keep labels and descriptions in message bundles, configure optional icons, and launch suspending work through the event coroutine scope so synchronous action handling does not freeze the IDE.
Move IntelliJ JPS test modules to Bazel and diagnose failures that occur only under Bazel. The guide helps trace missing dependencies, plugins, test data, and runfiles, then verify the generated target and migrated tests.
Write or review Kotlin and Java code in line with IntelliJ project conventions. The guidance covers language choice, formatting, sealed types, threading annotations, and safe handling of control flow exceptions so changes fit the surrounding codebase.
Create, amend, or rename IntelliJ commits with messages that follow the project’s ticket and label formats. Keep each commit focused, explain the reason and decision in the body when needed, and pass the message safely through the shell or PowerShell.
Compare how Python type checkers interpret a code snippet or file. The script runs selected tools, including ty, pyrefly, basedpyright, mypy, and zuban, then produces a Markdown report with verdicts, timings, and each checker's diagnostic output.
Shape Compose Desktop UI code by keeping composables focused on rendering and placing application decisions in presenters or coordinators. Get guidance on state ownership, effects, reusable component APIs, lazy-list keys, animation, and performance for individual UI changes.