fix: docker issues

[MODSetter]

Description

Motivation and Context

FIX #

Screenshots

API Changes

• This PR includes API changes

Change Type

• Bug fix
• New feature
• Performance improvement
• Refactoring
• Documentation
• Dependency/Build system
• Breaking change
• Other (specify):

Testing Performed

• Tested locally
• Manual/QA verification

Checklist

• Follows project coding standards and conventions
• Documentation updated as needed
• Dependencies updated as needed
• No lint/build errors or new warnings
• All relevant tests are passing

High-level PR Summary

This PR fixes Docker build issues by ensuring deterministic dependency installation using uv.lock to prevent production drift from newer upstream package versions. The changes include pinning deepagents to <0.5 to avoid breaking imports, removing redundant PyTorch installation from the CUDA index (since PyPI wheels already include CUDA support), and restructuring the Dockerfile to install dependencies from the lockfile before installing the project in editable mode with --no-deps.

⏱️ Estimated Review Time: 5-15 minutes

💡 Review Order Suggestion

Order	File Path
1	surfsense_backend/pyproject.toml
2	surfsense_backend/uv.lock
3	surfsense_backend/Dockerfile

Summary by CodeRabbit

• Chores
  • Modified Docker image build process for dependency installation
  • Updated deepagents dependency version constraint to restrict major version changes

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
surf-sense-frontend	Building	Preview, Comment	May 3, 2026 7:40am

[coderabbitai]

Caution

Review failed

The pull request is closed.

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 9d541bb3-ce1f-4d5a-b887-6906e91d3c50

📥 Commits

Reviewing files that changed from the base of the PR and between 2bd068d and cab1dd6.

⛔ Files ignored due to path filters (1)

• surfsense_backend/uv.lock is excluded by !**/*.lock

📒 Files selected for processing (2)

• surfsense_backend/Dockerfile
• surfsense_backend/pyproject.toml

---

📝 Walkthrough

Walkthrough

The pull request transitions Dockerfile dependency installation from editable mode with pip install -e . to a deterministic lock-based approach, exporting frozen requirements from uv.lock and installing via uv pip install. Additionally, pyproject.toml tightens the deepagents version constraint from >=0.4.12 to >=0.4.12,<0.5.

Changes

Deterministic Lock-Based Dependency Installation

Layer / File(s)	Summary
Version Constraint surfsense_backend/pyproject.toml	deepagents dependency is pinned to >=0.4.12,<0.5 to ensure consistent lock file resolution.
Lock Export & Install surfsense_backend/Dockerfile	uv exports frozen requirements from uv.lock to a temporary file, then installs all dependencies via uv pip install for deterministic resolution.
Build Step Ordering surfsense_backend/Dockerfile	Playwright browser installation (playwright install chromium --with-deps) is moved to run after the lock-based dependency install, ensuring playwright package is available.
Project Installation surfsense_backend/Dockerfile	Source is copied and the project is installed in editable mode with --no-deps to prevent any re-resolution beyond what was locked and pre-installed.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Possibly related PRs

• Fix docker config to run on non-windows architectures #245: Modifies the Dockerfile with an alternative approach to PyTorch installation, conflicting with the removal of architecture-specific PyTorch handling in this PR.

Poem

🐰 Locked and loaded, oh what delight!
No more resolving from left and right,
uv exports the truth from the lock so tight,
Deterministic builds shining bright,
Deepagents bounded, all things right! ✨

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch dev

---

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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Review rate limit: 7/8 reviews remaining, refill in 7 minutes and 30 seconds.}

_{Comment @coderabbitai help to get the list of available commands and usage tips.}