deepagents-code==0.1.61
langchain-ai/deepagentsdeepagents-code==0.1.61Aug 24, 2026by github-actions[bot]
AI Summary
This release introduces support for Google Anthropic Vertex AI and enforces configured model allowlists to improve security and flexibility. It also brings several user interface enhancements, such as prompt clipboard support and a server-owned `/offload` command.
Key Highlights
- Added `google_anthropic_vertex` provider support
- Enforced configured model allowlists
- Added prompt clipboard support
- Made `/offload` server-owned
New Features
- Added `google_anthropic_vertex` provider support for Claude on Vertex AI
- Enforced configured model allowlists
- Injected goal and rubric context directly, replacing `get_goal` and `get_rubric`
- Made `/offload` server-owned
- Added prompt clipboard support
- Show Auto approval review progress
Full Release Notes
### Features - Added `google_anthropic_vertex` provider support for Claude on Vertex AI ([#5760](https://github.com/langchain-ai/deepagents/issues/5760)). - Enforced configured model allowlists ([#5649](https://github.com/langchain-ai/deepagents/issues/5649)). - Injected goal and rubric context directly, replacing `get_goal` and `get_rubric` ([#5041](https://github.com/langchain-ai/deepagents/issues/5041)). - Made `/offload` server-owned ([#5261](https://github.com/langchain-ai/deepagents/issues/5261)). - Added prompt clipboard support ([#5733](https://github.com/langchain-ai/deepagents/issues/5733)). - Show Auto approval review progress ([#5729](https://github.com/langchain-ai/deepagents/issues/5729)). ### Bug Fixes - Kept long thread resumes responsive ([#5772](https://github.com/langchain-ai/deepagents/issues/5772)). - Render first streamed text immediately ([#5761](https://github.com/langchain-ai/deepagents/issues/5761)). - Show the incognito shell command widget ([#5768](https://github.com/langchain-ai/deepagents/issues/5768)). - Only highlight actionable tool rows ([#5769](https://github.com/langchain-ai/deepagents/issues/5769)). - Warn and ignore `--auto-approve` and `--yolo` in headless mode ([#5750](https://github.com/langchain-ai/deepagents/issues/5750)). - Sweep expired history archives at startup ([#5751](https://github.com/langchain-ai/deepagents/issues/5751)). - Clarified auth environment setup ([#5767](https://github.com/langchain-ai/deepagents/issues/5767)). --- **Special thanks** to everyone who reported the issues addressed in this release: @cakkinep ([#4489](https://github.com/langchain-ai/deepagents/issues/4489)) Internal maintainers: @mdrxy Released by: @mdrxy <details> <summary>Git log since deepagents-code==0.1.60</summary> This commit history includes changes to this package. Commits are listed newest first. - [`23a80f4`](https://github.com/langchain-ai/deepagents/commit/23a80f4d8ef4eef97f1a95440f34cc0f8ec4dffe) release(deepagents-code): 0.1.61 (#5758) - [`b494f43`](https://github.com/langchain-ai/deepagents/commit/b494f43437de3569c8c512d65f5f3accda8e0730) feat(code): make `/offload` server-owned (#5261) - [`85e1e5b`](https://github.com/langchain-ai/deepagents/commit/85e1e5b130b7501308e7e57d0be389f41f5b72df) fix(code): keep long thread resumes responsive (#5772) - [`b78413c`](https://github.com/langchain-ai/deepagents/commit/b78413c708afae7fba806ae81336007d5d4c0c58) fix(code): only highlight actionable tool rows (#5769) - [`d9dc96e`](https://github.com/langchain-ai/deepagents/commit/d9dc96ed555c94f9973e9b49d4b12fd543038f8b) fix(code): sweep expired history archives at startup (#5751) - [`dd8b8ad`](https://github.com/langchain-ai/deepagents/commit/dd8b8ad84aab54dfe1da4a277dfa9539a03e645c) feat(code): enforce configured model allowlists (#5649) - [`a095af9`](https://github.com/langchain-ai/deepagents/commit/a095af93cf3e8ef0c06d9958cfca52dfe9230fcb) fix(code): clarify auth environment setup (#5767) - [`a564f8e`](https://github.com/langchain-ai/deepagents/commit/a564f8eead8eeadaf2c485e4254dcfbc3ff8694c) refactor(code): remove legacy `resolve_scalar` config path (#5755) - [`9bef676`](https://github.com/langchain-ai/deepagents/commit/9bef676fe69a3b79c0702fda59b555d1ef6ad3bc) fix(code): show incognito shell command widget (#5768) - [`f6b0611`](https://github.com/langchain-ai/deepagents/commit/f6b06110d809a3f886d1ee370d1bb353fe7db0d6) style(code): align execute output details (#5770) - [`c2f44ff`](https://github.com/langchain-ai/deepagents/commit/c2f44ffd2759401a5ba31b5f7c41a12b441d6663) feat(code): prompt clipboard (#5733) - [`75095bb`](https://github.com/langchain-ai/deepagents/commit/75095bb81dc200ef33652bcaa173a758312f3962) chore(deps): raise dependency minimums for `all` (#5764) - [`19de73d`](https://github.com/langchain-ai/deepagents/commit/19de73de163d2ca907fa4d9a44f2e19af635aa39) fix(code): render first streamed text immediately (#5761) - [`aa69916`](https://github.com/langchain-ai/deepagents/commit/aa69916baf74c904a70a7497f78752d294df82ac) feat(code): add `google_anthropic_vertex` provider for Claude on Vertex AI (#5760) - [`9223593`](https://github.com/langchain-ai/deepagents/commit/922359307cb0e7728ab58613b29631500e9c5abc) fix(code): warn and ignore `--auto-approve`/`--yolo` in headless mode (#5750) - [`2c80153`](https://github.com/langchain-ai/deepagents/commit/2c801537885ea511c91fc880f6f8daa0a71342be) feat(code,evals): inject goal/rubric context, dropping `get_goal`/`get_rubric` (#5041) - [`8fe11f0`](https://github.com/langchain-ai/deepagents/commit/8fe11f0c2566d95b0ca9af6db9ade7a4c58753e6) feat(code): show Auto approval review progress (#5729) - [`acc752e`](https://github.com/langchain-ai/deepagents/commit/acc752e3e94e986368e13a0cf30bb526b9182480) refactor(code): add configuration provider resolver (#5736) </details>