deepagents-code 0.1.0__tar.gz → 0.1.2__tar.gz

deepagents_code-0.1.2/CHANGELOG.md

@@ -0,0 +1,32 @@
+# Deep Agents Code Changelog
+
+## [0.1.2](https://github.com/langchain-ai/deepagents/compare/deepagents-code==0.1.1...deepagents-code==0.1.2) (2026-05-19)
+
+### Features
+
+* `/model` toggle for recommended-only list ([#3453](https://github.com/langchain-ai/deepagents/issues/3453)) ([c326b7e](https://github.com/langchain-ai/deepagents/commit/c326b7ec1b9940861175e0466ab4221f03e2bcba))
+* `--timeout` flag for non-interactive ([#3351](https://github.com/langchain-ai/deepagents/issues/3351)) ([44e86ab](https://github.com/langchain-ai/deepagents/commit/44e86abbb1870f689dace8b1be6ed430d65e74c1))
+* Browser loopback OAuth callback for MCP auth ([#3467](https://github.com/langchain-ai/deepagents/issues/3467)) ([d83aa07](https://github.com/langchain-ai/deepagents/commit/d83aa07c818af35800f81d062a147fa45a47ace7))
+* MCP screen metadata ([#3349](https://github.com/langchain-ai/deepagents/issues/3349)) ([ce2f07e](https://github.com/langchain-ai/deepagents/commit/ce2f07e7211f22b3f44a1a232088b89a469a0a99))
+
+### Bug Fixes
+
+* Drop sections from `system_prompt.md` already supplied by SDK middleware ([#3448](https://github.com/langchain-ai/deepagents/issues/3448)) ([9dbf2c2](https://github.com/langchain-ai/deepagents/commit/9dbf2c2f19e941e012d0c93418ef09fb56f30d6a))
+* Rename stale usage commands ([#3460](https://github.com/langchain-ai/deepagents/issues/3460)) ([da43b7f](https://github.com/langchain-ai/deepagents/commit/da43b7f9d913e6190ff03c496a269faf08bbf182))
+* Suppress interrupt-cleanup state writes from traces ([#3465](https://github.com/langchain-ai/deepagents/issues/3465)) ([319b24e](https://github.com/langchain-ai/deepagents/commit/319b24e6f179eaf56f105a6db683901c82fe95be))
+
+## [0.1.1](https://github.com/langchain-ai/deepagents/compare/deepagents-code==0.1.0...deepagents-code==0.1.1) (2026-05-16)
+
+### Bug Fixes
+
+* Correct LangSmith sandbox working directory ([#3415](https://github.com/langchain-ai/deepagents/issues/3415)) ([b0e8d83](https://github.com/langchain-ai/deepagents/commit/b0e8d83f97a2a698268173a839000c84e8368324))
+* Guard `fetch_url` against SSRF ([#3411](https://github.com/langchain-ai/deepagents/issues/3411)) ([54d8521](https://github.com/langchain-ai/deepagents/commit/54d8521976940dfe147ead4b56565360241335be))
+
+## [0.1.0](https://github.com/langchain-ai/deepagents/compare/deepagents-code==0.0.1...deepagents-code==0.1.0) (2026-05-12)
+
+Hello world! Ported from `libs/cli`.
+
+---
+
+`deepagents-code` was forked from `deepagents-cli` at v0.1.0 (2026-05-12).
+For history prior to the fork, see [the `deepagents-cli` changelog](https://github.com/langchain-ai/deepagents/blob/main/libs/cli/CHANGELOG.md).

{deepagents_code-0.1.0 → deepagents_code-0.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deepagents-code
-Version: 0.1.0
+Version: 0.1.2
 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/
@@ -27,13 +27,13 @@ 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==0.6.1
+Requires-Dist: deepagents==0.6.2
 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
 Requires-Dist: langchain-mcp-adapters<1.0.0,>=0.2.2
 Requires-Dist: langchain-openai<2.0.0,>=1.2.1
-Requires-Dist: langchain<2.0.0,>=1.3.0
+Requires-Dist: langchain<2.0.0,>=1.3.1
 Requires-Dist: langgraph-checkpoint-sqlite<4.0.0,>=3.1.0
 Requires-Dist: langgraph-cli[inmem]<1.0.0,>=0.4.24
 Requires-Dist: langgraph-runtime-inmem<1.0.0,>=0.28.1
@@ -145,13 +145,13 @@ Description-Content-Type: text/markdown
 ## Quick Install
 
 ```bash
-curl -LsSf https://langch.in/gh-da-cli | bash
+curl -LsSf https://langch.in/dcode | bash
 ```
 
 ```bash
 # With model provider extras
 # OpenAI, Anthropic, and Gemini are included by default
-DEEPAGENTS_EXTRAS="nvidia,ollama" curl -LsSf https://langch.in/gh-da-cli | bash
+DEEPAGENTS_EXTRAS="nvidia,ollama" curl -LsSf https://langch.in/dcode | bash
 ```
 
 Or install directly with `uv`:
@@ -184,7 +184,7 @@ The fastest way to start using Deep Agents. `deepagents-code` is a pre-built cod
 
 ## 📖 Resources
 
-- **[CLI Documentation](https://docs.langchain.com/oss/python/deepagents/cli/overview)**
+- **[Documentation](https://docs.langchain.com/oss/python/deepagents/code/overview)**
 - **[Changelog](https://github.com/langchain-ai/deepagents/blob/main/libs/code/CHANGELOG.md)**
 - **[Source code](https://github.com/langchain-ai/deepagents/tree/main/libs/code)**
 - **[Deep Agents SDK](https://github.com/langchain-ai/deepagents)** — underlying agent harness

{deepagents_code-0.1.0 → deepagents_code-0.1.2}/README.md

@@ -12,13 +12,13 @@
 ## Quick Install
 
 ```bash
-curl -LsSf https://langch.in/gh-da-cli | bash
+curl -LsSf https://langch.in/dcode | bash
 ```
 
 ```bash
 # With model provider extras
 # OpenAI, Anthropic, and Gemini are included by default
-DEEPAGENTS_EXTRAS="nvidia,ollama" curl -LsSf https://langch.in/gh-da-cli | bash
+DEEPAGENTS_EXTRAS="nvidia,ollama" curl -LsSf https://langch.in/dcode | bash
 ```
 
 Or install directly with `uv`:
@@ -51,7 +51,7 @@ The fastest way to start using Deep Agents. `deepagents-code` is a pre-built cod
 
 ## 📖 Resources
 
-- **[CLI Documentation](https://docs.langchain.com/oss/python/deepagents/cli/overview)**
+- **[Documentation](https://docs.langchain.com/oss/python/deepagents/code/overview)**
 - **[Changelog](https://github.com/langchain-ai/deepagents/blob/main/libs/code/CHANGELOG.md)**
 - **[Source code](https://github.com/langchain-ai/deepagents/tree/main/libs/code)**
 - **[Deep Agents SDK](https://github.com/langchain-ai/deepagents)** — underlying agent harness

{deepagents_code-0.1.0 → deepagents_code-0.1.2}/deepagents_code/_version.py

@@ -2,9 +2,9 @@
 
 # Keep the `x-release-please-version` annotation — release-please uses it to
 # bump `__version__` in sync with `pyproject.toml` on every release PR.
-__version__ = "0.1.0"  # x-release-please-version
+__version__ = "0.1.2"  # x-release-please-version
 
-DOCS_URL = "https://docs.langchain.com/oss/python/deepagents/cli"
+DOCS_URL = "https://docs.langchain.com/oss/python/deepagents/code"
 """URL for `deepagents-code` documentation."""
 
 PYPI_URL = "https://pypi.org/pypi/deepagents-code/json"

{deepagents_code-0.1.0 → deepagents_code-0.1.2}/deepagents_code/agent.py

@@ -420,7 +420,7 @@ def reset_agent(
             console.print(
                 f"[bold red]Error:[/bold red] Source agent '{source_agent}' not found "
                 "or has no AGENTS.md\n"
-                "  Available agents: deepagents agents list"
+                "  Available agents: dcode agents list"
             )
             raise SystemExit(1)
 

{deepagents_code-0.1.0 → deepagents_code-0.1.2}/deepagents_code/app.py

@@ -1266,7 +1266,7 @@ class DeepAgentsApp(App):
         self._mcp_unauthenticated = sum(
             1 for s in (mcp_server_info or []) if s.status == "unauthenticated"
         )
-        """MCP servers awaiting a `deepagents mcp login` run."""
+        """MCP servers awaiting a `dcode mcp login` run."""
 
         self._mcp_errored = sum(
             1 for s in (mcp_server_info or []) if s.status == "error"
@@ -2350,7 +2350,11 @@ class DeepAgentsApp(App):
             await self._await_prewarm_imports()
 
             from deepagents_code.config import create_model
-            from deepagents_code.model_config import ModelConfigError, save_recent_model
+            from deepagents_code.model_config import (
+                ModelConfigError,
+                save_recent_model,
+                touch_recent_model,
+            )
 
             try:
                 async with self._deepagents_import_lock:
@@ -2359,7 +2363,9 @@ class DeepAgentsApp(App):
                 self.post_message(self.ServerStartFailed(error=exc))
                 return
             result.apply_to_settings()
-            save_recent_model(f"{result.provider}:{result.model_name}")
+            resolved_spec = f"{result.provider}:{result.model_name}"
+            save_recent_model(resolved_spec)
+            touch_recent_model(resolved_spec)
             self._model_kwargs = None  # consumed
 
         from deepagents_code.server_manager import start_server_and_get_agent
@@ -5508,6 +5514,8 @@ class DeepAgentsApp(App):
             if result.offload_warning:
                 await self._mount_message(ErrorMessage(result.offload_warning))
 
+            # Intentionally traced: the summarization event is a meaningful state
+            # transition that should surface in LangSmith alongside real agent turns.
             await self._agent.aupdate_state(
                 config, {"_summarization_event": result.new_event}
             )
@@ -7421,7 +7429,7 @@ class DeepAgentsApp(App):
             # `previous_thread_has_agent_output` above.
             if previous_thread_id and previous_thread_has_agent_output:
                 resume_hint = Content.from_markup(
-                    "[dim]Relaunch with[/dim] deepagents -r $thread "
+                    "[dim]Relaunch with[/dim] dcode -r $thread "
                     "[dim]to resume the previous thread.[/dim]",
                     thread=previous_thread_id,
                 )
@@ -8253,6 +8261,7 @@ class DeepAgentsApp(App):
             ProviderAuthState,
             get_provider_auth_status,
             save_recent_model,
+            touch_recent_model,
         )
 
         logger.info("Switching model to %s", model_spec)
@@ -8359,11 +8368,12 @@ class DeepAgentsApp(App):
                 display = f"{provider}:{model_name}"
 
             try:
-                create_model(
+                result = create_model(
                     display,
                     extra_kwargs=extra_kwargs,
                     profile_overrides=self._profile_override,
-                ).apply_to_settings()
+                )
+                result.apply_to_settings()
             except Exception as exc:
                 logger.exception("Failed to resolve model metadata for %s", display)
                 await self._mount_message(
@@ -8395,6 +8405,13 @@ class DeepAgentsApp(App):
                 await self._mount_message(
                     AppMessage(f"Switched to {display}{params_suffix}")
                 )
+            # Best-effort MRU update for the `/model` Recent section.
+            # `display` may be a bare model name when provider auto-detection
+            # fails; use the post-resolution spec so touch_recent_model always
+            # gets a valid "provider:model" string. Silent on failure —
+            # debug log captures it when DEEPAGENTS_CODE_DEBUG=1.
+            resolved_spec = f"{result.provider}:{result.model_name}"
+            await asyncio.to_thread(touch_recent_model, resolved_spec)
             logger.info(
                 "Model switched to %s (via configurable middleware); model_params=%s",
                 display,

{deepagents_code-0.1.0 → deepagents_code-0.1.2}/deepagents_code/app.tcss

@@ -24,7 +24,7 @@ Screen {
 /* Chat area - main scrollable messages area */
 #chat {
     height: 1fr;
-    padding: 1 2;
+    padding: 1 2 0 2;
     background: $background;
 }
 
@@ -44,7 +44,7 @@ Screen {
 /* Bottom app container - holds ChatInput (now inside scroll) */
 #bottom-app-container {
     height: auto;
-    margin-top: 1;
+    margin-top: 0;
 }
 
 /* Input area */

{deepagents_code-0.1.0 → deepagents_code-0.1.2}/deepagents_code/config.py

@@ -2277,8 +2277,8 @@ def create_model(
 
     Examples:
         >>> model = create_model("anthropic:claude-sonnet-4-5")
-        >>> model = create_model("openai:gpt-4o")
-        >>> model = create_model("gpt-4o")  # Auto-detects openai
+        >>> model = create_model("openai:gpt-5.5")
+        >>> model = create_model("gpt-5.5")  # Auto-detects openai
         >>> model = create_model()  # Uses environment defaults
     """
     from deepagents_code.model_config import (

{deepagents_code-0.1.0 → deepagents_code-0.1.2}/deepagents_code/integrations/sandbox_factory.py

@@ -73,7 +73,7 @@ def _run_sandbox_setup(backend: SandboxBackendProtocol, setup_script_path: str)
 _PROVIDER_TO_WORKING_DIR = {
     "agentcore": "/tmp",  # noqa: S108 # AgentCore Code Interpreter working directory
     "daytona": "/home/daytona",
-    "langsmith": "/tmp",  # noqa: S108  # LangSmith sandbox working directory
+    "langsmith": "/root",  # `$HOME` in the LangSmith sandbox
     "modal": "/workspace",
     "runloop": "/home/user",
 }

{deepagents_code-0.1.0 → deepagents_code-0.1.2}/deepagents_code/main.py

@@ -829,6 +829,17 @@ def parse_args() -> argparse.Namespace:
         "to prevent runaway agents. Requires -n or piped stdin.",
     )
 
+    parser.add_argument(
+        "--timeout",
+        dest="timeout",
+        type=positive_int,
+        metavar="SECONDS",
+        help="Hard wall-clock timeout in seconds. The agent is cancelled and "
+        "the process exits with code 124 if the timeout is reached. "
+        "Complements --max-turns (turn count) with a time-based limit; both "
+        "use exit code 124 on expiry. Requires -n or piped stdin.",
+    )
+
     parser.add_argument(
         "--stdin",
         action="store_true",
@@ -1165,7 +1176,11 @@ async def _run_acp_cli_async(
     """
     from deepagents_code.agent import create_cli_agent, load_async_subagents
     from deepagents_code.config import create_model, settings
-    from deepagents_code.model_config import ModelConfigError, save_recent_model
+    from deepagents_code.model_config import (
+        ModelConfigError,
+        save_recent_model,
+        touch_recent_model,
+    )
     from deepagents_code.tools import fetch_url, web_search
 
     try:
@@ -1181,7 +1196,9 @@ async def _run_acp_cli_async(
     model_result.apply_to_settings()
 
     # Persist the resolved model so [models].recent is always populated.
-    save_recent_model(f"{model_result.provider}:{model_result.model_name}")
+    resolved_spec = f"{model_result.provider}:{model_result.model_name}"
+    save_recent_model(resolved_spec)
+    touch_recent_model(resolved_spec)
 
     tools: list[Any] = [fetch_url]
     if settings.has_tavily:
@@ -1264,7 +1281,7 @@ def apply_stdin_pipe(args: argparse.Namespace) -> None:
         piped text to it (the CLI still runs non-interactively):
 
         ```bash
-        cat context.txt | deepagents -n "summarize this"
+        cat context.txt | dcode -n "summarize this"
         # non_interactive_message = "{contents of context.txt}\n\nsummarize this"
         ```
 
@@ -1272,7 +1289,7 @@ def apply_stdin_pipe(args: argparse.Namespace) -> None:
         the piped text to it (the CLI still runs interactively):
 
         ```bash
-        cat error.log | deepagents -m "explain this"
+        cat error.log | dcode -m "explain this"
         # initial_prompt = "{contents of error.log}\n\nexplain this"
         ```
 
@@ -1281,7 +1298,7 @@ def apply_stdin_pipe(args: argparse.Namespace) -> None:
         startup request:
 
         ```bash
-        cat diff.txt | deepagents --skill code-review
+        cat diff.txt | dcode --skill code-review
         # initial_prompt = "{contents of diff.txt}"
         ```
 
@@ -1289,7 +1306,7 @@ def apply_stdin_pipe(args: argparse.Namespace) -> None:
         the CLI to run non-interactively with it as the prompt:
 
         ```bash
-        echo "fix the typo in README.md" | deepagents
+        echo "fix the typo in README.md" | dcode
         # non_interactive_message = "fix the typo in README.md"
         ```
 
@@ -1325,7 +1342,7 @@ def apply_stdin_pipe(args: argparse.Namespace) -> None:
             console.print(
                 "[bold red]Error:[/bold red] --stdin was passed but stdin "
                 "is a terminal. Pipe input or use -n instead.\n"
-                "  cat prompt.txt | deepagents --stdin -q"
+                "  cat prompt.txt | dcode --stdin -q"
             )
             sys.exit(1)
         return
@@ -1521,7 +1538,7 @@ def _check_mcp_project_trust(*, trust_flag: bool = False) -> bool | None:
     from rich.console import Console as _Console
 
     docs_url = (
-        "https://docs.langchain.com/oss/python/deepagents/cli/"
+        "https://docs.langchain.com/oss/python/deepagents/code/"
         "mcp-tools#project-level-trust"
     )
     prompt_console = _Console(stderr=True)
@@ -1676,8 +1693,8 @@ def cli_main() -> None:
                 msg = (
                     "Error: --no-mcp and --mcp-config are mutually exclusive."
                     " Use one or the other.\n"
-                    "  deepagents --mcp-config path/to/config.json\n"
-                    "  deepagents --no-mcp\n"
+                    "  dcode --mcp-config path/to/config.json\n"
+                    "  dcode --no-mcp\n"
                 )
                 sys.stderr.write(msg)
                 sys.stderr.flush()
@@ -1712,8 +1729,8 @@ def cli_main() -> None:
             _Console(stderr=True).print(
                 "[bold red]Error:[/bold red] --no-mcp and --mcp-config "
                 "are mutually exclusive. Use one or the other.\n"
-                "  deepagents --mcp-config path/to/config.json\n"
-                "  deepagents --no-mcp"
+                "  dcode --mcp-config path/to/config.json\n"
+                "  dcode --no-mcp"
             )
             sys.exit(2)
 
@@ -1728,8 +1745,8 @@ def cli_main() -> None:
                 "[bold red]Error:[/bold red] --skill requires "
                 "--non-interactive (-n) when combined with --quiet or "
                 "--no-stream.\n"
-                "  deepagents --skill code-review -m 'review this patch'\n"
-                "  deepagents --skill code-review -n 'review this patch'"
+                "  dcode --skill code-review -m 'review this patch'\n"
+                "  dcode --skill code-review -n 'review this patch'"
             )
             sys.exit(2)
 
@@ -1740,7 +1757,18 @@ def cli_main() -> None:
             _Console(stderr=True).print(
                 "[bold red]Error:[/bold red] --max-turns requires "
                 "--non-interactive (-n) or piped stdin\n"
-                "  deepagents -n 'refactor auth module' --max-turns 5"
+                "  dcode -n 'refactor auth module' --max-turns 5"
+            )
+            sys.exit(2)
+
+        timeout_set = getattr(args, "timeout", None) is not None
+        if timeout_set and not args.non_interactive_message:
+            from rich.console import Console as _Console
+
+            _Console(stderr=True).print(
+                "[bold red]Error:[/bold red] --timeout requires "
+                "--non-interactive (-n) or piped stdin\n"
+                "  dcode -n 'run the test suite' --timeout 120"
             )
             sys.exit(2)
 
@@ -1759,7 +1787,7 @@ def cli_main() -> None:
             _Console(stderr=True).print(
                 f"[bold red]Error:[/bold red] {flag} requires "
                 "--non-interactive (-n) or piped stdin\n"
-                "  deepagents -n 'summarize README.md' --quiet"
+                "  dcode -n 'summarize README.md' --quiet"
             )
             sys.exit(2)
 
@@ -1964,7 +1992,7 @@ def cli_main() -> None:
                 if getattr(args, "mcp_config", None) and not args.config_path:
                     print(  # noqa: T201
                         "--mcp-config is not supported for 'mcp login'. "
-                        "Use: deepagents mcp login <server> --config <path>",
+                        "Use: dcode mcp login <server> --config <path>",
                         file=sys.stderr,
                     )
                     sys.exit(2)
@@ -2066,26 +2094,49 @@ def cli_main() -> None:
             # Non-interactive mode - execute single task and exit
             from deepagents_code.non_interactive import run_non_interactive
 
-            exit_code = asyncio.run(
-                run_non_interactive(
-                    message=args.non_interactive_message,
-                    assistant_id=assistant_id,
-                    model_name=getattr(args, "model", None),
-                    model_params=model_params,
-                    profile_override=profile_override,
-                    sandbox_type=args.sandbox,
-                    sandbox_id=args.sandbox_id,
-                    sandbox_setup=getattr(args, "sandbox_setup", None),
-                    initial_skill=getattr(args, "initial_skill", None),
-                    startup_cmd=getattr(args, "startup_cmd", None),
-                    quiet=args.quiet,
-                    stream=not args.no_stream,
-                    mcp_config_path=getattr(args, "mcp_config", None),
-                    no_mcp=getattr(args, "no_mcp", False),
-                    trust_project_mcp=getattr(args, "trust_project_mcp", False),
-                    max_turns=getattr(args, "max_turns", None),
+            timeout = getattr(args, "timeout", None)
+            try:
+                exit_code = asyncio.run(
+                    asyncio.wait_for(
+                        run_non_interactive(
+                            message=args.non_interactive_message,
+                            assistant_id=assistant_id,
+                            model_name=getattr(args, "model", None),
+                            model_params=model_params,
+                            profile_override=profile_override,
+                            sandbox_type=args.sandbox,
+                            sandbox_id=args.sandbox_id,
+                            sandbox_setup=getattr(args, "sandbox_setup", None),
+                            initial_skill=getattr(args, "initial_skill", None),
+                            startup_cmd=getattr(args, "startup_cmd", None),
+                            quiet=args.quiet,
+                            stream=not args.no_stream,
+                            mcp_config_path=getattr(args, "mcp_config", None),
+                            no_mcp=getattr(args, "no_mcp", False),
+                            trust_project_mcp=getattr(args, "trust_project_mcp", False),
+                            max_turns=getattr(args, "max_turns", None),
+                        ),
+                        timeout=timeout,
+                    )
                 )
-            )
+            except TimeoutError:
+                # `asyncio.wait_for` raises `asyncio.TimeoutError`, which is
+                # an alias of the builtin on Python >= 3.11 (the project's
+                # minimum).
+                from rich.console import Console as _Console
+
+                _Console(stderr=True).print(
+                    f"[bold red]Error:[/bold red] agent timed out after "
+                    f"{timeout}s. Retry with a larger --timeout, or use "
+                    "--max-turns for a turn-count limit."
+                )
+                sys.exit(124)
+            except KeyboardInterrupt:
+                # `asyncio.run` re-raises `KeyboardInterrupt` past the inner
+                # `run_non_interactive` handler when the signal hits during
+                # `wait_for`; mirror its exit code 130 here so Ctrl-C is a
+                # quiet exit instead of a traceback.
+                sys.exit(130)
             sys.exit(exit_code)
         else:
             # Resolve recent-agent fallback only for actual session launches.
@@ -2187,7 +2238,7 @@ def cli_main() -> None:
             if thread_id and return_code == 0 and asyncio.run(thread_exists(thread_id)):
                 console.print()
                 console.print("[dim]Resume this thread with:[/dim]")
-                hint = Text("deepagents -r ", style="cyan")
+                hint = Text("dcode -r ", style="cyan")
                 hint.append(str(thread_id), style="cyan")
                 console.print(hint)
 
@@ -2222,7 +2273,7 @@ def cli_main() -> None:
                         console.print(cmd_hint)
                         if not is_auto_update_enabled():
                             auto_hint = Text("Enable auto-updates: ", style="dim")
-                            auto_hint.append("deepagents --auto-update", style="cyan")
+                            auto_hint.append("dcode --auto-update", style="cyan")
                             console.print(auto_hint)
                         mark_update_notified(latest)
             except Exception: