# What changed in langchain-anthropic

> 6 changes in langchain-anthropic 1 from its official release notes, upgrade guides and advisories: 1 security, 1 new APIs and options. Each one cited. Sources checked 2026-10-19. HTML: https://unstale.sh/pypi/langchain-anthropic/

To see which of these apply to a repository, run `npx @unstale/cli init` in it.

## langchain-anthropic 1

### Security

- LangChain: Path traversal and sandbox escape in LangChain file-search middleware and loaders (1.4.6) — [security advisory](https://github.com/advisories/GHSA-GR75-JV2W-4656)

### New APIs and options

- Model profiles now declare mid-conversation support, helping developers determine whether a model supports it. (1.7.6) — [release notes](https://github.com/langchain-ai/langchain/releases/tag/langchain-anthropic%3D%3D1.7.6)

### Behavior and default changes

- Tool changes can now occur mid-conversation on SystemMessage. (1.7.4) — [release notes](https://github.com/langchain-ai/langchain/releases/tag/langchain-anthropic%3D%3D1.7.4)

### Fixes and other changes

- Adds compatibility with Claude Sonnet 5.5. (1.7.5) — [release notes](https://github.com/langchain-ai/langchain/releases/tag/langchain-anthropic%3D%3D1.7.5)
- Invalid tool calls are serialized as tool use during replay. (1.7.5) — [release notes](https://github.com/langchain-ai/langchain/releases/tag/langchain-anthropic%3D%3D1.7.5)
- Adds Opus 5.5 and GPT-6 profile augmentations. (1.7.4) — [release notes](https://github.com/langchain-ai/langchain/releases/tag/langchain-anthropic%3D%3D1.7.4)