deepagents-code 0.1.69__tar.gz → 0.1.71__tar.gz

{deepagents_code-0.1.69 → deepagents_code-0.1.71}/AGENTS.md

@@ -76,7 +76,7 @@ Apply these rules to new UI; do not treat them as a mandate to refactor existing
 ### Testing Textual apps
 
 - Use `textual.pilot` for async UI testing - see [Testing guide](https://textual.textualize.io/guide/testing/)
-- Snapshot testing available for visual regression - see repo `notes/snapshot_testing.md`
+- Snapshot testing available for visual regression
 - For modal flows, test the real interaction path with keypresses when possible. Unit tests that call action methods or resume handlers directly can miss focus and modal-stack bugs.
 - A modal test that pushes the screen onto a bare `App` host cannot catch app-level key interception. The app owns a priority `shift+tab -> toggle_auto_approve` binding (`app.py`) that wins dispatch over any screen binding; `action_toggle_auto_approve` routes the key to the active screen. Cursor-style modals (highlighted rows, no focusable children) just implement `action_move_up` — the router's `_SupportsReverseNav` protocol branch picks them up automatically. The protocol matches on method *presence*, so enrollment is opt-out, not opt-in: a screen that defines `action_move_up` for its own arrow keys gets `shift+tab` reverse-nav whether or not it was written with that key in mind. Screens needing different behavior (`focus_previous`, tab bars) — or none — need an explicit branch *above* the protocol check in `action_toggle_auto_approve`. Add an app-level regression test in `test_app.py` when a screen needs behavior other than swallowing the key.
 - Do not open another modal or refocus the base chat input directly inside a modal dismiss callback. Preserve the non-blocking `push_screen(..., callback)` flow and schedule follow-up UI work with `call_after_refresh` so the dismissing modal fully unwinds first.

{deepagents_code-0.1.69 → deepagents_code-0.1.71}/ARCHITECTURE.md

@@ -85,6 +85,4 @@ The main cost is the client/server boundary. When debugging, first decide which
 - For cost estimates and local pricing overrides (`prices.json`), see
   [`PRICING.md`](./PRICING.md).
 - For security boundaries, see [`THREAT_MODEL.md`](./THREAT_MODEL.md).
-- For how a transient model failure is retried after output has already streamed,
-  see [`STREAMING_RETRY_DESIGN.md`](./STREAMING_RETRY_DESIGN.md).
 - For package-specific coding conventions, see [`AGENTS.md`](./AGENTS.md).

{deepagents_code-0.1.69 → deepagents_code-0.1.71}/CHANGELOG.md

@@ -2,6 +2,30 @@
 
 # Deep Agents Code Changelog
 
+## [0.1.71](https://github.com/langchain-ai/deepagents/compare/deepagents-code==0.1.70...deepagents-code==0.1.71) (2026-09-18)
+
+### Features
+
+- Moved MCP support onto FastMCP and `langchain.mcp`. ([#5922](https://github.com/langchain-ai/deepagents/pull/5922))
+- Condensed the current working directory in the status bar from the middle, preserving the end of the path. ([#6252](https://github.com/langchain-ai/deepagents/pull/6252))
+
+### Bug Fixes
+
+- Replaced the composer with prompt search and prevented query feedback. ([#6346](https://github.com/langchain-ai/deepagents/pull/6346))
+- Kept absolute paths intact in normal input mode. ([#6379](https://github.com/langchain-ai/deepagents/pull/6379))
+- Pinned resumed history to the newest message. ([#6225](https://github.com/langchain-ai/deepagents/pull/6225))
+- Improved handling of overflowing pasted content in the chat input so it scrolls correctly. ([#6382](https://github.com/langchain-ai/deepagents/pull/6382))
+- Wrapped debug logs before the scrollbar. ([#6219](https://github.com/langchain-ai/deepagents/pull/6219))
+
+## [0.1.70](https://github.com/langchain-ai/deepagents/compare/deepagents-code==0.1.69...deepagents-code==0.1.70) (2026-09-16)
+
+### Bug Fixes
+
+- Fixed prompt clipboard word deletion behavior. ([#6345](https://github.com/langchain-ai/deepagents/pull/6345))
+- Clarified unknown model effort in footer. ([#6337](https://github.com/langchain-ai/deepagents/pull/6337))
+- Prevented duplicate clicks in the footer model and effort pickers. ([#6344](https://github.com/langchain-ai/deepagents/pull/6344))
+- Delayed hook status display in the footer to reduce UI noise. ([#6322](https://github.com/langchain-ai/deepagents/pull/6322))
+
 ## [0.1.69](https://github.com/langchain-ai/deepagents/compare/deepagents-code==0.1.68...deepagents-code==0.1.69) (2026-09-14)
 
 ### Features

{deepagents_code-0.1.69 → deepagents_code-0.1.71}/DEVELOPMENT.md

@@ -143,7 +143,7 @@ To send logs elsewhere, also `export DEEPAGENTS_CODE_DEBUG_DIRECTORY=<path>`. Ea
 
 The directory is created or tightened to owner-only access, and each file is created or tightened to user-only access. Symlinks are refused. If either cannot be secured, no file handler is attached and a warning goes to stderr. Use the in-app Debug Console in that case.
 
-Stdio MCP server stderr is captured here at `DEBUG`. This keeps server-side failures visible when the TUI cannot show process stderr. Each record is one line, capped at 4096 characters. Characters in the Unicode `C` categories are removed, which includes control and format characters. The ESC byte of an ANSI sequence is removed but the rest stays as literal text, so the log is not free of escape-sequence residue. The text comes from the server. It can contain credentials or other sensitive values. Enable `DEBUG` logging only if you accept that risk, and do not share the log file.
+Raw stdio MCP server stderr is discarded to protect the TUI and prevent unbounded disk growth, including in debug mode. Structured MCP log notifications still reach the application logger with the server name and severity.
 
 ### In-app Debug Console (`Ctrl+\`)
 

{deepagents_code-0.1.69 → deepagents_code-0.1.71}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: deepagents-code
-Version: 0.1.69
+Version: 0.1.71
 Summary: Terminal coding agent built on the Deep Agents SDK - works with any model, with persistent memory, customizable skills, subagents, and approval controls.
 Project-URL: Homepage, https://www.langchain.com/dcode
 Project-URL: Documentation, https://docs.langchain.com/oss/deepagents/code/overview
@@ -30,23 +30,23 @@ Requires-Python: <4.0,>=3.12
 Requires-Dist: aiosqlite<1.0.0,>=0.22.1
 Requires-Dist: anyio<5.0.0,>=4.14.2
 Requires-Dist: deepagents-acp<1.0.0,>=0.0.10
-Requires-Dist: deepagents==0.7.14
+Requires-Dist: deepagents==0.7.15
+Requires-Dist: fastmcp<5.0.0,>=4.0.0b4
 Requires-Dist: filelock<3.30,>=3.29.7
-Requires-Dist: genai-prices<0.2.0,>=0.1.6
+Requires-Dist: genai-prices<0.2.0,>=0.1.7
 Requires-Dist: httpx<1.0.0,>=0.28.1
 Requires-Dist: langchain-anthropic<2.0.0,>=1.7.2
 Requires-Dist: langchain-google-genai<5.0.0,>=4.4.0
-Requires-Dist: langchain-mcp-adapters<1.0.0,>=0.3.2
 Requires-Dist: langchain-openai<2.0.0,>=1.6.2
 Requires-Dist: langchain-quickjs<0.4.0,>=0.3.4
-Requires-Dist: langchain<2.0.0,>=1.4.0
+Requires-Dist: langchain<2.0.0,>=1.4.1
 Requires-Dist: langgraph-checkpoint-sqlite<4.0.0,>=3.1.1
 Requires-Dist: langgraph-cli[inmem]<1.0.0,>=0.4.31
-Requires-Dist: langgraph-runtime-inmem<1.0.0,>=0.34.0
+Requires-Dist: langgraph-runtime-inmem<1.0.0,>=0.34.1
 Requires-Dist: langgraph-sdk<1.0.0,>=0.4.4
-Requires-Dist: langsmith[sandbox]>=0.12.4
+Requires-Dist: langsmith[sandbox]>=0.12.6
 Requires-Dist: markdownify<2.0.0,>=1.2.3
-Requires-Dist: mcp<2.0.0,>=1.28.1
+Requires-Dist: mcp<3.0.0,>=2.0.0
 Requires-Dist: packaging>=26.2
 Requires-Dist: pillow<13.0.0,>=12.3.0
 Requires-Dist: prompt-toolkit<4.0.0,>=3.0.53
@@ -67,7 +67,7 @@ Requires-Dist: langchain-agentcore-codeinterpreter<1.0.0,>=0.0.5; extra == 'agen
 Provides-Extra: all-providers
 Requires-Dist: aiohttp<3.15.0,>=3.14.3; extra == 'all-providers'
 Requires-Dist: langchain-anthropic<2.0.0,>=1.7.2; extra == 'all-providers'
-Requires-Dist: langchain-aws<2.0.0,>=1.7.6; extra == 'all-providers'
+Requires-Dist: langchain-aws<2.0.0,>=1.7.8; extra == 'all-providers'
 Requires-Dist: langchain-baseten<1.0.0,>=0.2.4; extra == 'all-providers'
 Requires-Dist: langchain-cohere<1.0.0,>=0.6.0; extra == 'all-providers'
 Requires-Dist: langchain-deepseek<2.0.0,>=1.1.0; extra == 'all-providers'
@@ -77,7 +77,7 @@ Requires-Dist: langchain-google-vertexai<4.0.0,>=3.2.4; extra == 'all-providers'
 Requires-Dist: langchain-groq<2.0.0,>=1.1.3; extra == 'all-providers'
 Requires-Dist: langchain-huggingface<2.0.0,>=1.2.2; extra == 'all-providers'
 Requires-Dist: langchain-ibm<2.0.0,>=1.1.0; extra == 'all-providers'
-Requires-Dist: langchain-litellm<2.0.0,>=0.7.1; extra == 'all-providers'
+Requires-Dist: langchain-litellm<2.0.0,>=0.7.2; extra == 'all-providers'
 Requires-Dist: langchain-meta<2.0.0,>=0.5.0; extra == 'all-providers'
 Requires-Dist: langchain-mistralai<2.0.0,>=1.1.6; extra == 'all-providers'
 Requires-Dist: langchain-nvidia-ai-endpoints<2.0.0,>=1.4.3; extra == 'all-providers'
@@ -98,7 +98,7 @@ Requires-Dist: langchain-anthropic<2.0.0,>=1.7.2; extra == 'anthropic'
 Provides-Extra: baseten
 Requires-Dist: langchain-baseten<1.0.0,>=0.2.4; extra == 'baseten'
 Provides-Extra: bedrock
-Requires-Dist: langchain-aws<2.0.0,>=1.7.6; extra == 'bedrock'
+Requires-Dist: langchain-aws<2.0.0,>=1.7.8; extra == 'bedrock'
 Provides-Extra: cohere
 Requires-Dist: langchain-cohere<1.0.0,>=0.6.0; extra == 'cohere'
 Provides-Extra: daytona
@@ -116,7 +116,7 @@ Requires-Dist: langchain-huggingface<2.0.0,>=1.2.2; extra == 'huggingface'
 Provides-Extra: ibm
 Requires-Dist: langchain-ibm<2.0.0,>=1.1.0; extra == 'ibm'
 Provides-Extra: litellm
-Requires-Dist: langchain-litellm<2.0.0,>=0.7.1; extra == 'litellm'
+Requires-Dist: langchain-litellm<2.0.0,>=0.7.2; extra == 'litellm'
 Provides-Extra: media
 Requires-Dist: av<19.0.0,>=18.0.0; extra == 'media'
 Requires-Dist: pillow<13.0.0,>=12.3.0; extra == 'media'

{deepagents_code-0.1.69 → deepagents_code-0.1.71}/deepagents_code/_build_info.py

@@ -1,3 +1,3 @@
 """Generated at build time. Do not edit or commit."""
 
-BUILD_COMMIT = "1d3232c"
+BUILD_COMMIT = "d1674e6"

{deepagents_code-0.1.69 → deepagents_code-0.1.71}/deepagents_code/_version.py

@@ -2,7 +2,7 @@
 
 # Keep the `x-release-please-version` annotation — release-please uses it to
 # bump `__version__` in sync with `pyproject.toml` on every release PR.
-__version__ = "0.1.69"  # x-release-please-version
+__version__ = "0.1.71"  # x-release-please-version
 
 DOCS_URL = "https://docs.langchain.com/oss/python/deepagents/code"
 """URL for `deepagents-code` documentation."""

{deepagents_code-0.1.69 → deepagents_code-0.1.71}/deepagents_code/app.py

@@ -1173,6 +1173,9 @@ or the user queues input while waiting, the status bar reveals the connection
 state as the single app-wide progress indicator.
 """
 
+_HOOK_STATUS_REVEAL_DELAY_SECONDS = 2.0
+"""Seconds to hide hook progress so fast hooks do not flash in the footer."""
+
 _UPDATE_RECHECK_INTERVAL_SECONDS = 60 * 60
 """How often long-running TUI sessions quietly re-check for app updates."""
 
@@ -2039,7 +2042,7 @@ Shared by every such command -- `/effort`, `/summarization-model` -- so the
 habit transfers and the accepted spellings cannot drift apart.
 """
 
-_UNKNOWN_EFFORT_LABEL = "effort"
+_UNKNOWN_EFFORT_LABEL = "effort?"
 
 
 def _parse_reconnect_args(rest: str) -> tuple[bool, bool]:
@@ -2960,6 +2963,7 @@ class _ChatScroll(VerticalScroll):
         """
         super().__init__(*args, **kwargs)
         self._follow_bottom_when_scrollable = False
+        self._bottom_follow_generation = 0
 
     def anchor(self, anchor: bool = True) -> None:
         """Anchor only once the transcript is tall enough to scroll.
@@ -2976,6 +2980,8 @@ class _ChatScroll(VerticalScroll):
                 delegate to the base class.
         """
         self._follow_bottom_when_scrollable = anchor
+        if anchor:
+            self._bottom_follow_generation += 1
         if not anchor:
             super().anchor(False)
             return
@@ -4078,6 +4084,9 @@ class DeepAgentsApp(App):
         self._agent_running = False
         """True while the agent worker is streaming a response."""
 
+        self._footer_picker_requests: set[str] = set()
+        """Footer picker commands currently being submitted."""
+
         self._agent_reconciling = False
         """True while turn-end checkpoint state is being synchronized."""
 
@@ -4153,6 +4162,15 @@ class DeepAgentsApp(App):
         self._connection_status_reveal_timer: Timer | None = None
         """One-shot timer that reveals deferred status-bar connection progress."""
 
+        self._pending_hook_status_message = ""
+        """Latest hook progress text waiting for its footer reveal delay."""
+
+        self._hook_status_visible = False
+        """Whether hook progress currently owns the visible status-message slot."""
+
+        self._hook_status_reveal_timer: Timer | None = None
+        """One-shot timer that reveals continuously active hook progress."""
+
         self._connection_ready_event = asyncio.Event()
         """Set once the initial server connection has either succeeded or failed."""
         if not self._connecting:
@@ -4328,6 +4346,9 @@ class DeepAgentsApp(App):
         self._history_prefetch_active = False
         """Whether resumed history is warming toward the soft window size."""
 
+        self._history_prefetch_anchor_generation: int | None = None
+        """Bottom-follow generation owned by resumed-history prefetch."""
+
         self._transcript_prune_timer: Timer | None = None
         """Idle timer that defers opposite-edge pruning while scrolling."""
 
@@ -4811,6 +4832,8 @@ class DeepAgentsApp(App):
         self._apply_scrollbar_visibility(chat)
 
         self._status_bar = self.query_one("#status-bar", StatusBar)
+        if self._pending_hook_status_message:
+            self._schedule_hook_status_reveal_timer()
         self._goal_status_panel = self.query_one("#goal-status-panel", GoalStatusPanel)
         self._chat_input = self.query_one("#input-area", ChatInput)
         model_spec = self._effective_model_spec()
@@ -5318,9 +5341,44 @@ class DeepAgentsApp(App):
         self.notify(message, severity=severity, markup=False)
 
     def _update_hook_status(self, message: str) -> None:
-        """Update the status bar with hook-owned progress text."""
-        if self._status_bar:
-            self._status_bar.set_status_message(message, source="hooks")
+        """Delay hook-owned progress text so fast hooks do not flash."""
+        self._pending_hook_status_message = message
+        if not message:
+            self._hook_status_visible = False
+            self._cancel_hook_status_reveal_timer()
+            if self._status_bar:
+                self._status_bar.set_status_message("", source="hooks")
+            return
+        if self._hook_status_visible:
+            if self._status_bar:
+                self._status_bar.set_status_message(message, source="hooks")
+            return
+        self._schedule_hook_status_reveal_timer()
+
+    def _schedule_hook_status_reveal_timer(self) -> None:
+        """Schedule the one-shot timer that reveals active hook progress."""
+        if self._hook_status_reveal_timer is not None or not self._running:
+            return
+        self._hook_status_reveal_timer = self.set_timer(
+            _HOOK_STATUS_REVEAL_DELAY_SECONDS,
+            self._on_hook_status_reveal_timer,
+        )
+
+    def _cancel_hook_status_reveal_timer(self) -> None:
+        """Cancel and clear the deferred hook-status reveal timer."""
+        if self._hook_status_reveal_timer is None:
+            return
+        self._hook_status_reveal_timer.stop()
+        self._hook_status_reveal_timer = None
+
+    def _on_hook_status_reveal_timer(self) -> None:
+        """Reveal the latest hook progress after continuous activity."""
+        self._hook_status_reveal_timer = None
+        message = self._pending_hook_status_message
+        if not message or self._status_bar is None:
+            return
+        self._hook_status_visible = True
+        self._status_bar.set_status_message(message, source="hooks")
 
     async def _init_session_state(self) -> None:
         """Create session state and load its Hooks v2 manager.
@@ -9401,7 +9459,7 @@ class DeepAgentsApp(App):
             self._hydration_scheduled = False
 
         if hydrated_count == 0 and direction == "above":
-            self._history_prefetch_active = False
+            self._stop_history_prefetch()
         if hydrated_count or self._hydration_requests:
             self.call_after_refresh(lambda: self._continue_hydration(direction))
 
@@ -9421,7 +9479,7 @@ class DeepAgentsApp(App):
             ):
                 self._request_hydration("above")
                 return
-            self._history_prefetch_active = False
+            self._stop_history_prefetch()
             return
 
         if direction == "above":
@@ -9435,8 +9493,26 @@ class DeepAgentsApp(App):
             self._message_store.has_messages_above
             and self._message_store.visible_count < self._message_store.WINDOW_SIZE
         )
-        if self._history_prefetch_active:
-            self._request_hydration("above")
+        if not self._history_prefetch_active:
+            return
+        self._history_prefetch_anchor_generation = None
+        with suppress(NoMatches):
+            chat = self.query_one("#chat", _ChatScroll)
+            chat.anchor()
+            self._history_prefetch_anchor_generation = chat._bottom_follow_generation
+        self._request_hydration("above")
+
+    def _stop_history_prefetch(self) -> None:
+        """End resumed-history warming and its temporary bottom anchor."""
+        if not self._history_prefetch_active:
+            return
+        self._history_prefetch_active = False
+        generation = self._history_prefetch_anchor_generation
+        self._history_prefetch_anchor_generation = None
+        with suppress(NoMatches):
+            chat = self.query_one("#chat", _ChatScroll)
+            if generation == chat._bottom_follow_generation:
+                chat.anchor(False)
 
     def _check_hydration_needed(self) -> None:
         """Prefetch older messages near the mounted-window boundary."""
@@ -20807,6 +20883,9 @@ class DeepAgentsApp(App):
         self._pending_shell_messages.clear()
         self._hydration_requests.clear()
         self._history_prefetch_active = False
+        self._history_prefetch_anchor_generation = None
+        with suppress(NoMatches):
+            self.query_one("#chat", _ChatScroll).anchor(False)
         if self._transcript_prune_timer is not None:
             self._transcript_prune_timer.stop()
             self._transcript_prune_timer = None
@@ -21740,6 +21819,7 @@ class DeepAgentsApp(App):
         # active workers so their subprocesses are terminated
         # (SIGTERM → SIGKILL) instead of being orphaned.
         self._cancel_connection_status_reveal_timer()
+        self._cancel_hook_status_reveal_timer()
         self._discard_queue()
 
         if self._shell_running and self._shell_worker:
@@ -23522,6 +23602,26 @@ class DeepAgentsApp(App):
     # Model Switching
     # =========================================================================
 
+    async def _submit_footer_picker(self, command: str) -> None:
+        """Submit a footer picker unless the same request is open or queued."""
+        from deepagents_code.tui.widgets.effort_selector import EffortSelectorScreen
+        from deepagents_code.tui.widgets.model_selector import ModelSelectorScreen
+
+        screen_type = (
+            ModelSelectorScreen if command == "/model" else EffortSelectorScreen
+        )
+        if (
+            command in self._footer_picker_requests
+            or isinstance(self.screen, screen_type)
+            or any(message.text == command for message in self._pending_messages)
+        ):
+            return
+        self._footer_picker_requests.add(command)
+        try:
+            await self._submit_input(command, "command")
+        finally:
+            self._footer_picker_requests.discard(command)
+
     async def action_open_model_selector(self) -> None:
         """Open the model selector via `/model`.
 
@@ -23531,7 +23631,7 @@ class DeepAgentsApp(App):
         tip is dismissed. The bare form is `IMMEDIATE_UI`, so it still bypasses
         the queue and opens while the agent is busy.
         """
-        await self._submit_input("/model", "command")
+        await self._submit_footer_picker("/model")
 
     async def action_open_effort_selector(self) -> None:
         """Open the reasoning effort picker via `/effort`.
@@ -23539,7 +23639,7 @@ class DeepAgentsApp(App):
         `/effort` is `QUEUED`, so it must go through `_submit_input` to keep its
         place behind any pending input instead of jumping an in-flight turn.
         """
-        await self._submit_input("/effort", "command")
+        await self._submit_footer_picker("/effort")
 
     def _build_model_selector_screen(
         self,
@@ -25115,7 +25215,7 @@ class DeepAgentsApp(App):
             started_at = self._first_invocation_at
             if started_at is None:
                 return "not started"
-            return format_duration(max(0.0, time.monotonic() - started_at))
+            return format_duration(int(max(0.0, time.monotonic() - started_at)))
 
         def _model_field() -> SnapshotField:
             # Built directly (not via `_safe`) so the copyable metadata tracks

{deepagents_code-0.1.69 → deepagents_code-0.1.71}/deepagents_code/config.py

@@ -4817,7 +4817,7 @@ def configure_langsmith_secret_redaction() -> bool:
         _fail_closed_disable_tracing()
         return False
 
-    logger.info("LangSmith secret redaction enabled for agent traces.")
+    logger.debug("LangSmith secret redaction enabled for agent traces.")
     return True