[sandbox] Add Docker-based package sandbox testing environment #399
Conversation
Anshgrover23
requested review from
Suyashd999 and
mikejmorgan-ai
as code owners
📝 WalkthroughWalkthroughAdds a Docker-backed sandbox subsystem and CLI wiring to create, install, test, promote, exec, list, and clean isolated package test environments; includes implementation, public API exports, CLI help, documentation, and unit tests. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant User
participant CLI as CortexCLI
participant Sandbox as DockerSandbox
participant Docker as Docker Engine
participant Meta as Metadata Store
rect rgba(0,128,96,0.06)
Note over CLI,Sandbox: Create sandbox
User->>CLI: cortex sandbox create <name>
CLI->>Sandbox: create(name, image?)
Sandbox->>Docker: check & pull/run container
Docker-->>Sandbox: container created
Sandbox->>Meta: save SandboxInfo
Sandbox-->>CLI: SandboxExecutionResult(success)
CLI-->>User: success
end
rect rgba(0,96,192,0.06)
Note over CLI,Sandbox: Install & Test
User->>CLI: cortex sandbox install <name> <pkg>
CLI->>Sandbox: install(name, package)
Sandbox->>Docker: apt-get install in container
Docker-->>Sandbox: stdout/stderr/exit
Sandbox->>Meta: update packages_installed
Sandbox-->>CLI: SandboxExecutionResult(success)
User->>CLI: cortex sandbox test <name> [pkg...]
CLI->>Sandbox: test(name, package?)
Sandbox->>Docker: run validation commands
Docker-->>Sandbox: per-package test results
Sandbox-->>CLI: SandboxExecutionResult(test_results)
CLI-->>User: report
end
rect rgba(192,96,0,0.06)
Note over CLI,Sandbox: Promote & Cleanup
User->>CLI: cortex sandbox promote <name> <pkg> [--dry-run]
CLI->>Sandbox: promote(name, package, dry_run)
Sandbox->>Docker: verify package present
Sandbox-->>CLI: prompt/confirm
Sandbox->>Docker: (host) apt-get install (or dry-run)
Sandbox-->>CLI: SandboxExecutionResult(success)
User->>CLI: cortex sandbox cleanup <name>
CLI->>Sandbox: cleanup(name, force?)
Sandbox->>Docker: stop & remove container
Sandbox->>Meta: delete SandboxInfo
Sandbox-->>CLI: SandboxExecutionResult(success)
CLI-->>User: success
end
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Possibly related PRs
Suggested labels
Suggested reviewers
Poem
Pre-merge checks and finishing touches❌ Failed checks (1 inconclusive)
✅ Passed checks (4 passed)
✨ Finishing touches
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
CLA Verification PassedAll contributors have signed the CLA.
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (5)
docs/SANDBOX.md (1)
158-170: Add language specifiers to fenced code blocks.Static analysis flagged missing language specifiers on code blocks. Consider using
textorconsolefor output examples:🔎 Proposed fix
-``` +```text # Output: # ✓ docker.io: package installed # ✓ docker.io: no conflicts -``` +```Apply similar fixes to other output blocks (around lines 169 and 215).
cortex/cli.py (3)
289-298: Redundant import of SandboxTestStatus.
SandboxTestStatusis imported on line 297 within thesandbox()method, and again on line 391 within_sandbox_test(). Consider importing it once at the method level or moving all sandbox imports to the top of thesandbox()method.🔎 Proposed fix
Remove the duplicate import on line 391:
def _sandbox_test(self, sandbox, args: argparse.Namespace) -> int: """Run tests in sandbox.""" - from cortex.sandbox import SandboxTestStatus - name = args.nameThe
SandboxTestStatusfrom the parentsandbox()method import is already in scope.
427-430: Unused result in dry-run path.The
resultfromsandbox.promote(name, package, dry_run=True)on line 428 is captured but never used. The method immediately prints a hardcoded message instead.🔎 Proposed fix
Either use the result message or remove the unnecessary call:
if dry_run: - result = sandbox.promote(name, package, dry_run=True) - cx_print(f"Would run: sudo apt-get install -y {package}", "info") + cx_print(f"Would run: sudo apt-get install -y {package}", "info") return 0Alternatively, use
result.messageif it contains more detailed information.
354-357: Add type hints for thesandboxparameter.Per coding guidelines, type hints are required. The private sandbox action handlers are missing type annotations for the
sandboxparameter.🔎 Proposed fix
+from cortex.sandbox import DockerSandbox # At top of sandbox() or module level + - def _sandbox_create(self, sandbox, args: argparse.Namespace) -> int: + def _sandbox_create(self, sandbox: DockerSandbox, args: argparse.Namespace) -> int: """Create a new sandbox environment."""Apply similar type hints to all private sandbox methods:
_sandbox_install,_sandbox_test,_sandbox_promote,_sandbox_cleanup,_sandbox_list,_sandbox_exec.Also applies to: 372-375, 389-394, 420-425, 457-460, 472-474, 493-496
cortex/sandbox/docker_sandbox.py (1)
326-350: Consider validating sandbox name to prevent edge cases.The sandbox name is used directly in container naming and filesystem paths without validation. Malformed names (e.g., containing
/,.., or shell metacharacters) could cause unexpected behavior.🔎 Proposed fix
Add name validation at the start of
create():def create( self, name: str, image: str | None = None, ) -> SandboxExecutionResult: # Validate sandbox name if not name or not name.replace("-", "").replace("_", "").isalnum(): return SandboxExecutionResult( success=False, message="Sandbox name must contain only alphanumeric characters, hyphens, and underscores", exit_code=1, ) self.require_docker() # ... rest of method
📜 Review details
Configuration used: defaults
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (7)
README.mdcortex/cli.pycortex/sandbox/__init__.pycortex/sandbox/docker_sandbox.pydocs/COMMANDS.mddocs/SANDBOX.mdtests/test_docker_sandbox.py
🧰 Additional context used
📓 Path-based instructions (4)
**/*.py
📄 CodeRabbit inference engine (AGENTS.md)
**/*.py: Follow PEP 8 style guide
Type hints required in Python code
Docstrings required for all public APIs
Files:
tests/test_docker_sandbox.pycortex/sandbox/docker_sandbox.pycortex/sandbox/__init__.pycortex/cli.py
tests/**/*.py
📄 CodeRabbit inference engine (AGENTS.md)
Maintain >80% test coverage for pull requests
Files:
tests/test_docker_sandbox.py
**/*sandbox*.py
📄 CodeRabbit inference engine (AGENTS.md)
Implement Firejail sandboxing for package operations
Files:
tests/test_docker_sandbox.pycortex/sandbox/docker_sandbox.py
{setup.py,setup.cfg,pyproject.toml,**/__init__.py}
📄 CodeRabbit inference engine (AGENTS.md)
Use Python 3.10 or higher as the minimum supported version
Files:
cortex/sandbox/__init__.py
🧠 Learnings (2)
📓 Common learnings
Learnt from: CR
Repo: cortexlinux/cortex PR: 0
File: AGENTS.md:0-0
Timestamp: 2025-12-11T12:03:24.071Z
Learning: Applies to **/*sandbox*.py : Implement Firejail sandboxing for package operations
📚 Learning: 2025-12-11T12:03:24.071Z
Learnt from: CR
Repo: cortexlinux/cortex PR: 0
File: AGENTS.md:0-0
Timestamp: 2025-12-11T12:03:24.071Z
Learning: Applies to **/*sandbox*.py : Implement Firejail sandboxing for package operations
Applied to files:
docs/SANDBOX.mdtests/test_docker_sandbox.pydocs/COMMANDS.mdcortex/sandbox/docker_sandbox.pycortex/sandbox/__init__.pycortex/cli.py
🧬 Code graph analysis (1)
cortex/sandbox/__init__.py (2)
cortex/sandbox/docker_sandbox.py (10)
DockerNotFoundError(107-110)DockerSandbox(125-873)SandboxAlreadyExistsError(119-122)SandboxExecutionResult(95-104)SandboxInfo(60-91)SandboxNotFoundError(113-116)SandboxState(32-38)SandboxTestResult(50-56)SandboxTestStatus(41-46)docker_available(877-879)cortex/sandbox/sandbox_executor.py (3)
CommandBlocked(43-46)ExecutionResult(49-96)SandboxExecutor(99-662)
🪛 markdownlint-cli2 (0.18.1)
docs/SANDBOX.md
158-158: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
169-169: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
215-215: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
249-249: Bare URL used
(MD034, no-bare-urls)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (3)
- GitHub Check: test (3.11)
- GitHub Check: test (3.10)
- GitHub Check: test (3.12)
🔇 Additional comments (19)
README.md (1)
148-148: LGTM!The command reference entry accurately reflects the new sandbox functionality added to the CLI.
docs/COMMANDS.md (1)
17-17: LGTM!Comprehensive documentation with clear examples demonstrating the sandbox workflow. The quick reference entry and detailed command section align well with the CLI implementation.
Also applies to: 265-319
docs/SANDBOX.md (1)
1-59: Well-structured sandbox documentation.The documentation provides a clear overview of the sandbox feature, practical workflow examples, and honest limitations. The troubleshooting section will help users resolve common issues.
Also applies to: 192-296
tests/test_docker_sandbox.py (4)
36-86: Good Docker detection test coverage.Tests properly cover the three scenarios: Docker not installed, Docker installed but daemon not running, and Docker fully available. The exception handling tests verify correct error messages.
263-285: Verify mock sequence for test_test_all_pass.The test mocks four subprocess.run calls in sequence. Ensure the order matches what the implementation actually calls:
docker info(require_docker)which nginx(binary check)nginx --version(version check)dpkg --audit(conflict check)The mock setup appears correct based on the implementation flow.
340-377: Promotion tests cover key scenarios.The tests correctly verify:
- Dry-run mode returns without executing (no subprocess call)
- Attempting to promote a package not in the sandbox fails
- Successful promotion calls the correct host
apt-get installcommand
1-34: Comprehensive test suite with good structure.The test file covers the main DockerSandbox functionality with appropriate mocking. The data class tests (SandboxInfo to_dict/from_dict) and convenience function tests round out the coverage.
Also applies to: 557-612
cortex/sandbox/__init__.py (1)
1-40: Clean package interface with organized exports.The module properly exposes both the Firejail-based
SandboxExecutorand the new Docker-basedDockerSandboxAPIs. The__all__list clearly separates the two sandboxing approaches.cortex/cli.py (2)
1720-1768: Well-structured argparse configuration for sandbox commands.The subparsers correctly define all sandbox actions with appropriate arguments. The
--imagedefault, confirmation flags (-y,--yes), and force options are properly configured.
1897-1898: LGTM!The sandbox command dispatch is correctly integrated into the main CLI routing.
cortex/sandbox/docker_sandbox.py (9)
1-28: Well-documented module header with appropriate imports.The module docstring clearly describes the purpose and features. All necessary imports are present for the implementation.
49-105: Clean dataclass definitions with serialization support.The
SandboxInfo.from_dict()method defensively handles missingpackageskey with a default empty list. All fields have proper type annotations.
188-262: Robust Docker detection with proper error handling.The implementation correctly separates availability checking (
check_docker) from requirement enforcement (require_docker). Caching the docker path avoids repeated lookups, and appropriate timeouts prevent hanging on unresponsive daemons.
673-690: Host promotion uses safe subprocess invocation.The command is constructed as a list (
["sudo", "apt-get", "install", "-y", package]) which prevents shell injection. The package name is validated by apt-get itself.
810-817: Blocked command check only inspects the first argument.The filter can be bypassed with
["bash", "-c", "systemctl start nginx"]. However, since Docker container isolation is the primary security boundary (not command filtering), this provides useful guidance without false security.Consider documenting this limitation or enhancing the check if stricter filtering is desired.
483-639: Comprehensive package testing with smart fallbacks.The test logic handles packages where the binary name differs from the package name (falls back to
dpkg -s). The version/help check tries multiple flags (--version,-v,--help) to maximize compatibility across different programs.
843-873: LGTM!The
is_sandbox_compatibleclassmethod provides a useful filter for the LLM command generator. The simple string split is appropriate for this hint-based filtering use case.
714-754: Best-effort cleanup behavior.The cleanup deletes metadata regardless of whether the container removal succeeded. This is appropriate for cleanup operations where the goal is to reset state, but could leave the container running if removal fails for unexpected reasons. The
--forceflag helps mitigate this.
125-167: Well-designed DockerSandbox class with clear constants.The class docstring provides a clear usage example. The
SANDBOX_BLOCKED_COMMANDSset comprehensively covers system-level commands that don't work in containers.
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (2)
tests/test_docker_sandbox.py (2)
11-20: Simplify imports and path setup.Two minor improvements:
- Line 13: The alias
shutil_moduleis unusual and reduces readability. Use the standardshutilname.- Line 20: Direct
sys.pathmanipulation works but is fragile. Consider usingpytestwith proper package discovery or aconftest.pyinstead.🔎 Suggested simplification
-import shutil as shutil_module +import shutilThen update line 62:
- shutil_module.rmtree(self.temp_dir, ignore_errors=True) + shutil.rmtree(self.temp_dir, ignore_errors=True)
270-280: Consider making the promote test more flexible.Line 279 asserts the exact command format including
sudo. This tightly couples the test to implementation details and assumessudois always used.Consider either:
- Testing that the command contains the package name and uses
apt-get installwithout asserting the exact format, or- Verifying in the test's scope documentation that sudo is an intentional requirement for promotion
Alternative assertion
- call_args = mock_run.call_args[0][0] - self.assertEqual(call_args, ["sudo", "apt-get", "install", "-y", "nginx"]) + call_args = mock_run.call_args[0][0] + # Verify essential components without enforcing exact command structure + self.assertIn("apt-get", call_args) + self.assertIn("install", call_args) + self.assertIn("nginx", call_args)
📜 Review details
Configuration used: defaults
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
tests/test_docker_sandbox.py
🧰 Additional context used
📓 Path-based instructions (3)
**/*.py
📄 CodeRabbit inference engine (AGENTS.md)
**/*.py: Follow PEP 8 style guide
Type hints required in Python code
Docstrings required for all public APIs
Files:
tests/test_docker_sandbox.py
tests/**/*.py
📄 CodeRabbit inference engine (AGENTS.md)
Maintain >80% test coverage for pull requests
Files:
tests/test_docker_sandbox.py
**/*sandbox*.py
📄 CodeRabbit inference engine (AGENTS.md)
Implement Firejail sandboxing for package operations
Files:
tests/test_docker_sandbox.py
🧠 Learnings (2)
📓 Common learnings
Learnt from: CR
Repo: cortexlinux/cortex PR: 0
File: AGENTS.md:0-0
Timestamp: 2025-12-11T12:03:24.071Z
Learning: Applies to **/*sandbox*.py : Implement Firejail sandboxing for package operations
📚 Learning: 2025-12-11T12:03:24.071Z
Learnt from: CR
Repo: cortexlinux/cortex PR: 0
File: AGENTS.md:0-0
Timestamp: 2025-12-11T12:03:24.071Z
Learning: Applies to **/*sandbox*.py : Implement Firejail sandboxing for package operations
Applied to files:
tests/test_docker_sandbox.py
🧬 Code graph analysis (1)
tests/test_docker_sandbox.py (1)
cortex/sandbox/docker_sandbox.py (21)
DockerNotFoundError(107-110)DockerSandbox(125-873)SandboxAlreadyExistsError(119-122)SandboxInfo(60-91)SandboxNotFoundError(113-116)SandboxState(32-38)SandboxTestStatus(41-46)docker_available(877-879)check_docker(188-221)require_docker(223-262)create(326-417)get_sandbox(775-785)install(419-481)test(483-639)promote(641-712)cleanup(714-754)list_sandboxes(756-773)exec_command(787-841)is_sandbox_compatible(844-873)to_dict(70-79)from_dict(82-91)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (3)
- GitHub Check: test (3.11)
- GitHub Check: test (3.12)
- GitHub Check: test (3.10)
🔇 Additional comments (1)
tests/test_docker_sandbox.py (1)
76-423: Excellent test coverage and structure.The test suite demonstrates comprehensive coverage with 11 test classes systematically validating:
- Docker availability detection and error handling
- Complete sandbox lifecycle (create, install, test, promote, cleanup)
- Edge cases (non-existent sandboxes, blocked commands, empty environments)
- Data structure serialization (SandboxInfo to_dict/from_dict)
- Public API helpers (docker_available function)
The organization with dedicated test classes, proper mocking, and isolated fixtures follows testing best practices. This should easily exceed the >80% coverage requirement.
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (2)
tests/test_docker_sandbox.py (2)
21-21: Consider using proper package installation instead of sys.path manipulation.While
sys.path.insertis common in test files, it's fragile and can lead to import issues. Consider using proper package installation (e.g.,pip install -e .) or settingPYTHONPATHin your test runner configuration.
268-273: Consider verifying subprocess.run isn't called in dry-run mode.The test verifies the dry-run message but doesn't explicitly assert that
subprocess.runwasn't called. This would strengthen confidence that dry-run truly skips execution.🔎 Suggested enhancement
@patch("subprocess.run") def test_promote_dry_run(self, mock_run: Mock) -> None: """Test promotion in dry-run mode.""" result = self.create_sandbox_instance().promote("test-env", "nginx", dry_run=True) self.assertTrue(result.success) self.assertIn("Would run", result.message) + mock_run.assert_not_called()
📜 Review details
Configuration used: defaults
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
tests/test_docker_sandbox.py
🧰 Additional context used
📓 Path-based instructions (3)
**/*.py
📄 CodeRabbit inference engine (AGENTS.md)
**/*.py: Follow PEP 8 style guide
Type hints required in Python code
Docstrings required for all public APIs
Files:
tests/test_docker_sandbox.py
tests/**/*.py
📄 CodeRabbit inference engine (AGENTS.md)
Maintain >80% test coverage for pull requests
Files:
tests/test_docker_sandbox.py
**/*sandbox*.py
📄 CodeRabbit inference engine (AGENTS.md)
Implement Firejail sandboxing for package operations
Files:
tests/test_docker_sandbox.py
🧠 Learnings (3)
📓 Common learnings
Learnt from: CR
Repo: cortexlinux/cortex PR: 0
File: AGENTS.md:0-0
Timestamp: 2025-12-11T12:03:24.071Z
Learning: Applies to **/*sandbox*.py : Implement Firejail sandboxing for package operations
📚 Learning: 2025-12-11T12:03:24.071Z
Learnt from: CR
Repo: cortexlinux/cortex PR: 0
File: AGENTS.md:0-0
Timestamp: 2025-12-11T12:03:24.071Z
Learning: Applies to **/*sandbox*.py : Implement Firejail sandboxing for package operations
Applied to files:
tests/test_docker_sandbox.py
📚 Learning: 2025-12-11T12:03:24.071Z
Learnt from: CR
Repo: cortexlinux/cortex PR: 0
File: AGENTS.md:0-0
Timestamp: 2025-12-11T12:03:24.071Z
Learning: Applies to **/*.py : Type hints required in Python code
Applied to files:
tests/test_docker_sandbox.py
🧬 Code graph analysis (1)
tests/test_docker_sandbox.py (1)
cortex/sandbox/docker_sandbox.py (16)
DockerNotFoundError(107-110)DockerSandbox(125-873)SandboxAlreadyExistsError(119-122)SandboxInfo(60-91)SandboxNotFoundError(113-116)SandboxState(32-38)SandboxTestStatus(41-46)docker_available(877-879)check_docker(188-221)require_docker(223-262)create(326-417)install(419-481)test(483-639)is_sandbox_compatible(844-873)to_dict(70-79)from_dict(82-91)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (3)
- GitHub Check: test (3.12)
- GitHub Check: test (3.10)
- GitHub Check: test (3.11)
🔇 Additional comments (13)
tests/test_docker_sandbox.py (13)
35-53: Type hints properly implemented.The helper functions now have complete type hints, resolving the issue flagged in the previous review. Well done!
56-83: Base test class properly implemented.All methods have type hints and docstrings. The test infrastructure provides clean setup/teardown and helper methods for creating test fixtures.
86-132: Comprehensive Docker detection tests.The test class thoroughly covers Docker availability scenarios: not installed, installed but not running, and fully available. Error handling and user-facing messages are properly validated.
134-172: Sandbox creation tests look solid.The tests verify successful creation, metadata persistence, duplicate detection, and custom image support. Good coverage of the create operation.
174-214: Package installation tests cover success and failure paths.The tests verify successful installation, failure handling, and proper exception raising for non-existent sandboxes.
217-252: Sandbox testing functionality well covered.Tests verify both scenarios: packages with passing tests and empty sandboxes with no packages to test.
275-291: Package promotion validation and execution tests are correct.Line 280 properly validates package presence before promotion, and lines 290-291 verify the correct
apt-get installcommand is constructed with sudo.
294-320: Cleanup tests verify metadata removal.The tests properly verify that cleanup removes metadata files and handles the force flag.
323-338: Sandbox listing tests are clean and effective.The tests verify both empty and populated listing scenarios with appropriate assertions.
341-371: Command execution tests including security blocking.Tests verify both successful execution and proper rejection of blocked system-level commands like
systemctl.
374-389: Compatibility checking tests align with blocked command list.Tests properly verify that normal commands pass while system-level commands (systemctl, service, modprobe) are correctly rejected with informative reasons.
392-417: SandboxInfo serialization tests ensure proper roundtrip conversion.Tests verify that
to_dictandfrom_dictcorrectly handle state enum serialization and all field mappings.
420-434: Convenience function tests complete the API coverage.Tests verify the
docker_available()function works correctly in both scenarios.
…rove error messaging
|
@Suyashd999 @Sahilbhatane Can I get a review on this one ? |
…ature/sandbox-docker-testing-issue-130
We already have Firejail implemented, could you change test to use Firejail sandbox only instead of docker based sandbox. also lint after changes |
|
@Sahilbhatane and I discussed over discord we are moving with this approach of docker one only. |
Sounds good! @sujay-d07 cam you review this please? Handing over to @sujay-d07 |
|
@Anshgrover23 can you add a video of this please? Mike has instructed to have videos in PRs henceforth |
|
@Suyashd999 @sujay-d07 @Sahilbhatane Demo Video
|
|
@Anshgrover23 thanks for the video @sujay-d07 is reviewing this and we'll merge this today once approved! |
|
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (1)
.github/scripts/cla_check.py (1)
137-137: Consider full pagination for edge cases.Adding
per_page=100is good, but PRs with over 100 commits won't have all authors checked. For a CLA check, missing authors could allow unsigned contributors through. Consider following theLinkheader for pagination or documenting this as a known limitation.
📜 Review details
Configuration used: defaults
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
.github/scripts/cla_check.py
🧰 Additional context used
🧠 Learnings (1)
📓 Common learnings
Learnt from: CR
Repo: cortexlinux/cortex PR: 0
File: AGENTS.md:0-0
Timestamp: 2025-12-11T12:03:24.071Z
Learning: Applies to **/*sandbox*.py : Implement Firejail sandboxing for package operations
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (4)
- GitHub Check: Build Package
- GitHub Check: test (3.11)
- GitHub Check: test (3.10)
- GitHub Check: test (3.12)
🔇 Additional comments (2)
.github/scripts/cla_check.py (2)
340-348: LGTM on bot patterns formatting.Trailing commas on list items is a good practice for cleaner diffs when adding new patterns.
89-89: LGTM on the type hint modernization.The
str | Noneunion syntax is appropriate for this project, which requires Python 3.10+ (confirmed in bothpyproject.tomlandsetup.py).
3 participants
Related Issue
Closes #130
Summary
Implements Docker-based package sandbox testing as requested in #130. Test packages in isolated containers before installing to the main system.
Acceptance Criteria
Demo
Screenshots
Video
Checklist
pytest tests/)Summary by CodeRabbit
cortex sandboxcommand with subcommands (create, install, test, promote, cleanup, list, exec) for Docker-based isolated package testing, dry-run promotions, and in-sandbox command execution.✏️ Tip: You can customize this high-level summary in your review settings.