dagger/container-use v0.4.2

on GitHub

container-use v0.4.2

This is a mostly a patch release, but a feature or 2 snuck in. Notably:

1. Improved lock granularity around export -- fixes issues with cu list and other commands hanging while an agent is running a tool.
2. Added container-use stdio --single-tenant, an experimental mode intended for use with claude code or other CLI based agents. This mode locks each MCP server process to a single environment, reducing the tokens needed for each toolcall and improving subtask performance and accuracy. To try this out, add the --single-tenant flag to your agent's MCP server configuration block.
3. version --system shows versions of our dependencies
4. Environment creation can now take a git ref, so you can prompt "starting from $branch_or_sha, do task" and expect the agent to no longer be concerned with what you've got in your local checkout

Download the pre-compiled binaries from the assets below.

Changelog

• 53d2d5c: Fix linter (#268) (@cwlbraa)
• 301e947: add awesome claude code badge to readme, fixes #255 (#267) (@cwlbraa)
• 2c98c9d: avoid holding locks during Export (#274) (@cwlbraa)
• 57d84de: better manual test instructions (#269) (@cwlbraa)
• 95ad536: feat: add system info to version --system (#266) (@grouville)
• 3dafb4f: feature: allow agent to specify a git ref for environment creation -- this makes it easier to multitask across multiple branches (#273) (@cwlbraa)
• 6c5dff0: fix agent linter instructions so linter output matches CI (#276) (@cwlbraa)
• 4fdd1b5: fix explanation parameter description (#264) (@cwlbraa)
• 5181467: fix permissions tests when not running as root (#265) (@cwlbraa)
• 9932cd4: single tenant mode: lock an mcp process to a single environment (#246) (@cwlbraa)

Full Changelog: v0.4.1...v0.4.2