langchain-ai/docs
1
0
Fork 0
mirror of https://github.com/langchain-ai/docs.git synced 2026-07-22 04:15:25 -04:00
Code Issues 490 Packages Projects Releases Wiki Activity

[PR #4024] docs: clarify Deep Agents context compression, add multimodal page [DOC-803] #4028

New Issue
Open
opened 2026-06-05 18:53:57 -04:00 by yindo · 0 comments
yindo commented 2026-06-05 18:53:57 -04:00
Owner
Copy Link
Copy Source

📋 Pull Request Information

Original PR: https://github.com/langchain-ai/docs/pull/4024
Author: @open-swe[bot]
Created: 5/15/2026
Status: 🔄 Open

Base: main ← Head: open-swe/deep-agents-context-multimodal

📝 Commits (1)

  • 6aeff22 docs(deepagents): clarify context compression, add multimodal page [DOC-803]

📊 Changes

6 files changed (+407 additions, -3 deletions)

View changed files

📝 src/docs.json (+2 -0)
📝 src/oss/deepagents/backends.mdx (+1 -1)
📝 src/oss/deepagents/context-engineering.mdx (+86 -0)
📝 src/oss/deepagents/harness.mdx (+4 -2)
➕ src/oss/deepagents/multimodal.mdx (+227 -0)
📝 src/oss/langchain/tools.mdx (+87 -0)

📄 Description

Description

Adds Deep Agents documentation for context compression and multimodality, plus multimodal tool return types in the LangChain tools guide. Addresses the three documentation needs in DOC-803.

  • Context compression: Expanded the existing section in context-engineering.mdx to spell out that compression is middleware-based, tunable, replaceable, and that custom middleware can rewrite the chat history via wrap_model_call / wrapModelCall. Cross-linked from harness.mdx.
  • Multimodal in Deep Agents: New oss/deepagents/multimodal.mdx covering user input, read_file binary support, tool returns (text + image, interleaved), audio/video caveats, and provider compatibility. Wired into the Core capabilities group for both Python and TypeScript. Also fixed the broken #multimodal-and-binary-files anchor in backends.mdx to point at the new page.
  • LangChain tools: Added a Return multimodal content subsection to tools.mdx documenting list-of-content-block returns and the ToolMessage workaround for audio/video.

Verified make build succeeds, no cross-ref warnings, no scope warnings on the changed files.

Release Note

none

Test Plan

  • Verify the new Deep Agents Multimodal page renders correctly under both Python and TypeScript dropdowns.
  • Verify the rewritten Context compression section in Context engineering reads cleanly and the rewrite-chat-history example compiles mentally for both languages.

Opened collaboratively by Naomi Pentrel and open-swe.

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/langchain-ai/docs/pull/4024 **Author:** [@open-swe[bot]](https://github.com/apps/open-swe) **Created:** 5/15/2026 **Status:** 🔄 Open **Base:** `main` ← **Head:** `open-swe/deep-agents-context-multimodal` --- ### 📝 Commits (1) - [`6aeff22`](https://github.com/langchain-ai/docs/commit/6aeff22f9380a9425727ae77bf5eeea14dae44f8) docs(deepagents): clarify context compression, add multimodal page [DOC-803] ### 📊 Changes **6 files changed** (+407 additions, -3 deletions) <details> <summary>View changed files</summary> 📝 `src/docs.json` (+2 -0) 📝 `src/oss/deepagents/backends.mdx` (+1 -1) 📝 `src/oss/deepagents/context-engineering.mdx` (+86 -0) 📝 `src/oss/deepagents/harness.mdx` (+4 -2) ➕ `src/oss/deepagents/multimodal.mdx` (+227 -0) 📝 `src/oss/langchain/tools.mdx` (+87 -0) </details> ### 📄 Description ## Description Adds Deep Agents documentation for context compression and multimodality, plus multimodal tool return types in the LangChain tools guide. Addresses the three documentation needs in DOC-803. - **Context compression**: Expanded the existing section in `context-engineering.mdx` to spell out that compression is middleware-based, tunable, replaceable, and that custom middleware can rewrite the chat history via `wrap_model_call` / `wrapModelCall`. Cross-linked from `harness.mdx`. - **Multimodal in Deep Agents**: New `oss/deepagents/multimodal.mdx` covering user input, `read_file` binary support, tool returns (text + image, interleaved), audio/video caveats, and provider compatibility. Wired into the Core capabilities group for both Python and TypeScript. Also fixed the broken `#multimodal-and-binary-files` anchor in `backends.mdx` to point at the new page. - **LangChain tools**: Added a Return multimodal content subsection to `tools.mdx` documenting list-of-content-block returns and the ToolMessage workaround for audio/video. Verified `make build` succeeds, no cross-ref warnings, no scope warnings on the changed files. ## Release Note none ## Test Plan - [ ] Verify the new Deep Agents Multimodal page renders correctly under both Python and TypeScript dropdowns. - [ ] Verify the rewritten Context compression section in Context engineering reads cleanly and the rewrite-chat-history example compiles mentally for both languages. _Opened collaboratively by Naomi Pentrel and open-swe._ --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
yindo added the pull-request label 2026-06-05 18:53:57 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
deepagents
docs-infra
external
external
help wanted
integration
internal
internal
javascript
js/ts
langchain
langgraph
langsmith
needs-title
pull-request
Mirrored from GitHub Pull Request
 python
reference
No Label pull-request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langchain-ai/docs#4028
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?