langchain-ai/deepagents - CodSpeed

deepagents

Blog Docs Changelog

Performance History

Latest Results

style(cli): hide redundant exit code 0 from tool output (#2354) The SDK appends `[Command succeeded with exit code 0]` to every successful tool execution result. In the TUI this is visual noise — success is already conveyed by the widget's status indicator. Strip the trailer in `ToolCallMessage.set_success` so it never renders, while preserving non-zero exit code lines which flow through `set_error` and remain useful for debugging.

main

33 minutes ago

style(cli): hide redundant exit code 0 from tool output

mdrxy/cli/truncate-success-code

37 minutes ago

fix(cli): mark token count as approximate after interrupted generation (#2353) When a user interrupts generation mid-stream, the status bar previously showed the last-known token count with no indication that the real context is larger. The displayed value is now marked stale with a trailing "+" (e.g. "12.5K+ tokens") until the next model response reports fresh usage. Also deduplicates the identical `CancelledError`/`KeyboardInterrupt` cleanup handlers into a single shared function.

main

39 minutes ago

fix(cli): mark token count as approximate after interrupted generation

mdrxy/cli/stale-tokens

43 minutes ago

style(cli): defer mode glyph and popup updates atomically, update input bg (#2352) Fix visual timing between mode glyph and completion popup by deferring both CSS class mutations and prompt updates into `call_after_refresh` callbacks. Previously, `watch_mode` applied CSS classes synchronously but updated the glyph in a deferred callback, creating a one-frame desync. The backspace-exits-mode path had the same category of issue — resetting the completion popup synchronously while the glyph change was deferred. Remove selection arrow and leading slashes for slash command selector. Remove `ChatTextArea` background. Remove top border in model switcher's metadata section. <details> <summary>chat input - before</summary> <img width="233" height="282" alt="Screenshot 14" src="https://github.com/user-attachments/assets/134b76cf-82e3-4fff-9cd7-e387505ce956" /> </details> <details> <summary>chat input - after</summary> <img width="363" height="278" alt="Screenshot 10" src="https://github.com/user-attachments/assets/6883afb5-13ac-459d-aa2b-6f26595829f4" /> </details> <details> <summary>model switcher - before</summary> <img width="728" height="166" alt="Screenshot 15" src="https://github.com/user-attachments/assets/a3b5e0ea-5274-435a-a808-43eff14ff113" /> </details> <details> <summary>model switcher - after</summary> <img width="687" height="181" alt="Screenshot 13" src="https://github.com/user-attachments/assets/16ca34f6-ca08-4ad6-9c44-954550f93959" /> </details>

main

1 hour ago

style(cli): defer mode glyph and popup updates atomically, update input bg

mdrxy/cli/ui

1 hour ago

chore: bump pygments to 2.20.0 everywhere (GHSA-5239-wwwm-4pmq) (#2349) ## Summary - Bumps `pygments` from `2.19.2` → `2.20.0` across all 12 `uv.lock` files - Fixes GHSA-5239-wwwm-4pmq (low severity) flagged by Dependabot in all sub-packages and examples Affected lock files: - `libs/partners/{runloop,quickjs,modal,daytona}/uv.lock` - `libs/{evals,deepagents,cli,acp}/uv.lock` - `examples/{text-to-sql-agent,nvidia_deep_agent,deep_research,content-builder-agent}/uv.lock` ## Release Note Patch deps ### Test Plan - [x] CI Green :pray:

main

3 hours ago

bump pygments to 2.20.0 across all lock files (GHSA-5239-wwwm-4pmq) Resolves 12 open Dependabot alerts for Pygments < 2.20.0 (low severity). Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

bump/pygments-2.20.0

4 hours ago

Latest Branches

0%

style(cli): hide redundant exit code 0 from tool output#2354

45 minutes ago

704692a

mdrxy/cli/truncate-success-code

0%

fix(cli): mark token count as approximate after interrupted generation#2353

51 minutes ago

8074ef2

mdrxy/cli/stale-tokens

0%

style(cli): defer mode glyph and popup updates atomically, update input bg#2352

2 hours ago

e8af566

mdrxy/cli/ui

© 2026 CodSpeed Technology

Home Terms Privacy Docs