feat: Add --show-thinking flag to CLI (#311) and Show Tool Outputs in a Collapsible Pane #326 #333
+57
−7
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
added 2 commits
November 7, 2025 18:29
Add support for displaying agent reasoning/thinking process in the CLI. Changes: - Add --show-thinking command-line flag to main.py - Extend SessionState with show_thinking field in config.py - Display reasoning blocks in dim style when flag is enabled in execution.py - Update help documentation and examples in ui.py When enabled with --show-thinking, the agent's reasoning blocks from LangChain content blocks are displayed in real-time as dim text, providing transparency into the agent's thought process. Resolves langchain-ai#311
Add universal tool output toggle that works with all tools including MCP servers. Changes: - Add show_tool_outputs field and toggle method to SessionState in config.py - Add Ctrl+O keybinding to toggle tool outputs in input.py - Display tool outputs in dim cyan when toggle is enabled in execution.py - Add toolbar indicator showing current toggle state - Update help documentation and tips to include Ctrl+O When enabled, tool outputs are displayed using existing format_tool_message_content() function. No smart parsing needed - works universally with any tool type. Features: - Universal: Works with ALL tools (file ops, web search, MCP servers, custom tools) - Session-persistent: Toggle stays active throughout session - Visual indicator: Toolbar shows current state (cyan when ON, gray when OFF) - Clean display: Outputs shown in dim cyan without interfering with existing UI
dayal-arnav05
changed the title
ma1688
added a commit
to HK-CCASS/deepagents-hk
that referenced
this pull request
Nov 10, 2025
…chain-ai#333) - Add --show-thinking CLI flag to display agent reasoning process - Add Ctrl+O keybinding to toggle tool output visibility - Update toolbar to show tool output state with color indicators - Display tool outputs with ↳ symbol in dim cyan style - Update help documentation and README with new features - Add comprehensive testing guide (docs/PR333_TESTING_GUIDE.md) Changes: - src/cli/config.py: Add show_thinking and show_tool_outputs to SessionState - src/cli/execution.py: Add logic to display tool outputs and reasoning - src/cli/input.py: Add Ctrl+O keybinding and toolbar indicators - src/cli/main.py: Add --show-thinking argument and tips - src/cli/ui.py: Update help documentation - README.md: Document new CLI features - docs/PR333_TESTING_GUIDE.md: Add comprehensive testing guide
ma1688
added a commit
to HK-CCASS/deepagents-hk
that referenced
this pull request
Nov 10, 2025
Implements PR langchain-ai#333: Add --show-thinking flag and Ctrl+O tool output toggle for enhanced CLI observability
ma1688
added a commit
to HK-CCASS/deepagents-hk
that referenced
this pull request
Nov 10, 2025
- Add CLI enhancements to architecture diagram:
* Show Thinking (--show-thinking flag)
* Tool Outputs Toggle (Ctrl+O)
* Auto-Approve Toggle (Ctrl+T)
- Update Storage layer to reflect Virtual Mode:
* Mark all backends with [Virtual Mode]
* Highlight Default Filesystem with sandbox security
* Explain path mapping: /file.txt → {cwd}/file.txt
- Update Key Features section:
* File System Sandbox (Virtual Mode)
* CLI Enhancements (Show Thinking + Tool Outputs)
* Better debugging and understanding capabilities
Reflects changes from PR langchain-ai#331 and PR langchain-ai#333
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add support for displaying agent reasoning/thinking process in the CLI.
Changes:
When enabled with --show-thinking, the agent's reasoning blocks from LangChain content blocks are displayed in real-time as dim text, providing transparency into the agent's thought process.
Resolves #311