deepagents-code 0.1.6__tar.gz → 0.1.7__tar.gz

{deepagents_code-0.1.6 → deepagents_code-0.1.7}/AGENTS.md

@@ -65,6 +65,14 @@ Textual's `App.notify(message)` parses the message string as Rich markup by defa
 - Features that run shell commands silently must be opt-in, never default-enabled. Gate behind an explicit env var or config key.
 - Background workers that spawn subprocesses must set a timeout to avoid blocking indefinitely.
 
+## Logging
+
+Debug logging is configured **once**, on the `deepagents_code` package logger, by the `configure_debug_logging` call in `deepagents_code/__init__.py`. Child module loggers (`logging.getLogger(__name__)`) reach the shared debug file via propagation.
+
+- Do **not** add per-module `configure_debug_logging(logger)` calls. They are redundant now that the package logger is configured at import, and they reintroduce the duplicate-handler problem the single-config approach solves.
+- Every module should create its logger with `logging.getLogger(__name__)` so it stays inside the `deepagents_code.*` hierarchy and inherits the package handler. Don't set `logger.propagate = False` or attach your own handlers.
+- The handler only attaches when `DEEPAGENTS_CODE_DEBUG` is truthy; the no-op path is a single env-var read, so it's safe on the startup hot path. See `DEV.md` for the `DEEPAGENTS_CODE_DEBUG` / `DEEPAGENTS_CODE_DEBUG_FILE` env vars.
+
 ## CLI help screen
 
 The `deepagents-code --help` screen is hand-maintained in `ui.show_help()`, separate from the argparse definitions in `main.parse_args()`. When adding a new CLI flag, update **both** files. A drift-detection test (`test_args.TestHelpScreenDrift`) fails if a flag is registered in argparse but missing from the help screen.

{deepagents_code-0.1.6 → deepagents_code-0.1.7}/CHANGELOG.md

@@ -1,5 +1,28 @@
+<!-- markdownlint-disable MD024 -->
+
 # Deep Agents Code Changelog
 
+## [0.1.7](https://github.com/langchain-ai/deepagents/compare/deepagents-code==0.1.6...deepagents-code==0.1.7) (2026-05-30)
+
+### Features
+
+* Add toggleable message timestamp footers ([#3662](https://github.com/langchain-ai/deepagents/issues/3662)) ([977e110](https://github.com/langchain-ai/deepagents/commit/977e11006cfbd78fbaba4e7bb2a13acf6b788652))
+
+### Bug Fixes
+
+* Fix zero tool MCP server rendering ([#3649](https://github.com/langchain-ai/deepagents/issues/3649)) ([7e7a567](https://github.com/langchain-ai/deepagents/commit/7e7a567556110ad927a78b45c3a3d4ac37b65e86))
+* Centralize debug logging setup to package root ([#3650](https://github.com/langchain-ai/deepagents/issues/3650)) ([5145ed1](https://github.com/langchain-ai/deepagents/commit/5145ed1f8296f41d78c905c2ce899d2742f7dc9b))
+* Char-truncate execute tool preview output ([#3627](https://github.com/langchain-ai/deepagents/issues/3627)) ([bb276e2](https://github.com/langchain-ai/deepagents/commit/bb276e2c41177b0dfe6ffd44fd37a293fbfdcb27))
+* Handle stale slash-command `Enter` before completion popup renders ([#3647](https://github.com/langchain-ai/deepagents/issues/3647)) ([9a28742](https://github.com/langchain-ai/deepagents/commit/9a287424e86d5d52d0a328388c3fe453b160f597))
+* Keep chat input focused when clicking a message ([#3655](https://github.com/langchain-ai/deepagents/issues/3655)) ([daf6571](https://github.com/langchain-ai/deepagents/commit/daf65716d7c999eadb2b7c37e412ec07b2c7aed3))
+* Mention `Ctrl+R` in MCP reconnect toast ([#3622](https://github.com/langchain-ai/deepagents/issues/3622)) ([3b4b086](https://github.com/langchain-ai/deepagents/commit/3b4b0867665e58959073e660d85b74c700acaa1e))
+* Prevent duplicate-id crash on MCP reconnect and clipboard `NoScreen` ([#3632](https://github.com/langchain-ai/deepagents/issues/3632)) ([6b9a3c0](https://github.com/langchain-ai/deepagents/commit/6b9a3c051586c26c542e958849e952d08a4b5a88))
+* Reconstruct message counts for `DeltaChannel` threads from writes table ([#3668](https://github.com/langchain-ai/deepagents/issues/3668)) ([27e1940](https://github.com/langchain-ai/deepagents/commit/27e1940a924abfc999126cf46024003f453ba0c8))
+* Render MCP tool errors and drop empty-string optional params ([#3624](https://github.com/langchain-ai/deepagents/issues/3624)) ([fdf3db4](https://github.com/langchain-ai/deepagents/commit/fdf3db464cd9f3de4e84c246547dd2971d26c726))
+* Respect line width in tool output previews ([#3646](https://github.com/langchain-ai/deepagents/issues/3646)) ([ba1ad2d](https://github.com/langchain-ai/deepagents/commit/ba1ad2dbabd19b3821490537465a3bcd39c6fed6))
+* Restore resumed thread model ([#3651](https://github.com/langchain-ai/deepagents/issues/3651)) ([550a8ab](https://github.com/langchain-ai/deepagents/commit/550a8abf3c595d738162a97f694b5d9527613323))
+* Tool spinner, result formatting, and expand-hint fixes ([#3661](https://github.com/langchain-ai/deepagents/issues/3661)) ([54485a3](https://github.com/langchain-ai/deepagents/commit/54485a305854f46a6ce00ae4df51f3301c652a38))
+
 ## [0.1.6](https://github.com/langchain-ai/deepagents/compare/deepagents-code==0.1.5...deepagents-code==0.1.6) (2026-05-27)
 
 ### Features

{deepagents_code-0.1.6 → deepagents_code-0.1.7}/COMMANDS.md

@@ -6,7 +6,7 @@ Canonical list of slash commands for `deepagents-code`, derived from
 regenerate after editing the registry.
 
 
-## Public (26)
+## Public (27)
 
 | Command | Aliases | Description |
 | --- | --- | --- |
@@ -32,6 +32,7 @@ regenerate after editing the registry.
 | `/skill-creator` |  | Guide for creating effective agent skills |
 | `/theme` |  | Switch color theme |
 | `/threads` |  | Browse and resume previous threads |
+| `/timestamps` |  | Toggle message timestamp footers |
 | `/tokens` |  | Token usage |
 | `/trace` |  | Open current thread in LangSmith |
 | `/update` |  | Check for and install updates |

{deepagents_code-0.1.6 → deepagents_code-0.1.7}/DEV.md

@@ -60,10 +60,23 @@ The app runs a `langgraph dev` subprocess for every interactive session. When th
 | Variable | Effect |
 | --- | --- |
 | `DEEPAGENTS_CODE_DEBUG=1` | Preserves the server subprocess log on shutdown and prints its path to stderr. Without this, the log is deleted when the process stops. Also enables the app-process file handler below. Accepts `1`/`true`/`yes`/`on` (case-insensitive) as enabled; `0`/`false`/`no`/`off`/empty/unset as disabled. |
-| `DEEPAGENTS_CODE_DEBUG_FILE=<path>` | Overrides the default path (`/tmp/deepagents_debug.log`) for the app-process file handler, which attaches at `DEBUG` level to `textual_adapter` and `remote_client`. **Only takes effect when `DEEPAGENTS_CODE_DEBUG` is truthy.** Useful for diagnosing streaming/client-side issues; does **not** capture the server subprocess. |
+| `DEEPAGENTS_CODE_DEBUG_FILE=<path>` | Overrides the default path (`/tmp/deepagents_debug.log`) for the app-process file handler, which attaches at `DEBUG` level to the `deepagents_code` package logger. **Only takes effect when `DEEPAGENTS_CODE_DEBUG` is truthy.** Useful for diagnosing client-side app issues; does **not** capture the server subprocess. |
 
 `DEEPAGENTS_CODE_DEBUG` is what you want for startup crashes (graph init, MCP config, sandbox): the preserved subprocess log contains the real traceback. The optional `DEEPAGENTS_CODE_DEBUG_FILE` override is for post-startup client-side debugging.
 
+To capture client-side logs while reproducing an issue:
+
+```bash
+cd libs/code
+DEEPAGENTS_CODE_DEBUG=1 uv run deepagents-code
+```
+
+Then in another terminal:
+
+```bash
+tail -f /tmp/deepagents_debug.log
+```
+
 ### Finding the server subprocess log
 
 On macOS, `tempfile` resolves to `$TMPDIR` (a path under `/var/folders/.../T/`). Each `ServerProcess` writes its combined stdout+stderr to a file matching `deepagents_server_log_*.txt`:

{deepagents_code-0.1.6 → deepagents_code-0.1.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deepagents-code
-Version: 0.1.6
+Version: 0.1.7
 Summary: Terminal interface for Deep Agents - interactive AI agent with file operations, shell access, and sub-agent capabilities.
 Project-URL: Homepage, https://docs.langchain.com/oss/python/deepagents/overview
 Project-URL: Documentation, https://reference.langchain.com/python/deepagents/
@@ -27,7 +27,7 @@ Classifier: Topic :: Terminals
 Requires-Python: <4.0,>=3.11
 Requires-Dist: aiosqlite<1.0.0,>=0.19.0
 Requires-Dist: deepagents-acp>=0.0.4
-Requires-Dist: deepagents==0.6.4
+Requires-Dist: deepagents==0.6.7
 Requires-Dist: httpx<1.0.0,>=0.28.1
 Requires-Dist: langchain-anthropic<2.0.0,>=1.4.3
 Requires-Dist: langchain-google-genai<5.0.0,>=4.2.2

{deepagents_code-0.1.6 → deepagents_code-0.1.7}/deepagents_code/__init__.py

@@ -2,13 +2,17 @@
 
 from __future__ import annotations
 
+import logging
 from typing import TYPE_CHECKING
 
+from deepagents_code._debug import configure_debug_logging
 from deepagents_code._version import __version__
 
 if TYPE_CHECKING:
     from collections.abc import Callable
 
+configure_debug_logging(logging.getLogger(__name__))  # noqa: RUF067  # package logger must be configured before child modules emit logs
+
 __all__ = [
     "__version__",
     "cli_main",  # noqa: F822  # resolved lazily by __getattr__

{deepagents_code-0.1.6 → deepagents_code-0.1.7}/deepagents_code/_cli_context.py

@@ -25,3 +25,14 @@ class CLIContext(TypedDict, total=False):
     model_params: dict[str, Any]
     """Invocation params (e.g. `temperature`, `max_tokens`) to merge
     into `model_settings`."""
+
+    effective_model: str | None
+    """Resolved `provider:model` spec actually in use for this invocation.
+
+    Unlike `model` (a swap *instruction* that is `None` when the base model is
+    used), this carries the model in effect — whether from a `/model` override
+    or the startup default — so `ResumeStateMiddleware` can record it to
+    checkpoint state for restore on resume. `None` when no usable spec is
+    resolved yet (e.g. credentials not configured), in which case nothing is
+    recorded rather than a malformed spec.
+    """

{deepagents_code-0.1.6 → deepagents_code-0.1.7}/deepagents_code/_debug.py

@@ -10,17 +10,27 @@ from __future__ import annotations
 
 import logging
 import os
+import sys
 from pathlib import Path
 
 from deepagents_code._env_vars import DEBUG, DEBUG_FILE, is_env_truthy
 
+_DEBUG_HANDLER_ATTR = "_deepagents_code_debug_handler"
+
 
 def configure_debug_logging(target: logging.Logger) -> None:
     """Attach a file handler to *target* when `DEEPAGENTS_CODE_DEBUG` is set.
 
+    Intended to be called once on the `deepagents_code` package logger; child
+    module loggers reach the same file via propagation, so individual modules do
+    not configure logging themselves.
+
     The log file defaults to `'/tmp/deepagents_debug.log'` but can be overridden
-    with `DEEPAGENTS_CODE_DEBUG_FILE`. The handler appends so that multiple
-    modules share the same log file across a session.
+    with `DEEPAGENTS_CODE_DEBUG_FILE`. The handler appends (`mode='a'`) so logs
+    are preserved across separate process runs. Calling this again with the same
+    resolved path is a no-op: the existing tagged handler is reused rather than
+    stacking duplicates. If the resolved path changes, the stale handler is
+    closed and replaced.
 
     Does nothing when `DEEPAGENTS_CODE_DEBUG` is not truthy (see `is_env_truthy`).
 
@@ -36,16 +46,30 @@ def configure_debug_logging(target: logging.Logger) -> None:
             "/tmp/deepagents_debug.log",  # noqa: S108
         )
     )
+    for existing in list(target.handlers):
+        if not (
+            isinstance(existing, logging.FileHandler)
+            and getattr(existing, _DEBUG_HANDLER_ATTR, False)
+        ):
+            continue
+        if Path(existing.baseFilename) == debug_path:
+            # Already configured for this path; reuse rather than duplicate.
+            target.setLevel(logging.DEBUG)
+            return
+        # The debug path changed; drop the stale handler before re-attaching so
+        # we don't leak its file descriptor or fan logs out to two files.
+        target.removeHandler(existing)
+        existing.close()
+
     try:
         handler = logging.FileHandler(str(debug_path), mode="a")
     except OSError as exc:
-        import sys
-
         print(  # noqa: T201
             f"Warning: could not open debug log file {debug_path}: {exc}",
             file=sys.stderr,
         )
         return
+    setattr(handler, _DEBUG_HANDLER_ATTR, True)
     handler.setLevel(logging.DEBUG)
     handler.setFormatter(logging.Formatter("%(asctime)s %(name)s %(message)s"))
     target.addHandler(handler)

{deepagents_code-0.1.6 → deepagents_code-0.1.7}/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__ = "0.1.6"  # x-release-please-version
+__version__ = "0.1.7"  # x-release-please-version
 
 DOCS_URL = "https://docs.langchain.com/oss/python/deepagents/code"
 """URL for `deepagents-code` documentation."""

{deepagents_code-0.1.6 → deepagents_code-0.1.7}/deepagents_code/agent.py

@@ -1243,12 +1243,13 @@ def create_cli_agent(
     agent_middleware = []
     agent_middleware.append(ConfigurableModelMiddleware())
 
-    # Token state: declares the `_context_tokens` channel and writes it from
-    # `after_model` based on the latest `AIMessage.usage_metadata`. The CLI
-    # reads it back from `state_values` on thread resume.
-    from deepagents_code.token_state import TokenStateMiddleware
+    # Resume state: declares the `_context_tokens` and `_model_spec` channels
+    # and writes them from `after_model` (token count from the latest
+    # `AIMessage.usage_metadata`, model spec from `context["effective_model"]`).
+    # The CLI reads them back from `state_values` on thread resume.
+    from deepagents_code.resume_state import ResumeStateMiddleware
 
-    agent_middleware.append(TokenStateMiddleware())
+    agent_middleware.append(ResumeStateMiddleware())
 
     # Add ask_user middleware (must be early so its tool is available)
     if enable_ask_user: